2024
Ghadernejad, Saleh; Esmaeili, Kamran
Predicting Rock Hardness and Abrasivity Using Hyperspectral Imaging Data and Random Forest Regressor Model Journal Article
In: Remote Sensing, vol. 16, iss. 20, 2024.
@article{nokey,
title = {Predicting Rock Hardness and Abrasivity Using Hyperspectral Imaging Data and Random Forest Regressor Model},
author = {Saleh Ghadernejad and Kamran Esmaeili},
doi = {http://dx.doi.org/10.3390/rs16203778},
year = {2024},
date = {2024-10-02},
urldate = {2024-10-02},
journal = {Remote Sensing},
volume = {16},
issue = {20},
abstract = {This study aimed to develop predictive models for rock hardness and abrasivity based on hyperspectral imaging data, providing valuable information without interrupting the mining processes. The data collection stage first involved scanning 159 rock samples collected from 6 different blasted rock piles using visible and near-infrared (VNIR) and short-wave infrared (SWIR) sensors. The hardness and abrasivity of the samples were then determined through Leeb rebound hardness (LRH) and Cerchar abrasivity index (CAI) tests, respectively. The data preprocessing involved radiometric correction, background removal, and staking VNIR and SWIR images. An integrated approach based on K-means clustering and the band ratio concept was employed for feature extraction, resulting in 28 band-ratio-based features. Afterward, the random forest regressor (RFR) algorithm was employed to develop predictive models for rock hardness and abrasivity separately. The performance assessment showed that the developed models can estimate rock hardness and abrasivity of unseen data with R² scores of 0.74 and 0.79, respectively, with the most influential features located mainly within the SWIR region. The results indicate that integrated hyperspectral data and RFR technique have strong potential for practical and efficient rock hardness and abrasivity characterization during mining processes.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This study aimed to develop predictive models for rock hardness and abrasivity based on hyperspectral imaging data, providing valuable information without interrupting the mining processes. The data collection stage first involved scanning 159 rock samples collected from 6 different blasted rock piles using visible and near-infrared (VNIR) and short-wave infrared (SWIR) sensors. The hardness and abrasivity of the samples were then determined through Leeb rebound hardness (LRH) and Cerchar abrasivity index (CAI) tests, respectively. The data preprocessing involved radiometric correction, background removal, and staking VNIR and SWIR images. An integrated approach based on K-means clustering and the band ratio concept was employed for feature extraction, resulting in 28 band-ratio-based features. Afterward, the random forest regressor (RFR) algorithm was employed to develop predictive models for rock hardness and abrasivity separately. The performance assessment showed that the developed models can estimate rock hardness and abrasivity of unseen data with R² scores of 0.74 and 0.79, respectively, with the most influential features located mainly within the SWIR region. The results indicate that integrated hyperspectral data and RFR technique have strong potential for practical and efficient rock hardness and abrasivity characterization during mining processes.
Ghadernejad, Saleh; Esmaeili, Kamran
Investigating the Relationship between Geochemistry, Leeb Rebound Hardness, and Cerchar Abrasivity Index Journal Article
In: International Journal of Geomechanics, vol. 24, iss. 12, 2024.
@article{nokey,
title = {Investigating the Relationship between Geochemistry, Leeb Rebound Hardness, and Cerchar Abrasivity Index},
author = {Saleh Ghadernejad and Kamran Esmaeili},
doi = {DOI: 10.1061/IJGNAI.GMENG-9802.},
year = {2024},
date = {2024-09-05},
urldate = {2024-09-05},
journal = {International Journal of Geomechanics},
volume = {24},
issue = {12},
abstract = {Rock hardness and abrasivity are among the most crucial properties that can significantly impact the interaction between rocks and mechanical tools in different parts of geoengineering projects. Accurate estimation of these properties is essential for a better understanding and optimization of geoengineering operations. Hence, the main aim of this study was to develop different machine learning (ML) models based on the geochemical measurements for predicting rock hardness and abrasivity. To do this, 159 rock samples were collected from a gold mine, and portable X-ray fluorescence spectrometry (pXRF), Leeb rebound hardness (LRH), and Cerchar abrasivity index (CAI) tests were performed on the collected rock samples. Three different ML algorithms, including random forest regressor (RFR), support vector regression, and gradient boosting regressor, were applied to develop predictive models for LRH and CAI separately. Considering the fact that the geochemical data are of the compositional type, two scenarios were followed: developing predictive models based on the original data obtained from the pXRF and the centered log-ratio (Clr) transformed data, resulting in the development of six predictive models for LRH and CAI. The performance assessment of the developed predictive models showed that RFR models outperformed the other two ML algorithms in predicting LRH and CAI. In addition, the developed models based on the original data demonstrated a better performance in both cases of LRH and CAI than the trained model based on Clr data. The result indicates that integrated pXRF measurements and RFR technique have strong potential to be used for practical and efficient rock materials characterization during exploration and extraction processes.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Rock hardness and abrasivity are among the most crucial properties that can significantly impact the interaction between rocks and mechanical tools in different parts of geoengineering projects. Accurate estimation of these properties is essential for a better understanding and optimization of geoengineering operations. Hence, the main aim of this study was to develop different machine learning (ML) models based on the geochemical measurements for predicting rock hardness and abrasivity. To do this, 159 rock samples were collected from a gold mine, and portable X-ray fluorescence spectrometry (pXRF), Leeb rebound hardness (LRH), and Cerchar abrasivity index (CAI) tests were performed on the collected rock samples. Three different ML algorithms, including random forest regressor (RFR), support vector regression, and gradient boosting regressor, were applied to develop predictive models for LRH and CAI separately. Considering the fact that the geochemical data are of the compositional type, two scenarios were followed: developing predictive models based on the original data obtained from the pXRF and the centered log-ratio (Clr) transformed data, resulting in the development of six predictive models for LRH and CAI. The performance assessment of the developed predictive models showed that RFR models outperformed the other two ML algorithms in predicting LRH and CAI. In addition, the developed models based on the original data demonstrated a better performance in both cases of LRH and CAI than the trained model based on Clr data. The result indicates that integrated pXRF measurements and RFR technique have strong potential to be used for practical and efficient rock materials characterization during exploration and extraction processes.
Rezaei, Mohammad; Esmaeili, Kamran
The Anisotropy of Rock Drilling in Marble Quarry Mining Based on the Relationship between Vertical and Horizontal Drilling Rates Journal Article
In: International Journal of Geomechanics, vol. 24, iss. 10, 2024.
@article{Drilling,
title = {The Anisotropy of Rock Drilling in Marble Quarry Mining Based on the Relationship between Vertical and Horizontal Drilling Rates},
author = {Mohammad Rezaei and Kamran Esmaeili},
doi = {DOI: 10.1061/IJGNAI.GMENG-9658},
year = {2024},
date = {2024-08-14},
urldate = {2024-08-14},
journal = {International Journal of Geomechanics},
volume = {24},
issue = {10},
abstract = {There is a high correlation between rock engineering behavior and the anisotropy index, which is crucial for the safety assessment of rock properties. In rock drilling, rock anisotropy significantly affects the cost and efficiency of the drilling project. Therefore, the anisotropy during the rock drilling process is investigated in this study. First, measurements of horizontal and vertical drilling rates were conducted at a marble quarry. Then, experimental tests were conducted to measure the physical and mechanical properties of collected minor rock blocks corresponding to the major under-drilling marble blocks. Through this process, a total of 40 data sets comprising vertical and horizontal drilling rates and 12 different rock properties were provided. The results showed that the vertical drilling rate correlates more with rock mechanical properties, while the horizontal drilling rate is associated more with physical characteristics. Also, sensitivity analysis confirmed that porosity and dry density are the most and least effective variables on both vertical and horizontal drilling rates, respectively. Based on the analysis of variance, the polynomial equation was found to be the optimum relationship between vertical and horizontal drilling rates. Based on the comparative analysis and using actual data sets, the accuracy of the proposed equation was proved. Finally, a drilling anisotropy index was proposed based on the ratio of vertical to horizontal drilling rates. The results showed that the drilling anisotropy index is more prominent in rock blocks with high-strength properties. Also, it was proved that the drilling anisotropy index has an inverse relation with vertical and horizontal drilling rates. As the vertical and horizontal drilling rates are inherently affected by different rock properties, it can be concluded that the suggested relationship between vertical and horizontal drilling rates and the proposed drilling anisotropy index could be effectively applied to assessing the interaction of drilling rates with rock properties in anisotropic rocks. Practical Applications: This study investigated the relationships between rock properties and vertical and horizontal drilling rates, which is one of the main issues in rock drilling operations. According to the field and laboratory results, the correlation between the physical and mechanical properties of rocks with vertical and horizontal drilling rates was achieved, which can be considered during the drilling operation to enhance the drilling efficiency. Also, a new drilling anisotropy index was proposed, which can help identify a good interaction between drilling tools and the rock environment in anisotropic rocks. In addition, a new relation was proposed between vertical and horizontal drilling rates, which can be applied in practice to predict one of them based on the other. Generally, the outputs of this research could be applied to specifying the relationship between rock properties and vertical and horizontal drilling rates, emphasizing the drilling anisotropy index. In practice, the study outputs can lead to enhanced drilling efficiency and mine productivity in marble quarry mining by providing a deep knowledge of rock interaction with drilling tools while drilling in both the vertical and horizontal directions and choosing proper drilling tools for the under-drilling rock type.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
There is a high correlation between rock engineering behavior and the anisotropy index, which is crucial for the safety assessment of rock properties. In rock drilling, rock anisotropy significantly affects the cost and efficiency of the drilling project. Therefore, the anisotropy during the rock drilling process is investigated in this study. First, measurements of horizontal and vertical drilling rates were conducted at a marble quarry. Then, experimental tests were conducted to measure the physical and mechanical properties of collected minor rock blocks corresponding to the major under-drilling marble blocks. Through this process, a total of 40 data sets comprising vertical and horizontal drilling rates and 12 different rock properties were provided. The results showed that the vertical drilling rate correlates more with rock mechanical properties, while the horizontal drilling rate is associated more with physical characteristics. Also, sensitivity analysis confirmed that porosity and dry density are the most and least effective variables on both vertical and horizontal drilling rates, respectively. Based on the analysis of variance, the polynomial equation was found to be the optimum relationship between vertical and horizontal drilling rates. Based on the comparative analysis and using actual data sets, the accuracy of the proposed equation was proved. Finally, a drilling anisotropy index was proposed based on the ratio of vertical to horizontal drilling rates. The results showed that the drilling anisotropy index is more prominent in rock blocks with high-strength properties. Also, it was proved that the drilling anisotropy index has an inverse relation with vertical and horizontal drilling rates. As the vertical and horizontal drilling rates are inherently affected by different rock properties, it can be concluded that the suggested relationship between vertical and horizontal drilling rates and the proposed drilling anisotropy index could be effectively applied to assessing the interaction of drilling rates with rock properties in anisotropic rocks. Practical Applications: This study investigated the relationships between rock properties and vertical and horizontal drilling rates, which is one of the main issues in rock drilling operations. According to the field and laboratory results, the correlation between the physical and mechanical properties of rocks with vertical and horizontal drilling rates was achieved, which can be considered during the drilling operation to enhance the drilling efficiency. Also, a new drilling anisotropy index was proposed, which can help identify a good interaction between drilling tools and the rock environment in anisotropic rocks. In addition, a new relation was proposed between vertical and horizontal drilling rates, which can be applied in practice to predict one of them based on the other. Generally, the outputs of this research could be applied to specifying the relationship between rock properties and vertical and horizontal drilling rates, emphasizing the drilling anisotropy index. In practice, the study outputs can lead to enhanced drilling efficiency and mine productivity in marble quarry mining by providing a deep knowledge of rock interaction with drilling tools while drilling in both the vertical and horizontal directions and choosing proper drilling tools for the under-drilling rock type.
Akbar, Somaieh; Abdolmaleki, Mehdi; Ghadernejad, Saleh; Esmaeili, Kamran
Applying Knowledge-Based and Data-Driven Methods to Improve Ore Grade Control of Blast Hole Drill Cuttings Using Hyperspectral Imaging Journal Article
In: Remote Sensing, vol. 16, iss. 15, 2024.
@article{Drillcut,
title = {Applying Knowledge-Based and Data-Driven Methods to Improve Ore Grade Control of Blast Hole Drill Cuttings Using Hyperspectral Imaging},
author = {Somaieh Akbar and Mehdi Abdolmaleki and Saleh Ghadernejad and Kamran Esmaeili},
doi = {DOI: 10.3390/rs16152823},
year = {2024},
date = {2024-08-04},
urldate = {2024-08-04},
journal = {Remote Sensing},
volume = {16},
issue = {15},
abstract = {This study introduces a novel method utilizing hyperspectral imaging for instantaneous ore-waste analysis of drill cuttings. To implement this technique, we collected samples of drill cuttings at regular depth intervals from five blast holes in an open pit gold mine and subjected them to scanning using a hyperspectral imaging system. Subsequently, we employed two distinct methods for processing the hyperspectral images. A knowledge-based method was used to estimate ore grade within each sampled interval, and a data-driven technique was employed to distinguish the ore and waste for each sample interval. Firstly, leveraging the mixed mineralogical composition of the samples, the Linear Spectral Unmixing (LSU) technique was utilized to predict ore grade for each sample. Additionally, the Gradient Boosting Classifier (GBC) was used as an efficient data-driven approach to classify ore-waste samples. Both methods rendered accurate results when they were compared with results obtained through laboratory X-ray diffraction (XRD) analysis and gold assay analysis for the same sample intervals. Adopting the proposed methodology in open pit mine operations can significantly enhance the process of grade control during blast hole drilling. This includes reducing costs, saving time, minimizing uncertainty in ore grade estimation, and establishing more precise ore-waste boundaries in resource block models.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This study introduces a novel method utilizing hyperspectral imaging for instantaneous ore-waste analysis of drill cuttings. To implement this technique, we collected samples of drill cuttings at regular depth intervals from five blast holes in an open pit gold mine and subjected them to scanning using a hyperspectral imaging system. Subsequently, we employed two distinct methods for processing the hyperspectral images. A knowledge-based method was used to estimate ore grade within each sampled interval, and a data-driven technique was employed to distinguish the ore and waste for each sample interval. Firstly, leveraging the mixed mineralogical composition of the samples, the Linear Spectral Unmixing (LSU) technique was utilized to predict ore grade for each sample. Additionally, the Gradient Boosting Classifier (GBC) was used as an efficient data-driven approach to classify ore-waste samples. Both methods rendered accurate results when they were compared with results obtained through laboratory X-ray diffraction (XRD) analysis and gold assay analysis for the same sample intervals. Adopting the proposed methodology in open pit mine operations can significantly enhance the process of grade control during blast hole drilling. This includes reducing costs, saving time, minimizing uncertainty in ore grade estimation, and establishing more precise ore-waste boundaries in resource block models.
Rezaei, Mohammad; Mousavi, Seyed Zanyar Seyed; Esmaeili, Kamran
Combination of Monte Carlo simulation and Bishop technique for the slope stability analysis of the Gol-E-Gohar iron open pit mine Journal Article
In: Journal of Mining and Environment, 2024.
@article{nokey,
title = {Combination of Monte Carlo simulation and Bishop technique for the slope stability analysis of the Gol-E-Gohar iron open pit mine},
author = {Mohammad Rezaei and Seyed Zanyar Seyed Mousavi and Kamran Esmaeili},
doi = {10.22044/jme.2024.14470.2718},
year = {2024},
date = {2024-07-23},
urldate = {2024-07-23},
journal = {Journal of Mining and Environment},
abstract = {This study introduces a novel approach, known as Hybrid Probabilistic Slope Stability Analysis (HPSSA), tailored for Mine 4 of the Gol-E-Gohar iron complex in Iran. The mine walls are first divided into 8 separate structural zones, including A-A' to H-H' sections for slope stability analysis. Then, sufficient core specimens are prepared from 22 drilled boreholes and the required parameters for slope design, including cohesion (c), friction angle (φ), and unit weight (γ), are measured. Finally, the HPSSA approach is performed through the combination of Monte Carlo simulation (MCS), Mohr-Coulomb criterion and Bishop's technique. According to the HPSSA results, the normal distribution function is achieved as the best curve fit for c, φ and γ parameters. Also, the obtained values of mean probabilistic safety factor (SF) for defined structural zones vary from 0.93 to 1.86, with the probability of failure (PF) of 0 to 75.6%. Moreover, SF values varied from 0.68 to 1.22 (mean value of 0.93) with a PF of 75% for the A-A' section and from 0.65 to 1.24 (mean value of 0.97) with a PF of 60% for the H-H' section. Hence, it is concluded that the A-A' section and mine’s north wall are more prone to instability with PF>60%. On the other hand, SF>1.2 and PF<5% for other mine walls (sections B-B'-G-G') prove that they are highly unlikely to be unstable. Displacement monitoring of the pit walls using installed prisms confirmed that average displacements in structural zones have a similar trend with SF values of the HPSSA. The results show a good agreement between the trend of probabilistic SFs and monitored slope displacements. Lastly, comparative analysis confirmed the validity of the suggested HPSSA approach with relatively higher accuracy than most previous slope stability analysis methods.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This study introduces a novel approach, known as Hybrid Probabilistic Slope Stability Analysis (HPSSA), tailored for Mine 4 of the Gol-E-Gohar iron complex in Iran. The mine walls are first divided into 8 separate structural zones, including A-A' to H-H' sections for slope stability analysis. Then, sufficient core specimens are prepared from 22 drilled boreholes and the required parameters for slope design, including cohesion (c), friction angle (φ), and unit weight (γ), are measured. Finally, the HPSSA approach is performed through the combination of Monte Carlo simulation (MCS), Mohr-Coulomb criterion and Bishop's technique. According to the HPSSA results, the normal distribution function is achieved as the best curve fit for c, φ and γ parameters. Also, the obtained values of mean probabilistic safety factor (SF) for defined structural zones vary from 0.93 to 1.86, with the probability of failure (PF) of 0 to 75.6%. Moreover, SF values varied from 0.68 to 1.22 (mean value of 0.93) with a PF of 75% for the A-A' section and from 0.65 to 1.24 (mean value of 0.97) with a PF of 60% for the H-H' section. Hence, it is concluded that the A-A' section and mine’s north wall are more prone to instability with PF>60%. On the other hand, SF>1.2 and PF<5% for other mine walls (sections B-B'-G-G') prove that they are highly unlikely to be unstable. Displacement monitoring of the pit walls using installed prisms confirmed that average displacements in structural zones have a similar trend with SF values of the HPSSA. The results show a good agreement between the trend of probabilistic SFs and monitored slope displacements. Lastly, comparative analysis confirmed the validity of the suggested HPSSA approach with relatively higher accuracy than most previous slope stability analysis methods.
Houshmand, Negin; Esmaeili, Kamran; Goodfellow, Sebastian
ARMA-2024-1047, 58th US Rock Mechanics/Geomechanics Symposium held in Golden, Colorado, USA 2024.
@conference{nokey,
title = {Enhancing joint detection and RQD estimation in acoustic televiewer imaging through automated instance segmentation and deep learning},
author = {Negin Houshmand and Kamran Esmaeili and Sebastian Goodfellow},
doi = {https://doi.org/10.56952/ARMA-2024-1047},
year = {2024},
date = {2024-06-26},
urldate = {2024-06-26},
booktitle = {ARMA-2024-1047},
organization = {58th US Rock Mechanics/Geomechanics Symposium held in Golden, Colorado, USA},
abstract = {Borehole imaging provides a fast and efficient way to characterize the structural complexity of rock mass. The acoustic televiewer (ATV) imaging technique is commonly used for fracture detection and RQD determination. However, the conventional process of ATV data and manually detecting and characterizing joints in ATV images is laborious, subjective, and inconsistent. This study introduces an automated method for joint detection and RQD estimation using ATV images. For this study, 1400 meters of ATV data from 24 boreholes across five different mines were collected. In the initial step, a deep learning algorithm called Mask RCNN was utilized for joint detection. This instance segmentation method demonstrated promising results, achieving high accuracy in automated joint detection on an unseen test set. The model's predictions yielded a mean absolute error of 1.7% for RQD estimation. These findings indicate that the model effectively detects natural joints and estimates RQD values. By automating joint detection and characterization, as well as RQD estimation, the proposed method streamlines and enhances the automated subsurface structure analysis process.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Borehole imaging provides a fast and efficient way to characterize the structural complexity of rock mass. The acoustic televiewer (ATV) imaging technique is commonly used for fracture detection and RQD determination. However, the conventional process of ATV data and manually detecting and characterizing joints in ATV images is laborious, subjective, and inconsistent. This study introduces an automated method for joint detection and RQD estimation using ATV images. For this study, 1400 meters of ATV data from 24 boreholes across five different mines were collected. In the initial step, a deep learning algorithm called Mask RCNN was utilized for joint detection. This instance segmentation method demonstrated promising results, achieving high accuracy in automated joint detection on an unseen test set. The model's predictions yielded a mean absolute error of 1.7% for RQD estimation. These findings indicate that the model effectively detects natural joints and estimates RQD values. By automating joint detection and characterization, as well as RQD estimation, the proposed method streamlines and enhances the automated subsurface structure analysis process.
Mehrgini, Behzad; Esmaeili, Kamran
Numerical investigation of continuous miner parameters on fine generation: a case study at a salt mine Conference
6th Itasca Symposium on Applied Numerical Modeling, Toronto, Canada 2024.
@conference{nokey,
title = {Numerical investigation of continuous miner parameters on fine generation: a case study at a salt mine},
author = {Behzad Mehrgini and Kamran Esmaeili},
year = {2024},
date = {2024-06-06},
urldate = {2024-06-06},
organization = {6th Itasca Symposium on Applied Numerical Modeling, Toronto, Canada},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
de Kooker, Leon; Ferentinou, Maria; Musonda, Innocent; Esmaeili, Kamran
Investigation of the stability of a fly ash pond facility using 2D and 3D slope stability analysis Journal Article
In: Mining, Metallurgy & Exploration, 2024.
@article{deKooker2024,
title = {Investigation of the stability of a fly ash pond facility using 2D and 3D slope stability analysis},
author = {Leon de Kooker and Maria Ferentinou and Innocent Musonda and Kamran Esmaeili},
url = {https://link.springer.com/article/10.1007/s42461-024-00961-z?utm_source=rct_congratemailt&utm_medium=email&utm_campaign=oa_20240327&utm_content=10.1007/s42461-024-00961-z},
doi = {https://doi.org/10.1007/s42461-024-00961-z},
year = {2024},
date = {2024-03-27},
urldate = {2024-03-27},
journal = {Mining, Metallurgy & Exploration},
abstract = {A numerical investigation of the effect of pore pressure regime on the safety factor and the critical failure mechanism is presented for fly ash storage facility. Pore pressures’ measurements from standpipe piezometers and pore pressure estimated from seepage analysis are used to compare the factor of safety for a fly ash slope. This was applied for considering static and seismic scenarios. A probabilistic approach was applied to account for the uncertainties resulting from the limited data available and support a qualitative risk assessment evaluation. Slope stability analysis is conducted in two and three dimensions, adopting the limit equilibrium analysis approach, and also a finite element seepage analysis, to assess the stability of the slope. The two-dimensional cross-sections were extruded to three-dimensional models to estimate the factor of safety and associated shear failure. The results from the performed analysis suggest an increase in safety factor values of 5%.
},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
A numerical investigation of the effect of pore pressure regime on the safety factor and the critical failure mechanism is presented for fly ash storage facility. Pore pressures’ measurements from standpipe piezometers and pore pressure estimated from seepage analysis are used to compare the factor of safety for a fly ash slope. This was applied for considering static and seismic scenarios. A probabilistic approach was applied to account for the uncertainties resulting from the limited data available and support a qualitative risk assessment evaluation. Slope stability analysis is conducted in two and three dimensions, adopting the limit equilibrium analysis approach, and also a finite element seepage analysis, to assess the stability of the slope. The two-dimensional cross-sections were extruded to three-dimensional models to estimate the factor of safety and associated shear failure. The results from the performed analysis suggest an increase in safety factor values of 5%.
2023
Ghadernejad, Saleh; Esmaeili, Kamran
The application of small sample theory and confidence interval method to determine the representative mean Leeb rebound hardness value Journal Article
In: Bulletin of Engineering Geology and the Environment, vol. 83, iss. 25, 2023.
@article{Ghadernejad2023b,
title = {The application of small sample theory and confidence interval method to determine the representative mean Leeb rebound hardness value},
author = {Saleh Ghadernejad and Kamran Esmaeili },
doi = {https://doi.org/10.1007/s10064-023-03512-w},
year = {2023},
date = {2023-12-26},
urldate = {2023-12-26},
journal = {Bulletin of Engineering Geology and the Environment},
volume = {83},
issue = {25},
abstract = {The Leeb rebound hardness (LRH) test is a fast, non-destructive, and portable technique widely used to assess rock hardness in the field of rock engineering. However, no universally approved standard or testing method for measuring the representative mean Leeb rebound hardness value (HLD) exists. Hence, this research aimed to propose a method for measuring the representative mean HLD value and assess the effects of ambient temperature and geochemistry on the mean HLD value. The literature review revealed that regardless of the selected number of LRH measurements, the existing testing approaches could be classified as single impact, repeated impact, or hybrid dynamic hardness methods. The results of an extensive laboratory testing program on 16 different rock samples showed that only the single-impact method could result in a representative mean HLD value. Afterward, an integrated statistical technique called small sample theory and associated confidence interval was used to address the appropriate number of LRH measurements leading to a representative mean HLD value. Statistical analyses showed that it is impossible to determine a unique predefined number of LRH measurements leading to a representative mean HLD value in general. Instead, one could consider an error level and a confidence interval level and must perform the test until the small sample theory’s conditions are satisfied. Furthermore, while it was observed that the ambient temperature has no clear effect on the representative mean HLD value, the geochemistry analysis demonstrated that certain geochemical elements could control the mean and variation of the measured HLD values.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The Leeb rebound hardness (LRH) test is a fast, non-destructive, and portable technique widely used to assess rock hardness in the field of rock engineering. However, no universally approved standard or testing method for measuring the representative mean Leeb rebound hardness value (HLD) exists. Hence, this research aimed to propose a method for measuring the representative mean HLD value and assess the effects of ambient temperature and geochemistry on the mean HLD value. The literature review revealed that regardless of the selected number of LRH measurements, the existing testing approaches could be classified as single impact, repeated impact, or hybrid dynamic hardness methods. The results of an extensive laboratory testing program on 16 different rock samples showed that only the single-impact method could result in a representative mean HLD value. Afterward, an integrated statistical technique called small sample theory and associated confidence interval was used to address the appropriate number of LRH measurements leading to a representative mean HLD value. Statistical analyses showed that it is impossible to determine a unique predefined number of LRH measurements leading to a representative mean HLD value in general. Instead, one could consider an error level and a confidence interval level and must perform the test until the small sample theory’s conditions are satisfied. Furthermore, while it was observed that the ambient temperature has no clear effect on the representative mean HLD value, the geochemistry analysis demonstrated that certain geochemical elements could control the mean and variation of the measured HLD values.
Usta, Engin; Esmaeili, Kamran
Predicting Natural Joints' Shear Behaviour Using Core Logging Data and Unsupervised Learning Method Conference
28th International Mining Congress and Exhibition of Türkiye, 28 Nov-1 Dec, 2023, 2023.
@conference{Usta2023b,
title = {Predicting Natural Joints' Shear Behaviour Using Core Logging Data and Unsupervised Learning Method},
author = {Engin Usta and Kamran Esmaeili},
year = {2023},
date = {2023-11-27},
urldate = {2023-11-27},
publisher = {28th International Mining Congress and Exhibition of Türkiye},
address = {28 Nov-1 Dec, 2023},
abstract = {Joint surface characteristics extensively impact the shear behaviour of discontinuity and, subsequently, the geomechanical design of a surface and underground excavation. The conventional geotechnical core logging techniques for joint surface characterization are subjective and inconsistent. Sensor-based technologies can reduce this uncertainty and give more precise results.
In this study, manual and multi-sensor core logging methods were used on nearly 600m of core samples from an underground mine in Sudbury, Canada, to predict the shear behaviour of joints. 365 natural joints were identified, and joint roughness and joint alteration indices were assigned to each individual joint using manual observation techniques. A portable XRF, a handheld 3D scanner, and an Equotip Leeb hardness test device were used for data collection for sensor-based core logging. A comparison was made to determine the discrepancy between manual and digital core logging techniques.
PCA and K-means clustering techniques were used to classify the joints based on their surface characteristics. Four clusters were identified, and direct shear tests were conducted on selected joint samples from each cluster. Residual friction angles were measured for the tested joint samples. Finally, an unsupervised learning method was employed to predict joints' shear behaviour based on the measured and assessed joint surface characteristics.
},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Joint surface characteristics extensively impact the shear behaviour of discontinuity and, subsequently, the geomechanical design of a surface and underground excavation. The conventional geotechnical core logging techniques for joint surface characterization are subjective and inconsistent. Sensor-based technologies can reduce this uncertainty and give more precise results.
In this study, manual and multi-sensor core logging methods were used on nearly 600m of core samples from an underground mine in Sudbury, Canada, to predict the shear behaviour of joints. 365 natural joints were identified, and joint roughness and joint alteration indices were assigned to each individual joint using manual observation techniques. A portable XRF, a handheld 3D scanner, and an Equotip Leeb hardness test device were used for data collection for sensor-based core logging. A comparison was made to determine the discrepancy between manual and digital core logging techniques.
PCA and K-means clustering techniques were used to classify the joints based on their surface characteristics. Four clusters were identified, and direct shear tests were conducted on selected joint samples from each cluster. Residual friction angles were measured for the tested joint samples. Finally, an unsupervised learning method was employed to predict joints' shear behaviour based on the measured and assessed joint surface characteristics.
In this study, manual and multi-sensor core logging methods were used on nearly 600m of core samples from an underground mine in Sudbury, Canada, to predict the shear behaviour of joints. 365 natural joints were identified, and joint roughness and joint alteration indices were assigned to each individual joint using manual observation techniques. A portable XRF, a handheld 3D scanner, and an Equotip Leeb hardness test device were used for data collection for sensor-based core logging. A comparison was made to determine the discrepancy between manual and digital core logging techniques.
PCA and K-means clustering techniques were used to classify the joints based on their surface characteristics. Four clusters were identified, and direct shear tests were conducted on selected joint samples from each cluster. Residual friction angles were measured for the tested joint samples. Finally, an unsupervised learning method was employed to predict joints' shear behaviour based on the measured and assessed joint surface characteristics.
Houshmand, Negin; Esmaeili, Kamran; Goodfellow, Sebastian; Ordóñez-Calderón, Juan Carlos
Predicting rock hardness using Gaussian weighted moving average filter on borehole data and machine learning Journal Article
In: Minerals Engineering, vol. 204, no. 108448, 2023.
@article{Houshmand2023,
title = {Predicting rock hardness using Gaussian weighted moving average filter on borehole data and machine learning},
author = {Negin Houshmand and Kamran Esmaeili and Sebastian Goodfellow and Juan Carlos Ordóñez-Calderón},
doi = {https://doi.org/10.1016/j.mineng.2023.108448},
year = {2023},
date = {2023-10-19},
urldate = {2023-10-19},
journal = {Minerals Engineering},
volume = {204},
number = {108448},
abstract = {A comprehensive understanding of the hardness of ore being handled and processed in a mining operation can significantly improve operational efficiencies. This is feasible by providing valuable data to support decision- making through the mining value chain (drilling, blasting, loading, comminution). This study presents the re- sults of a machine learning (ML) approach for rock hardness prediction using rock’s geophysical and geochemical features. Core samples from several mine sites were logged using a multi-sensor core logging (MSCL) system. Measurements include ultrasonic P- and S-wave velocity, elemental concentration via portable X-Ray fluores- cence analyzers (pXRF), and Leeb rebound hardness, measured every 30 cm along 564 m of core samples. K- Means and PCA were used for better interpretation of the data. Supervised ML models (XGBoost and Random Forest) were utilized to predict rock hardness using the elemental concentrations and ultrasonic velocities as predictors. Since the data was collected automatically with predefined intervals, some of the measurement points were near fractures or veins. The Gaussian weighted moving average (WMA) was used to smooth out variations in geochemistry or hardness caused by local features that do not reflect the overall rock characteristics. This approach is effective for building ML models to become less susceptible to local rock features. It was concluded that the rock hardness could be effectively predicted using only geochemistry, and the process of collecting P- and S-wave velocity for hardness prediction can be skipped.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
A comprehensive understanding of the hardness of ore being handled and processed in a mining operation can significantly improve operational efficiencies. This is feasible by providing valuable data to support decision- making through the mining value chain (drilling, blasting, loading, comminution). This study presents the re- sults of a machine learning (ML) approach for rock hardness prediction using rock’s geophysical and geochemical features. Core samples from several mine sites were logged using a multi-sensor core logging (MSCL) system. Measurements include ultrasonic P- and S-wave velocity, elemental concentration via portable X-Ray fluores- cence analyzers (pXRF), and Leeb rebound hardness, measured every 30 cm along 564 m of core samples. K- Means and PCA were used for better interpretation of the data. Supervised ML models (XGBoost and Random Forest) were utilized to predict rock hardness using the elemental concentrations and ultrasonic velocities as predictors. Since the data was collected automatically with predefined intervals, some of the measurement points were near fractures or veins. The Gaussian weighted moving average (WMA) was used to smooth out variations in geochemistry or hardness caused by local features that do not reflect the overall rock characteristics. This approach is effective for building ML models to become less susceptible to local rock features. It was concluded that the rock hardness could be effectively predicted using only geochemistry, and the process of collecting P- and S-wave velocity for hardness prediction can be skipped.
Usta, Engin; Esmaeili, Kamran
Joint surface characterization using manual and multi-sensor core logging systems Conference
15th ISRM Congress 2023 & 72nd Geomechanics Colloquium, Salzburg, Austria, October 9-14, 2023, 2023.
@conference{Usta2023,
title = {Joint surface characterization using manual and multi-sensor core logging systems},
author = {Engin Usta and Kamran Esmaeili},
editor = {Schubert & Kluckner},
year = {2023},
date = {2023-10-10},
urldate = {2023-10-10},
publisher = {15th ISRM Congress 2023 & 72nd Geomechanics Colloquium},
address = {Salzburg, Austria, October 9-14, 2023},
abstract = {The shear behavior of a discontinuity can be significantly influenced by its surface characteristics. This paper compares manual and sensor-based geotechnical core logging methods for joint surface characterization. More than 500 m of core samples were logged, in which 367 joints were both digitally and manually characterized. Manual logging was carried out by visually assessing joint surface roughness and alteration indices. Sensor-based logging includes creating 2D and 3D joint roughness profiles using a handheld 3D scanner and measuring joint wall hardness and alteration type using Equotip Leeb hardness and pXRF analyzer. The collected data were statistically analyzed. A comparison between manual and multi-sensor joint surface characterization was made to demonstrate the discrepancy between the manual and digital logging techniques regarding joint surface roughness, joint wall strength and joint alteration.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
The shear behavior of a discontinuity can be significantly influenced by its surface characteristics. This paper compares manual and sensor-based geotechnical core logging methods for joint surface characterization. More than 500 m of core samples were logged, in which 367 joints were both digitally and manually characterized. Manual logging was carried out by visually assessing joint surface roughness and alteration indices. Sensor-based logging includes creating 2D and 3D joint roughness profiles using a handheld 3D scanner and measuring joint wall hardness and alteration type using Equotip Leeb hardness and pXRF analyzer. The collected data were statistically analyzed. A comparison between manual and multi-sensor joint surface characterization was made to demonstrate the discrepancy between the manual and digital logging techniques regarding joint surface roughness, joint wall strength and joint alteration.
Siamaki, Ali; Esmaeili, Kamran
Investigating Blast-Induced Wave Propagation in a Jointed Rock Sample Conference
GeoSaskatoon 2023, Saskatoon, Canada, October 1-4, 2023, 2023.
@conference{Siamaki2023,
title = {Investigating Blast-Induced Wave Propagation in a Jointed Rock Sample},
author = {Ali Siamaki and Kamran Esmaeili},
year = {2023},
date = {2023-10-02},
urldate = {2023-10-02},
publisher = {GeoSaskatoon 2023},
address = {Saskatoon, Canada, October 1-4, 2023},
abstract = {In open pit mines, repetitive blast-induced ground vibration can increase the risk of pit wall instability due to strain accumulation along discontinuities. The presence of discrete fractures in a rock mass can influence the propagation of blast-induced shock waves in the rock and, consequently, the degradation of the shear strength of the jointed rock mass. Quantifying blast-induced rock mass degradation is essential for predicting the potential risk of pit slope failure. This paper presents the results of a series of numerical experiments to investigate the effects of ground vibration from a single-row blast on a jointed rock, simulated with the 2D Synthetic Rock Mass approach. For this purpose, a 1 m x 2 m rock block with a single persistent joint in the middle was generated using Particle Flow Code (PFC). The model was used to study the effects of joint dip angle, infill thickness, and joint surface condition on blast-induced wave propagation. Both persistent and discontinuous joints with rock bridges were simulated in this study. Recorded blast vibration history from a quarry mine was applied to the 2D models, and wave propagation was monitored along the rock blocks. The results demonstrate that the transmission coefficient is slightly higher when wave propagation is perpendicular to the joint surface. Moreover, the wave transmission coefficient reduces considerably when the joint friction angle and normal and shear stiffnesses of the joint decrease. The higher thickness of joint infill material can decrease the wave transmission coefficient. Finally, a comparison of blast wave propagation through persistent and discontinuous fracture reveals that the fracture absorbs more energy in the latter condition. Hence, a higher degradation (damage) in the form of micro-cracks is observed along the discontinuous fracture within the rock bridges.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
In open pit mines, repetitive blast-induced ground vibration can increase the risk of pit wall instability due to strain accumulation along discontinuities. The presence of discrete fractures in a rock mass can influence the propagation of blast-induced shock waves in the rock and, consequently, the degradation of the shear strength of the jointed rock mass. Quantifying blast-induced rock mass degradation is essential for predicting the potential risk of pit slope failure. This paper presents the results of a series of numerical experiments to investigate the effects of ground vibration from a single-row blast on a jointed rock, simulated with the 2D Synthetic Rock Mass approach. For this purpose, a 1 m x 2 m rock block with a single persistent joint in the middle was generated using Particle Flow Code (PFC). The model was used to study the effects of joint dip angle, infill thickness, and joint surface condition on blast-induced wave propagation. Both persistent and discontinuous joints with rock bridges were simulated in this study. Recorded blast vibration history from a quarry mine was applied to the 2D models, and wave propagation was monitored along the rock blocks. The results demonstrate that the transmission coefficient is slightly higher when wave propagation is perpendicular to the joint surface. Moreover, the wave transmission coefficient reduces considerably when the joint friction angle and normal and shear stiffnesses of the joint decrease. The higher thickness of joint infill material can decrease the wave transmission coefficient. Finally, a comparison of blast wave propagation through persistent and discontinuous fracture reveals that the fracture absorbs more energy in the latter condition. Hence, a higher degradation (damage) in the form of micro-cracks is observed along the discontinuous fracture within the rock bridges.
Ghadernejad, Saleh; Esmaeili, Kamran
Application of Machine Learning Models for Rock Hardness Characterization Using Geochemical Data Conference
CIM Conference, Montreal, Canada, May 1-3, 2023, 2023.
@conference{Ghadernejad2023,
title = {Application of Machine Learning Models for Rock Hardness Characterization Using Geochemical Data},
author = {Saleh Ghadernejad and Kamran Esmaeili},
year = {2023},
date = {2023-05-01},
urldate = {2023-05-01},
publisher = {CIM Conference},
address = {Montreal, Canada, May 1-3, 2023},
abstract = {Rock hardness plays a significant role in various stages of mining projects, from drilling to comminution processes. An accurate and timely estimation of the rock hardness of the mined materials would enable engineers to improve the operational efficiencies of different mining stages. Hence, this paper aimed to develop Machine Learning (ML) models for predicting rock hardness using geochemical data. To do this, 162 rock samples were collected from the Canadian Malartic and Barnat open pits in Quebec, Canada. The Leeb Rebound Hardness (LRH) was used to measure the rebound hardness of the samples because of its flexibility, practicality, and non-destructivity. An integrated statistical technique called small sample theory and associated confidence interval was used to ensure the representativity of the measured mean Leeb hardness (HLD) value. Additionally, a multi-element geochemical analysis was conducted using portable X-ray fluorescence spectrometry (PXRF) to determine the bulk concentration of chemical elements in rock samples. Various ML regressor models, including Random Forest, Support Vector Machine, and Gboost, were used to predict the mean HLD value and rock hardness class. Results showed that rock hardness could be predicted using the developed models based on PXRF data, even in the early stage of a mining project.
},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Rock hardness plays a significant role in various stages of mining projects, from drilling to comminution processes. An accurate and timely estimation of the rock hardness of the mined materials would enable engineers to improve the operational efficiencies of different mining stages. Hence, this paper aimed to develop Machine Learning (ML) models for predicting rock hardness using geochemical data. To do this, 162 rock samples were collected from the Canadian Malartic and Barnat open pits in Quebec, Canada. The Leeb Rebound Hardness (LRH) was used to measure the rebound hardness of the samples because of its flexibility, practicality, and non-destructivity. An integrated statistical technique called small sample theory and associated confidence interval was used to ensure the representativity of the measured mean Leeb hardness (HLD) value. Additionally, a multi-element geochemical analysis was conducted using portable X-ray fluorescence spectrometry (PXRF) to determine the bulk concentration of chemical elements in rock samples. Various ML regressor models, including Random Forest, Support Vector Machine, and Gboost, were used to predict the mean HLD value and rock hardness class. Results showed that rock hardness could be predicted using the developed models based on PXRF data, even in the early stage of a mining project.
Smith, Lawrence Devon; Esmaeili, Kamran
Mining SEDAR Estimating Early-Stage Costs Using NI 43-101 Project Reports Conference
CIM Conference, Montreal, Canada, May 1-3, 2023, 2023.
@conference{Smith2023,
title = {Mining SEDAR Estimating Early-Stage Costs Using NI 43-101 Project Reports},
author = {Lawrence Devon Smith and Kamran Esmaeili
},
year = {2023},
date = {2023-05-01},
urldate = {2023-05-01},
publisher = {CIM Conference},
address = {Montreal, Canada, May 1-3, 2023},
abstract = {In the earliest stages of mineral property development, it is essential to develop reasonable estimates of basic technical and economic parameters in order to determine whether the project has reasonable prospects for eventual economic extraction (RPEEE). In other words, is the property worth pursuing? To achieve this, it is necessary to have access to reliable cost data, including reasonable and tested rules of thumb.
Our team has undertaken a statistical analysis of over one hundred mining project data from publicly available NI 43-101 reports on feasibility studies conducted between 2015-2020. These studies are available on SEDAR. The NI 43-101 reports were reviewed to collect project data, including project location, production rate (annual and daily), tonnages (waste & ore), commodity type, ore geology, mining and milling methods, surface topography, average grades, strip ratio, ramp-up, recoveries, payables, mining Opex, processing Opex, G&A cost, initial Capex, sustaining Capex, and Closure cost.
The collected data were statistically processed to find relationships between the main cost parameters of mining projects based on their operational characteristics. The study results allow a better understanding of mining project costs based on historical data. For example, Taylor’s rule (correlating the total ore reserve to the daily production rate) can be applied to both surface and underground mining projects. Also, reasonable early-stage estimates of ECPM & Indirects, Closure costs, and Sustaining Capex can be based on the total initial capital cost of projects, and the G&A operating costs can be estimated as a percentage of the total operating costs.
},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
In the earliest stages of mineral property development, it is essential to develop reasonable estimates of basic technical and economic parameters in order to determine whether the project has reasonable prospects for eventual economic extraction (RPEEE). In other words, is the property worth pursuing? To achieve this, it is necessary to have access to reliable cost data, including reasonable and tested rules of thumb.
Our team has undertaken a statistical analysis of over one hundred mining project data from publicly available NI 43-101 reports on feasibility studies conducted between 2015-2020. These studies are available on SEDAR. The NI 43-101 reports were reviewed to collect project data, including project location, production rate (annual and daily), tonnages (waste & ore), commodity type, ore geology, mining and milling methods, surface topography, average grades, strip ratio, ramp-up, recoveries, payables, mining Opex, processing Opex, G&A cost, initial Capex, sustaining Capex, and Closure cost.
The collected data were statistically processed to find relationships between the main cost parameters of mining projects based on their operational characteristics. The study results allow a better understanding of mining project costs based on historical data. For example, Taylor’s rule (correlating the total ore reserve to the daily production rate) can be applied to both surface and underground mining projects. Also, reasonable early-stage estimates of ECPM & Indirects, Closure costs, and Sustaining Capex can be based on the total initial capital cost of projects, and the G&A operating costs can be estimated as a percentage of the total operating costs.
Our team has undertaken a statistical analysis of over one hundred mining project data from publicly available NI 43-101 reports on feasibility studies conducted between 2015-2020. These studies are available on SEDAR. The NI 43-101 reports were reviewed to collect project data, including project location, production rate (annual and daily), tonnages (waste & ore), commodity type, ore geology, mining and milling methods, surface topography, average grades, strip ratio, ramp-up, recoveries, payables, mining Opex, processing Opex, G&A cost, initial Capex, sustaining Capex, and Closure cost.
The collected data were statistically processed to find relationships between the main cost parameters of mining projects based on their operational characteristics. The study results allow a better understanding of mining project costs based on historical data. For example, Taylor’s rule (correlating the total ore reserve to the daily production rate) can be applied to both surface and underground mining projects. Also, reasonable early-stage estimates of ECPM & Indirects, Closure costs, and Sustaining Capex can be based on the total initial capital cost of projects, and the G&A operating costs can be estimated as a percentage of the total operating costs.
Yang, Peng; Esmaeili, Kamran; Goodfellow, Sebastian; Ordóñez-Calderón, Juan Carlos
Mine PitWall Geological Mapping Using UAV-Based RGB Imaging and Unsupervised Learning Journal Article
In: Remote Sensing, vol. 15, iss. 6, no. 1641, 2023.
@article{Yang2023,
title = {Mine PitWall Geological Mapping Using UAV-Based RGB Imaging and Unsupervised Learning},
author = {Peng Yang and Kamran Esmaeili and Sebastian Goodfellow and Juan Carlos Ordóñez-Calderón },
doi = { https://doi.org/10.3390/rs15061641},
year = {2023},
date = {2023-03-17},
urldate = {2023-03-17},
journal = {Remote Sensing},
volume = {15},
number = {1641},
issue = {6},
abstract = {In surface mining operations, geological pit wall mapping is important since it provides significant information on the surficial geological features throughout the pit wall faces, thereby improving geological certainty and operational planning. Conventional pit wall geological mapping techniques generally rely on close visual observations and laboratory testing results, which can be both time- and labour-intensive and can expose the technical staff to different safety hazards on the ground. In this work, a case study was conducted by investigating the use of drone-acquired RGB images for pit wall mapping. High spatial resolution RGB image data were collected using a commercially available unmanned aerial vehicle (UAV) at two gold mines in Nevada, USA. Cluster maps were produced using unsupervised learning algorithms, including the implementation of convolutional autoencoders, to explore the use of unlabelled image data for pit wall geological mapping purposes. While the results are promising for simple geological settings, they deviate from human-labelled ground truth maps in more complex geological conditions. This indicates the need to further optimize and explore the algorithms to increase robustness for more complex geological cases.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In surface mining operations, geological pit wall mapping is important since it provides significant information on the surficial geological features throughout the pit wall faces, thereby improving geological certainty and operational planning. Conventional pit wall geological mapping techniques generally rely on close visual observations and laboratory testing results, which can be both time- and labour-intensive and can expose the technical staff to different safety hazards on the ground. In this work, a case study was conducted by investigating the use of drone-acquired RGB images for pit wall mapping. High spatial resolution RGB image data were collected using a commercially available unmanned aerial vehicle (UAV) at two gold mines in Nevada, USA. Cluster maps were produced using unsupervised learning algorithms, including the implementation of convolutional autoencoders, to explore the use of unlabelled image data for pit wall geological mapping purposes. While the results are promising for simple geological settings, they deviate from human-labelled ground truth maps in more complex geological conditions. This indicates the need to further optimize and explore the algorithms to increase robustness for more complex geological cases.
Firouzabadi, Mahdeyeh; Esmaeili, Kamran; Saeedi, Golamreza; Asadi, Mohsen
A discrete element modelling of gravity flow in sublevel caving considering the shape and size distribution of particles Journal Article
In: International Journal of Mining, Reclamation and Environment, 2023.
@article{Firouzabadi2023,
title = {A discrete element modelling of gravity flow in sublevel caving considering the shape and size distribution of particles},
author = {Mahdeyeh Firouzabadi and Kamran Esmaeili and Golamreza Saeedi and Mohsen Asadi},
doi = {http://dx.doi.org/10.1080/17480930.2023.2168870},
year = {2023},
date = {2023-01-11},
urldate = {2023-01-11},
journal = {International Journal of Mining, Reclamation and Environment},
abstract = {One of the most significant challenges for cave mining methods is dilution, a process which is generally controlled by the gravity flow of caved materials. In this study, the discrete element method (DEM) was used to investigate the effects of changing the porosity of blasted ring material and caved waste rocks and of changing the ring inclination on material flow in a sublevel cave (SLC) mine. Yade software was used to simulate two-dimensional gravity flow in a longitudinal section of the sublevel while taking into consideration the shape and size distribution of the flowing rock particles. Four simulations were generated with different porosities, and the material flow in the models was compared to each other. The results demonstrate that when ring porosity is increased, dilution decreases, and the height of the extraction zone grows longitudinally. Also, a lower porosity of the blasted ore material in the ring can cause early dilution entry and expansion of the extraction zone towards the caved waste.. Finally, the effect of ring inclination on the material flow and dilution was investigated. The results of these simulations show an increased dilution as the ring inclination increases and a change in the shape of the extraction zone.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
One of the most significant challenges for cave mining methods is dilution, a process which is generally controlled by the gravity flow of caved materials. In this study, the discrete element method (DEM) was used to investigate the effects of changing the porosity of blasted ring material and caved waste rocks and of changing the ring inclination on material flow in a sublevel cave (SLC) mine. Yade software was used to simulate two-dimensional gravity flow in a longitudinal section of the sublevel while taking into consideration the shape and size distribution of the flowing rock particles. Four simulations were generated with different porosities, and the material flow in the models was compared to each other. The results demonstrate that when ring porosity is increased, dilution decreases, and the height of the extraction zone grows longitudinally. Also, a lower porosity of the blasted ore material in the ring can cause early dilution entry and expansion of the extraction zone towards the caved waste.. Finally, the effect of ring inclination on the material flow and dilution was investigated. The results of these simulations show an increased dilution as the ring inclination increases and a change in the shape of the extraction zone.
2022
Abdolmaleki, Mehdi; Consens, Mariano; Esmaeili, Kamran
Ore-Waste Discrimination Using Supervised and Unsupervised Classification of Hyperspectral Images Journal Article
In: Remote Sensing, vol. 14, iss. 24, no. 6386, 2022.
@article{nokey,
title = {Ore-Waste Discrimination Using Supervised and Unsupervised Classification of Hyperspectral Images},
author = {Mehdi Abdolmaleki and Mariano Consens and Kamran Esmaeili },
doi = {https://doi.org/10.3390/rs14246386},
year = {2022},
date = {2022-12-17},
urldate = {2022-12-17},
journal = {Remote Sensing},
volume = {14},
number = {6386},
issue = {24},
abstract = {Ore and waste discrimination is essential for optimizing exploitation and minimizing ore dilution in a mining operation. The conventional ore/waste discrimination approach relies on the interpretation of ore control by geologists, which is subjective, time consuming, and can cause safety hazards. Hyperspectral remote sensing can be used as an alternative approach for ore/waste discrimination. The focus of this study is to investigate the application of hyperspectral remote sensing and deep learning (DL) for real-time ore and waste classification. Hyperspectral images of several meters of drill core samples from a silver ore deposit labeled by a site geologist as ore and waste material were used to train and test the models. A DL model was trained on the labels generated by a spectral angle mapper (SAM) machine learning technique. The performance on ore/waste discrimination of three classifiers (supervised DL and SAM, and unsupervised k-means clustering) was evaluated using Rand Error and Pixel Error as disagreement analysis and accuracy assessment indices. The results showed that the DL method outperformed the other two techniques. The performance of the DL model reached 0.89, 0.95, 0.89, and 0.91, respectively, on overall accuracy, precision, recall, and F1 score, which indicate the strong capability of the DL model in ore and waste discrimination. An integrated hyperspectral imaging and DL technique has strong potential to be used for practical and efficient discrimination of ore and waste in a near real-time manner.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Ore and waste discrimination is essential for optimizing exploitation and minimizing ore dilution in a mining operation. The conventional ore/waste discrimination approach relies on the interpretation of ore control by geologists, which is subjective, time consuming, and can cause safety hazards. Hyperspectral remote sensing can be used as an alternative approach for ore/waste discrimination. The focus of this study is to investigate the application of hyperspectral remote sensing and deep learning (DL) for real-time ore and waste classification. Hyperspectral images of several meters of drill core samples from a silver ore deposit labeled by a site geologist as ore and waste material were used to train and test the models. A DL model was trained on the labels generated by a spectral angle mapper (SAM) machine learning technique. The performance on ore/waste discrimination of three classifiers (supervised DL and SAM, and unsupervised k-means clustering) was evaluated using Rand Error and Pixel Error as disagreement analysis and accuracy assessment indices. The results showed that the DL method outperformed the other two techniques. The performance of the DL model reached 0.89, 0.95, 0.89, and 0.91, respectively, on overall accuracy, precision, recall, and F1 score, which indicate the strong capability of the DL model in ore and waste discrimination. An integrated hyperspectral imaging and DL technique has strong potential to be used for practical and efficient discrimination of ore and waste in a near real-time manner.
Houshmand, Negin; Goodfellow, Sebastian; Esmaeili, Kamran; Ordóñez-Calderón, Juan Carlos
Rock type classification based on petrophysical, geochemical, and core imaging data using machine and deep learning techniques Journal Article
In: Applied Computing and Geosciences, vol. 16, 2022.
@article{nokey,
title = {Rock type classification based on petrophysical, geochemical, and core imaging data using machine and deep learning techniques},
author = {Negin Houshmand and Sebastian Goodfellow and Kamran Esmaeili and Juan Carlos Ordóñez-Calderón},
doi = {https://doi.org/10.1016/j.acags.2022.100104},
year = {2022},
date = {2022-10-18},
urldate = {2022-10-18},
journal = {Applied Computing and Geosciences},
volume = {16},
abstract = {Rock type classification is one of the most crucial steps of geological and geotechnical core logging. In conventional core logging, rock type classification is subjective and time-consuming. This study aims to automate rock type classification using Machine Learning (ML). About 35 m of core samples from five different rock types obtained from an open pit mine were logged using a Multi-Sensor Core Logging (MSCL) system, along with a core scanner that automatically captured geochemical and petrophysical properties of the samples and 360° images of the core circumference. A train/test split strategy (interval split) was introduced, as it produces more realistic predictions than a random shuffle split. The collected logging data were split into train/test subsets based on the core length intervals. For the automated rock type classification, three approaches were implemented. First, different ML algorithms were used to classify rock types based on their petrophysical (P- and S- wave velocities, Leeb hardness) and geochemical properties (collected using a portable X-Ray Fluorescence analyzer (pXRF)). XGBoost outperformed the other models across all rock types. The second approach classified rock types using core images by applying a pre-trained ResNet-50 on ImageNet. Both classical ML and Convolutional Neural Network (CNN) models have higher accuracy for distinct rock samples than transition and interbedding zones. In the third approach, an expert decision procedure was mimicked by concatenating rock properties (first approach) and five features extracted from images (second approach). The concatenation of images and rock properties improved the F1-score of each approach by 10% and 35%, respectively. The core samples had been annotated with a marker in the field, and the effect of removing marked images from the dataset was investigated. The cleaned images improved the rock type prediction by up to 16% (F1-score) using the CNN approach. However, the improvement in the concatenation approach (7%) was not significant enough to justify the labor-intensive cleaning process.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Rock type classification is one of the most crucial steps of geological and geotechnical core logging. In conventional core logging, rock type classification is subjective and time-consuming. This study aims to automate rock type classification using Machine Learning (ML). About 35 m of core samples from five different rock types obtained from an open pit mine were logged using a Multi-Sensor Core Logging (MSCL) system, along with a core scanner that automatically captured geochemical and petrophysical properties of the samples and 360° images of the core circumference. A train/test split strategy (interval split) was introduced, as it produces more realistic predictions than a random shuffle split. The collected logging data were split into train/test subsets based on the core length intervals. For the automated rock type classification, three approaches were implemented. First, different ML algorithms were used to classify rock types based on their petrophysical (P- and S- wave velocities, Leeb hardness) and geochemical properties (collected using a portable X-Ray Fluorescence analyzer (pXRF)). XGBoost outperformed the other models across all rock types. The second approach classified rock types using core images by applying a pre-trained ResNet-50 on ImageNet. Both classical ML and Convolutional Neural Network (CNN) models have higher accuracy for distinct rock samples than transition and interbedding zones. In the third approach, an expert decision procedure was mimicked by concatenating rock properties (first approach) and five features extracted from images (second approach). The concatenation of images and rock properties improved the F1-score of each approach by 10% and 35%, respectively. The core samples had been annotated with a marker in the field, and the effect of removing marked images from the dataset was investigated. The cleaned images improved the rock type prediction by up to 16% (F1-score) using the CNN approach. However, the improvement in the concatenation approach (7%) was not significant enough to justify the labor-intensive cleaning process.
Bamford, Thomas; Esmaeili, Kamran; Schoellig, Angela
A real-time rock fragmentation analysis using a deep neural network Conference
The 13th International Symposium on Rock Fragmentation by Blasting (FRAGBLAST 13), Hangzhou, China 2022.
@conference{Bamford2022,
title = {A real-time rock fragmentation analysis using a deep neural network},
author = {Thomas Bamford and Kamran Esmaeili and Angela Schoellig },
year = {2022},
date = {2022-10-17},
urldate = {2022-10-17},
organization = {The 13th International Symposium on Rock Fragmentation by Blasting (FRAGBLAST 13), Hangzhou, China},
abstract = {Measuring rock fragmentation is essential to many mining operations because it can significantly influence the efficiency of all downstream mining and comminution processes. Image-based rock fragmentation analysis is commonly used for size distribution analysis in modern mining operations. However, in most cases, poor and wrong automated rock segmentation needs to be corrected through extensive manual editing, which is time-consuming. In this study, a Deep neural network (DNN) approach was developed for automated rock fragmentation analysis. Convolution neural network (CNN) architectures were trained to predict granulation histograms of blasted rock fragments directly from a 2D image using an end-to-end deep learning strategy. The
data set used to train the DNN model comprises 31,419 labelled images of blasted rock fragments. An exclusive data set of 3,772 labelled images were used to test the DNN model. To validate the results with field data, a blasted muck pile from a gold mine was manually labelled and compared with their predicted fragmentation parameters using the DNN model. The results indicate that the DNN method achieves a good accuracy compared to manual image labelling. The developed DNN model allows a real-time rock fragmentation analysis. The DNN model only required a fraction of the time for analysis compared with manual labelling.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Measuring rock fragmentation is essential to many mining operations because it can significantly influence the efficiency of all downstream mining and comminution processes. Image-based rock fragmentation analysis is commonly used for size distribution analysis in modern mining operations. However, in most cases, poor and wrong automated rock segmentation needs to be corrected through extensive manual editing, which is time-consuming. In this study, a Deep neural network (DNN) approach was developed for automated rock fragmentation analysis. Convolution neural network (CNN) architectures were trained to predict granulation histograms of blasted rock fragments directly from a 2D image using an end-to-end deep learning strategy. The
data set used to train the DNN model comprises 31,419 labelled images of blasted rock fragments. An exclusive data set of 3,772 labelled images were used to test the DNN model. To validate the results with field data, a blasted muck pile from a gold mine was manually labelled and compared with their predicted fragmentation parameters using the DNN model. The results indicate that the DNN method achieves a good accuracy compared to manual image labelling. The developed DNN model allows a real-time rock fragmentation analysis. The DNN model only required a fraction of the time for analysis compared with manual labelling.
data set used to train the DNN model comprises 31,419 labelled images of blasted rock fragments. An exclusive data set of 3,772 labelled images were used to test the DNN model. To validate the results with field data, a blasted muck pile from a gold mine was manually labelled and compared with their predicted fragmentation parameters using the DNN model. The results indicate that the DNN method achieves a good accuracy compared to manual image labelling. The developed DNN model allows a real-time rock fragmentation analysis. The DNN model only required a fraction of the time for analysis compared with manual labelling.
Zhao, Yilin; Esmaeili, Kamran
Probabilistic Step-Path Slope Stability Analysis Using Spatially Constrained DFN Models – A Case Study at Tasiast Mine Conference
The Third International Discrete Fracture Network Engineering Conference- DFNE 2022, Santa Fe, USA 2022.
@conference{DFNE2022,
title = {Probabilistic Step-Path Slope Stability Analysis Using Spatially Constrained DFN Models – A Case Study at Tasiast Mine},
author = {Yilin Zhao and Kamran Esmaeili },
year = {2022},
date = {2022-06-30},
urldate = {2022-06-30},
organization = {The Third International Discrete Fracture Network Engineering Conference- DFNE 2022, Santa Fe, USA},
abstract = {In this paper, geotechnical and structural data from an open pit mine were used to generate spatially constrained large-scale DFN models. Data from geotechnical boreholes were used to develop 3D block models of volumetric fracture intensity (P32), using a geostatistical simulation method. The block models were used for spatially constraining 3D geo-cellular DFN models. The fracture centers within the DFN cellular grid are controlled by the cellular P32 values, estimated using the 3D block model. The joints' orientation was bootstrapped from drillhole loggings whereas trace length data, collected across the pit area, were used to further calibrate the geo-cellular models. The resulting DFN models represented the spatial variation of fracture geometry and intensity along the pit area and were employed for numerical modeling of step-path slope stability analysis of the pit slopes. 2D cross sections of the 3D DFN models were embedded into 2D finite element models of the pit slope for a probabilistic step path stability analysis. In addition, the DFN models were used to estimate the rock bridge percentage along the pit walls which was ultimately used to assess the composite rock mass strength for a limit equilibrium analysis of the overall pit slope. The results of the two numerical approaches were compared to a simplistic approach assuming an average rock mass structural condition.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
In this paper, geotechnical and structural data from an open pit mine were used to generate spatially constrained large-scale DFN models. Data from geotechnical boreholes were used to develop 3D block models of volumetric fracture intensity (P32), using a geostatistical simulation method. The block models were used for spatially constraining 3D geo-cellular DFN models. The fracture centers within the DFN cellular grid are controlled by the cellular P32 values, estimated using the 3D block model. The joints' orientation was bootstrapped from drillhole loggings whereas trace length data, collected across the pit area, were used to further calibrate the geo-cellular models. The resulting DFN models represented the spatial variation of fracture geometry and intensity along the pit area and were employed for numerical modeling of step-path slope stability analysis of the pit slopes. 2D cross sections of the 3D DFN models were embedded into 2D finite element models of the pit slope for a probabilistic step path stability analysis. In addition, the DFN models were used to estimate the rock bridge percentage along the pit walls which was ultimately used to assess the composite rock mass strength for a limit equilibrium analysis of the overall pit slope. The results of the two numerical approaches were compared to a simplistic approach assuming an average rock mass structural condition.
Mathe, Lewis; Ferentinou, Maria; Esmaeili, Kamran
Use of stochastic DFN-DEM modelling for overall and inter-ramp slope stability analysis Conference
The Third International Discrete Fracture Network Engineering Conference- DFNE 2022, Santa Fe, USA 2022.
@conference{Mathe2022,
title = {Use of stochastic DFN-DEM modelling for overall and inter-ramp slope stability analysis},
author = {Lewis Mathe and Maria Ferentinou and Kamran Esmaeili },
year = {2022},
date = {2022-06-30},
urldate = {2022-06-30},
organization = {The Third International Discrete Fracture Network Engineering Conference- DFNE 2022, Santa Fe, USA},
abstract = {Geometric Discrete Fracture Network (DFN) models are constructed in this research project, where limited and insufficient field structural data were exploited statistically in terms of observed parameter distribution at the mine-scale perspective. The construction of a DFN model employing a disaggregate technique was suggested by descriptive statistics of intensities (P10), orientation and length. The DFN model was validated by field data. The calibrated DFN model was then exported in a two dimensional Distinct Element Model (DEM) to assess the stability of a critical pit wall where the multiple block failures were recorded at the bench and inter-ramp slope scale.
The DFN-DEM approach adopted in this study could model the slope stability state using the shear strength reduction (SSR) method. According to the results of the simulations, the critical strength reduction factor (SRF) values derived from each numerical simulation were highly dependent on the DFN geometry configuration, and the trace length. To quantify the uncertainty associated with the assumptions established during the DFN-DEM modelling process, a series of models were simulated with different DFN configuration. At the inter-ramp slope scale, the simulations were predominantly unstable (SRF≤1.0) over 30 DFN realisations. The probability of failure (PoF) averaged 70% with a mean safety factor of 0.93 for the simulated DEM slope models while using the LE method, an average PoF of 40% and a mean safety factor of 1.08 was estimated.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Geometric Discrete Fracture Network (DFN) models are constructed in this research project, where limited and insufficient field structural data were exploited statistically in terms of observed parameter distribution at the mine-scale perspective. The construction of a DFN model employing a disaggregate technique was suggested by descriptive statistics of intensities (P10), orientation and length. The DFN model was validated by field data. The calibrated DFN model was then exported in a two dimensional Distinct Element Model (DEM) to assess the stability of a critical pit wall where the multiple block failures were recorded at the bench and inter-ramp slope scale.
The DFN-DEM approach adopted in this study could model the slope stability state using the shear strength reduction (SSR) method. According to the results of the simulations, the critical strength reduction factor (SRF) values derived from each numerical simulation were highly dependent on the DFN geometry configuration, and the trace length. To quantify the uncertainty associated with the assumptions established during the DFN-DEM modelling process, a series of models were simulated with different DFN configuration. At the inter-ramp slope scale, the simulations were predominantly unstable (SRF≤1.0) over 30 DFN realisations. The probability of failure (PoF) averaged 70% with a mean safety factor of 0.93 for the simulated DEM slope models while using the LE method, an average PoF of 40% and a mean safety factor of 1.08 was estimated.
The DFN-DEM approach adopted in this study could model the slope stability state using the shear strength reduction (SSR) method. According to the results of the simulations, the critical strength reduction factor (SRF) values derived from each numerical simulation were highly dependent on the DFN geometry configuration, and the trace length. To quantify the uncertainty associated with the assumptions established during the DFN-DEM modelling process, a series of models were simulated with different DFN configuration. At the inter-ramp slope scale, the simulations were predominantly unstable (SRF≤1.0) over 30 DFN realisations. The probability of failure (PoF) averaged 70% with a mean safety factor of 0.93 for the simulated DEM slope models while using the LE method, an average PoF of 40% and a mean safety factor of 1.08 was estimated.
Arasteh, Hossein; Esmaeili, Kamran; Saeedi, Golamreza; Ebrahimi, Mohammad Ali
Discontinuous Modeling of Roof Strata Caving in a Mechanized Longwall Mine in Tabas Coal Mine Journal Article
In: International Journal of Geomechanics, vol. 22, iss. 5, no. May 2022, 2022.
@article{RoofCave,
title = {Discontinuous Modeling of Roof Strata Caving in a Mechanized Longwall Mine in Tabas Coal Mine},
author = {Hossein Arasteh and Kamran Esmaeili and Golamreza Saeedi and Mohammad Ali Ebrahimi},
doi = { 10.1061/(ASCE)GM.1943-5622.0002337},
year = {2022},
date = {2022-03-08},
journal = {International Journal of Geomechanics},
volume = {22},
number = {May 2022},
issue = {5},
abstract = {A better understanding of the interaction between shield supports, roof, and coal strata, along with the prediction of roof deformation and caving behavior is essential for successful longwall mining operations. The stability of a longwall face can be improved by caving of the roof strata, which can lead to decreasing the load on the powered supports and the magnitude of stress on the abutment pillar of the longwall face. This paper presents the application of the synthetic rock mass (SRM) numerical modeling approach for the simulation of the roof caving mechanism and the load on shield supports in a longwall coal mine in Iran. Using geological and geotechnical data from the mine, a two-dimensional SRM, based on the incorporation of a discrete fracture network into a bonded particle model, was used to simulate the coal seam and the jointed roof strata. The results of the numerical models were validated by comparing the span of first caving, caving height, and load on the support shields to field measurements, observations, and previous studies. The proposed numerical approach can be utilized as a reliable tool in the design and safe operation of a longwall panel with a better understanding of the caved, damaged, and deformed zones above it.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
A better understanding of the interaction between shield supports, roof, and coal strata, along with the prediction of roof deformation and caving behavior is essential for successful longwall mining operations. The stability of a longwall face can be improved by caving of the roof strata, which can lead to decreasing the load on the powered supports and the magnitude of stress on the abutment pillar of the longwall face. This paper presents the application of the synthetic rock mass (SRM) numerical modeling approach for the simulation of the roof caving mechanism and the load on shield supports in a longwall coal mine in Iran. Using geological and geotechnical data from the mine, a two-dimensional SRM, based on the incorporation of a discrete fracture network into a bonded particle model, was used to simulate the coal seam and the jointed roof strata. The results of the numerical models were validated by comparing the span of first caving, caving height, and load on the support shields to field measurements, observations, and previous studies. The proposed numerical approach can be utilized as a reliable tool in the design and safe operation of a longwall panel with a better understanding of the caved, damaged, and deformed zones above it.
Gharebaghi, Amin; Mostafavi, Mir Abolfazl; Larouche, Christian; Esmaeili, Kamran; Grenon, Martin
Precise indoor localization and mapping using mobile laser scanners: A scoping review Journal Article
In: Geomatica, 2022.
@article{LiDAR,
title = {Precise indoor localization and mapping using mobile laser scanners: A scoping review},
author = {Amin Gharebaghi and Mir Abolfazl Mostafavi and Christian Larouche and Kamran Esmaeili and Martin Grenon},
doi = {https://doi.org/10.1139/geomat-2021-0011},
year = {2022},
date = {2022-02-17},
urldate = {2022-02-17},
journal = {Geomatica},
abstract = {Indoor localization and mapping are essential for a wide range of applications. The absence of GPS signals in indoor environments such as buildings, caves, tunnels, etc., brings significant challenges for applications where accurate positioning (i.e., centimeter- level accuracy) is required. This paper presents a scoping review of the most recent studies on precise indoor localization and mapping using mobile technologies, specifically mobile laser scanners. The scoping review allows comprehensive and structured review of the literature to maximize the capture of relevant information and provide reproducible results. We extracted and reported a range of information from the selected articles published since 2009 with the goal of identifying the most frequently used sensors and methods of fusing their collected observations. The results show that in the majority of studies, LiDAR is the core sensor and IMUs with 75% and Odometers with 67% magnitude, are the main sensors integrated with the LiDAR system to enhance localization precision. In addition, the classical Iterative Closest Point (ICP) algorithm with about 60% frequency, and Extended Kalman Filter (EKF) method with over 40% frequency, are the main algorithms used for the scan matching and fusion of different sensors' data, respectively. This scoping review has also revealed the lack of mapping-systems calibration as the main limitation in over 70% of the papers analyzed.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Indoor localization and mapping are essential for a wide range of applications. The absence of GPS signals in indoor environments such as buildings, caves, tunnels, etc., brings significant challenges for applications where accurate positioning (i.e., centimeter- level accuracy) is required. This paper presents a scoping review of the most recent studies on precise indoor localization and mapping using mobile technologies, specifically mobile laser scanners. The scoping review allows comprehensive and structured review of the literature to maximize the capture of relevant information and provide reproducible results. We extracted and reported a range of information from the selected articles published since 2009 with the goal of identifying the most frequently used sensors and methods of fusing their collected observations. The results show that in the majority of studies, LiDAR is the core sensor and IMUs with 75% and Odometers with 67% magnitude, are the main sensors integrated with the LiDAR system to enhance localization precision. In addition, the classical Iterative Closest Point (ICP) algorithm with about 60% frequency, and Extended Kalman Filter (EKF) method with over 40% frequency, are the main algorithms used for the scan matching and fusion of different sensors' data, respectively. This scoping review has also revealed the lack of mapping-systems calibration as the main limitation in over 70% of the papers analyzed.
2021
de Kooker, Leon; Ferentinou, Maria; Musonda, Innocent; Esmaeili, Kamran
Investigation of the stability of a fly ash pond facility using 2D and 3D slope stability analysis Conference
Global Tailings Standards and Opportunities for the Mine of the Future, THE SOUTHERN AFRICAN INSTITUTE OF MINING AND METALLURGY, 2021.
@conference{Tailingdam,
title = {Investigation of the stability of a fly ash pond facility using 2D and 3D slope stability analysis },
author = {Leon de Kooker and Maria Ferentinou and Innocent Musonda and Kamran Esmaeili },
year = {2021},
date = {2021-11-24},
urldate = {2021-11-24},
booktitle = {Global Tailings Standards and Opportunities for the Mine of the Future},
publisher = {THE SOUTHERN AFRICAN INSTITUTE OF MINING AND METALLURGY},
abstract = {This paper investigates the pore pressure regime, on the calculated factor of safety for the slopes of a fly ash storage facility. A fly ash pond case study was used to compare the factor of safety for both hydrostatic conditions and seepage occurring conditions, considering static and pseudo-static scenarios. The results from the two-dimensional limit equilibrium slope stability analysis, and also finite element seepage analysis are compared, while the seepage field under different scenario is simulated and the position of the phreatic surface is obtained. The safety factors calculated are compared with the static conditions levels, and are lower at 25%. A three-dimensional model was further created from extrusion of the two-dimensional sections and the failure surface and corresponding factor of safety in the developed models was further compared. The results from 3D analysis present a systematic increase in safety factor values which ranges from 13% to 25%.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
This paper investigates the pore pressure regime, on the calculated factor of safety for the slopes of a fly ash storage facility. A fly ash pond case study was used to compare the factor of safety for both hydrostatic conditions and seepage occurring conditions, considering static and pseudo-static scenarios. The results from the two-dimensional limit equilibrium slope stability analysis, and also finite element seepage analysis are compared, while the seepage field under different scenario is simulated and the position of the phreatic surface is obtained. The safety factors calculated are compared with the static conditions levels, and are lower at 25%. A three-dimensional model was further created from extrusion of the two-dimensional sections and the failure surface and corresponding factor of safety in the developed models was further compared. The results from 3D analysis present a systematic increase in safety factor values which ranges from 13% to 25%.
Bhuiyan, Mahadi; Esmaeili, Kamran; Ordonez-Calderon, Juan C.
Evaluation of rock characterization tests as geometallurgical predictors of Bond Work Index at the Tasiast Mine, Mauritania Journal Article
In: Minerals Engineering, vol. 175, 2021.
@article{geometallurgy,
title = {Evaluation of rock characterization tests as geometallurgical predictors of Bond Work Index at the Tasiast Mine, Mauritania},
author = {Mahadi Bhuiyan and Kamran Esmaeili and Juan C. Ordonez-Calderon},
doi = {https://doi.org/10.1016/j.mineng.2021.107293},
year = {2021},
date = {2021-11-05},
urldate = {2021-11-05},
journal = {Minerals Engineering},
volume = {175},
abstract = {This paper presents a geometallurgical study for predicting ore grindability at Tasiast Gold Mine. Drill core samples of main gold-bearing lithologies were subjected to three phases of testing for characterization of physicomechanical, geochemical, mineralogical, and textural rock properties. In phase one, a set of physiomechanical and geochemical properties were measured using rapid and portable rock characterization tests. The measured properties include surface rebound hardness (Leeb hardness test), multi-element geochemistry (portable XRF test), acoustic wave velocity, and strength index (Point load test). In the second phase, the rock samples were subjected to more time-consuming and expensive micro-scale tests including mineralogical characterization by XRD and textural classification by petrographic analysis of thin sections. Finally in the third phase, the core samples were used for Bond ball mill work index (BWI) test to assess their grinding behaviour. Geometallurgical associations were identified between grindability and the geometallurgical test predictors using principal components analytics and K-means clustering. These associations were then used for fitting predictive models for BWI using multiple linear regression. Inferential tests were applied to evaluate how well micro-scale (phase 2) and drill core-scale (phase 1) properties can predict BWI behaviour, and how these predictions capture important geometallurgical relationships to BWI. The best BWI predictive model was considered by assessing statistical fit, testing speed, relative cost, and portability and amenability of the testing tool to the field. Accordingly, at the Tasiast mine multi-element geochemistry and lithological textural characteristics are the top
two predictors of BWI.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This paper presents a geometallurgical study for predicting ore grindability at Tasiast Gold Mine. Drill core samples of main gold-bearing lithologies were subjected to three phases of testing for characterization of physicomechanical, geochemical, mineralogical, and textural rock properties. In phase one, a set of physiomechanical and geochemical properties were measured using rapid and portable rock characterization tests. The measured properties include surface rebound hardness (Leeb hardness test), multi-element geochemistry (portable XRF test), acoustic wave velocity, and strength index (Point load test). In the second phase, the rock samples were subjected to more time-consuming and expensive micro-scale tests including mineralogical characterization by XRD and textural classification by petrographic analysis of thin sections. Finally in the third phase, the core samples were used for Bond ball mill work index (BWI) test to assess their grinding behaviour. Geometallurgical associations were identified between grindability and the geometallurgical test predictors using principal components analytics and K-means clustering. These associations were then used for fitting predictive models for BWI using multiple linear regression. Inferential tests were applied to evaluate how well micro-scale (phase 2) and drill core-scale (phase 1) properties can predict BWI behaviour, and how these predictions capture important geometallurgical relationships to BWI. The best BWI predictive model was considered by assessing statistical fit, testing speed, relative cost, and portability and amenability of the testing tool to the field. Accordingly, at the Tasiast mine multi-element geochemistry and lithological textural characteristics are the top
two predictors of BWI.
two predictors of BWI.
Bamford, Thomas; Esmaeili, Kamran; Schoellig, Angela
A deep learning approach for rock fragmentation analysis Journal Article
In: Int. Journal Rock Mechanics and Mining Sciences, vol. 145, no. 104839, 2021.
@article{DeepLearning,
title = {A deep learning approach for rock fragmentation analysis},
author = {Thomas Bamford and Kamran Esmaeili and Angela Schoellig },
doi = {https://doi.org/10.1016/j.ijrmms.2021.104839},
year = {2021},
date = {2021-06-21},
urldate = {2021-06-21},
journal = {Int. Journal Rock Mechanics and Mining Sciences},
volume = {145},
number = {104839},
abstract = {In mining operations, blast-induced rock fragmentation affects the productivity and efficiency of downstream operations including digging, hauling, crushing, and grinding. Continuous measurement of rock fragmentation is essential for optimizing blast design. Current methods of rock fragmentation analysis rely on either physical screening of blasted rock material or image analysis of the blasted muckpiles; both are time consuming. This study aims to present and evaluate the measurement of rock fragmentation using deep learning strategies. A deep neural network (DNN) architecture was used to predict characteristic sizes of rock fragments from a 2D image of a muckpile. The data set used for training the DNN model is composed of 61,853 labelled images of blasted rock fragments. An exclusive data set of 1,263 labelled images were used to test the DNN model. The percent error for coarse characteristic size prediction ranges within 25% when evaluated using the test set. Model validation on orthomosaics for two muckpiles shows that the deep learning method achieves a good accuracy (lower mean percent error) compared to manual image labelling. Validation on screened piles shows that the DNN model prediction is similar to manual labelling accuracy when compared with sieving analysis.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In mining operations, blast-induced rock fragmentation affects the productivity and efficiency of downstream operations including digging, hauling, crushing, and grinding. Continuous measurement of rock fragmentation is essential for optimizing blast design. Current methods of rock fragmentation analysis rely on either physical screening of blasted rock material or image analysis of the blasted muckpiles; both are time consuming. This study aims to present and evaluate the measurement of rock fragmentation using deep learning strategies. A deep neural network (DNN) architecture was used to predict characteristic sizes of rock fragments from a 2D image of a muckpile. The data set used for training the DNN model is composed of 61,853 labelled images of blasted rock fragments. An exclusive data set of 1,263 labelled images were used to test the DNN model. The percent error for coarse characteristic size prediction ranges within 25% when evaluated using the test set. Model validation on orthomosaics for two muckpiles shows that the deep learning method achieves a good accuracy (lower mean percent error) compared to manual image labelling. Validation on screened piles shows that the DNN model prediction is similar to manual labelling accuracy when compared with sieving analysis.
Cudjoe, Francisca; Esmaeili, Kamran
A stochastic spatial modeling approach for pit slope stability analysis using 3D Limit Equilibrium Analysis Conference
RocScience International Conference 2021, 2021.
@conference{RICAB42,
title = {A stochastic spatial modeling approach for pit slope stability analysis using 3D Limit Equilibrium Analysis},
author = {Francisca Cudjoe and Kamran Esmaeili},
year = {2021},
date = {2021-04-21},
urldate = {2021-04-21},
publisher = {RocScience International Conference 2021},
abstract = {This paper presents a stochastic spatial modelling approach to evaluate the stability analysis of a pit slope in an open pit mine in Canada. More than 200 km of geotechnical borehole data drilled and logged in the mine area was used to develop 3D block models of Rock Mass Rating (RMR) and Uniaxial Compressive Strength (UCS), using stochastic Sequential Gaussian Simulation (SGS) method. The pit design was then excavated into the 3D RMR block models. The block models of the geotechnical attributes (UCS and RMR) were then embedded as input into discretized 3D limit equilibrium models created in SLIDE 3D to conduct a stochastic stability analysis of the pit slope. The modeling results were used to investigate possibilities for optimization of the pit slope design.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
This paper presents a stochastic spatial modelling approach to evaluate the stability analysis of a pit slope in an open pit mine in Canada. More than 200 km of geotechnical borehole data drilled and logged in the mine area was used to develop 3D block models of Rock Mass Rating (RMR) and Uniaxial Compressive Strength (UCS), using stochastic Sequential Gaussian Simulation (SGS) method. The pit design was then excavated into the 3D RMR block models. The block models of the geotechnical attributes (UCS and RMR) were then embedded as input into discretized 3D limit equilibrium models created in SLIDE 3D to conduct a stochastic stability analysis of the pit slope. The modeling results were used to investigate possibilities for optimization of the pit slope design.
Zhao, Yilin; Esmaeili, Kamran
Developing spatially constrained Discrete Fracture Network (DFN) models for a stochastic pit slope stability analysis Conference
RocScience International Conference 2021, 2021.
@conference{RICAB43,
title = {Developing spatially constrained Discrete Fracture Network (DFN) models for a stochastic pit slope stability analysis},
author = {Yilin Zhao and Kamran Esmaeili},
year = {2021},
date = {2021-04-21},
urldate = {2021-04-21},
publisher = {RocScience International Conference 2021},
abstract = {In this paper, data from an open pit mine in Western Africa were used to generate spatially constrained large-scale DFN models. Structural data from geotechnical boreholes and borehole tele-viewers, drilled and logged in the pit area, were used to develop stochastic 3D block models of volumetric fracture intensity (P32) using Sequential Gaussian Simulation. A geocellular 3D DFN model with finite volume of cellular grid elements were then created and spatially constrained based on the 3D block models of P32. This approach ensures that the fracture centers in the DFN generation are under the influence of cellular P32 values at corresponding locations, estimated based on the 3D block model of volumetric fracture intensity. The joint orientations were bootstrapped based on the joint orientations observed along the drillholes. The geocellular DFN models were further calibrated using joint trace mapping data, collected across the pit area using a drone-based aerial photogrammetry. The resulting DFN models are expected to accurately reflect the spatial variation of fracture geometry and intensity along the pit area. Finally, 2D cross sections of the generated DFN models were incorporated into 2D finite element models of the pit slope for a stochastic slope stability analysis. The results of the stochastic modelling were compared to a simplistic approach assuming an average rock mass structural condition.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
In this paper, data from an open pit mine in Western Africa were used to generate spatially constrained large-scale DFN models. Structural data from geotechnical boreholes and borehole tele-viewers, drilled and logged in the pit area, were used to develop stochastic 3D block models of volumetric fracture intensity (P32) using Sequential Gaussian Simulation. A geocellular 3D DFN model with finite volume of cellular grid elements were then created and spatially constrained based on the 3D block models of P32. This approach ensures that the fracture centers in the DFN generation are under the influence of cellular P32 values at corresponding locations, estimated based on the 3D block model of volumetric fracture intensity. The joint orientations were bootstrapped based on the joint orientations observed along the drillholes. The geocellular DFN models were further calibrated using joint trace mapping data, collected across the pit area using a drone-based aerial photogrammetry. The resulting DFN models are expected to accurately reflect the spatial variation of fracture geometry and intensity along the pit area. Finally, 2D cross sections of the generated DFN models were incorporated into 2D finite element models of the pit slope for a stochastic slope stability analysis. The results of the stochastic modelling were compared to a simplistic approach assuming an average rock mass structural condition.
Tang, Mingliang; Esmaeili, Kamran
In: Remote Sensing, vol. 13, no. 8, 2021.
@article{Heap-CNN,
title = {Heap Leach Pad Surface Moisture Monitoring Using Drone-Based Aerial Images and Convolutional Neural Networks: A Case Study at the El Gallo Mine, Mexico},
author = {Mingliang Tang and Kamran Esmaeili},
url = {https://www.mdpi.com/2072-4292/13/8/1420},
doi = {https://doi.org/10.3390/rs13081420},
year = {2021},
date = {2021-04-02},
urldate = {2021-04-02},
journal = {Remote Sensing},
volume = {13},
number = {8},
abstract = {An efficient metal recovery in heap leach operations relies on uniform distribution of leaching reagent solution over the heap leach pad surface. However, the current practices for heap leach pad (HLP) surface moisture monitoring often rely on manual inspection, which is labor-intensive, time-consuming, discontinuous, and intermittent. In order to complement the manual monitoring process and reduce the frequency of exposing technical manpower to the hazardous leaching reagent (e.g., dilute cyanide solution in gold leaching), this manuscript describes a case study of implementing an HLP surface moisture monitoring method based on drone-based aerial images and convolutional neural networks (CNNs). Field data collection was conducted on a gold HLP at the El Gallo mine, Mexico. A commercially available hexa-copter drone was equipped with one visible-light (RGB) camera and one thermal infrared sensor to acquire RGB and thermal images from the HLP surface. The collected data had high spatial and temporal resolutions. The high-quality aerial images were used to generate surface moisture maps of the HLP based on two CNN approaches. The generated maps provide direct visualization of the different moisture zones across the HLP surface, and such information can be used to detect potential operational issues related to distribution of reagent solution and to facilitate timely decision making in heap leach operations.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
An efficient metal recovery in heap leach operations relies on uniform distribution of leaching reagent solution over the heap leach pad surface. However, the current practices for heap leach pad (HLP) surface moisture monitoring often rely on manual inspection, which is labor-intensive, time-consuming, discontinuous, and intermittent. In order to complement the manual monitoring process and reduce the frequency of exposing technical manpower to the hazardous leaching reagent (e.g., dilute cyanide solution in gold leaching), this manuscript describes a case study of implementing an HLP surface moisture monitoring method based on drone-based aerial images and convolutional neural networks (CNNs). Field data collection was conducted on a gold HLP at the El Gallo mine, Mexico. A commercially available hexa-copter drone was equipped with one visible-light (RGB) camera and one thermal infrared sensor to acquire RGB and thermal images from the HLP surface. The collected data had high spatial and temporal resolutions. The high-quality aerial images were used to generate surface moisture maps of the HLP based on two CNN approaches. The generated maps provide direct visualization of the different moisture zones across the HLP surface, and such information can be used to detect potential operational issues related to distribution of reagent solution and to facilitate timely decision making in heap leach operations.
2020
Ardestani, Akbar; Amini, Medi; Esmaeili, Kamran
A two-dimensional limit equilibrium computer code for analysis of complex toppling slope failures Journal Article
In: Journal of Rock Mechanics and Geotechnical Engineering, 2020.
@article{Ardestani2020,
title = {A two-dimensional limit equilibrium computer code for analysis of complex toppling slope failures},
author = {Akbar Ardestani and Medi Amini and Kamran Esmaeili},
doi = {https://doi.org/10.1016/j.jrmge.2020.04.006},
year = {2020},
date = {2020-10-08},
urldate = {2020-10-08},
journal = {Journal of Rock Mechanics and Geotechnical Engineering},
abstract = {Evaluation of blocky or layered rock slopes against toppling failures has remained of great concern for engineers in various rock mechanics projects. Several step-by-step analytical solutions have been developed for analyzing these types of slope failures. However, manual application of these analytical solutions for real case studies can be time-consuming, complicated, and in certain cases even impossible. This study will first examine existing methods for toppling failure analyses that are reviewed, modified and generalized to consider the effects of a wide range of external and dead loads on slope stability. Next, based on the generalized presented formulae, a Windows form computer code is programmed using Visual C# for analysis of common types of toppling failures. Input parameters, including slope geometry, joint sets parameters, rock and soil properties, ground water level, dynamic loads, support anchor loads as well as magnitudes and forms of external forces, are first loaded into the code. The input data are then saved and used to graphically draw the slope model. This is followed by automatic identification of the toppling failure mode and a deterministic analysis of the slope stability against this failure mode. The results are presented using a graphical approach. The developed code allows probabilistic introduction of the input parameters via probability distribution functions (PDFs) and thus a probabilistic analysis of the toppling failure modes using Monte-Carlo simulation technique. This allows calculation of the probability of slope failure. Finally, several published case studies and typical examples are analyzed with the developed code. The outcomes are compared with those of the main references to assess the performance and robustness of the developed computer code. The comparisons demonstrate good agreement between the results.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Evaluation of blocky or layered rock slopes against toppling failures has remained of great concern for engineers in various rock mechanics projects. Several step-by-step analytical solutions have been developed for analyzing these types of slope failures. However, manual application of these analytical solutions for real case studies can be time-consuming, complicated, and in certain cases even impossible. This study will first examine existing methods for toppling failure analyses that are reviewed, modified and generalized to consider the effects of a wide range of external and dead loads on slope stability. Next, based on the generalized presented formulae, a Windows form computer code is programmed using Visual C# for analysis of common types of toppling failures. Input parameters, including slope geometry, joint sets parameters, rock and soil properties, ground water level, dynamic loads, support anchor loads as well as magnitudes and forms of external forces, are first loaded into the code. The input data are then saved and used to graphically draw the slope model. This is followed by automatic identification of the toppling failure mode and a deterministic analysis of the slope stability against this failure mode. The results are presented using a graphical approach. The developed code allows probabilistic introduction of the input parameters via probability distribution functions (PDFs) and thus a probabilistic analysis of the toppling failure modes using Monte-Carlo simulation technique. This allows calculation of the probability of slope failure. Finally, several published case studies and typical examples are analyzed with the developed code. The outcomes are compared with those of the main references to assess the performance and robustness of the developed computer code. The comparisons demonstrate good agreement between the results.
Tang, Mingliang; Esmaeili, Kamran
Mapping Surface Moisture of a Gold Heap Leach Pad at the El Gallo Mine Using an UAV and Thermal Imaging Journal Article
In: Mining, Metallurgy & Exploration (Springer), 2020.
@article{Tang2020,
title = {Mapping Surface Moisture of a Gold Heap Leach Pad at the El Gallo Mine Using an UAV and Thermal Imaging},
author = {Mingliang Tang and Kamran Esmaeili},
doi = {10.1007/s42461-020-00332-4},
year = {2020},
date = {2020-10-04},
urldate = {2020-10-04},
journal = {Mining, Metallurgy & Exploration (Springer)},
abstract = {An understanding of the spatial and temporal variations of surface moisture content over a heap leach pad (HLP) is essential for leaching production and to achieve a high ore recovery. The current practice of leach pad monitoring and data collection remains highly manual and labour intensive, and exposes technical staff to hazardous material (i.e., cyanide solution). To address these challenges, we propose using unmanned aerial vehicles (UAVs) equipped with thermal imaging sensors to remotely obtain high temporal and spatial resolution image data for monitoring the surface moisture distribution over HLPs. A field study was conducted over a sprinkler-irrigated HLP at El Gallo gold mine in Sinaloa state, Mexico, and the acquired data were used to derive an empirical relationship between the surface moisture content and the remotely sensed surface temperature using linear regression. Moreover, the data were used to generate moisture distribution maps of the entire HLP surface. In-situ samples were taken manually during the field experiments to measure the ground-truth material moisture at selected sampling locations. The results show a good agreement between the remote sensing method and the measured ground-truth samples.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
An understanding of the spatial and temporal variations of surface moisture content over a heap leach pad (HLP) is essential for leaching production and to achieve a high ore recovery. The current practice of leach pad monitoring and data collection remains highly manual and labour intensive, and exposes technical staff to hazardous material (i.e., cyanide solution). To address these challenges, we propose using unmanned aerial vehicles (UAVs) equipped with thermal imaging sensors to remotely obtain high temporal and spatial resolution image data for monitoring the surface moisture distribution over HLPs. A field study was conducted over a sprinkler-irrigated HLP at El Gallo gold mine in Sinaloa state, Mexico, and the acquired data were used to derive an empirical relationship between the surface moisture content and the remotely sensed surface temperature using linear regression. Moreover, the data were used to generate moisture distribution maps of the entire HLP surface. In-situ samples were taken manually during the field experiments to measure the ground-truth material moisture at selected sampling locations. The results show a good agreement between the remote sensing method and the measured ground-truth samples.
Bamford, Thomas; Medinac, Filip; Esmaeili, Kamran
Continuous Monitoring and Improvement of the Blasting Process in Open Pit Mines Using Unmanned Aerial Vehicle Techniques Journal Article
In: Remote Sensing, vol. 12, no. 17, pp. 2801, 2020.
@article{UAV-Blasting,
title = {Continuous Monitoring and Improvement of the Blasting Process in Open Pit Mines Using Unmanned Aerial Vehicle Techniques},
author = {Thomas Bamford and Filip Medinac and Kamran Esmaeili},
doi = {doi:10.3390/rs12172801},
year = {2020},
date = {2020-08-29},
urldate = {2020-08-29},
journal = {Remote Sensing},
volume = {12},
number = {17},
pages = {2801},
abstract = {The current techniques used for monitoring the blasting process in open pit mines are manual, intermittent and inefficient and can expose technical manpower to hazardous conditions. This study presents the application of unmanned aerial vehicle (UAV) systems for monitoring and improving the blasting process in open pit mines. Field experiments were conducted in different open pit mines to assess rock fragmentation, blast-induced damage on final pit walls, blast dynamics and the accuracy of blastholes including production and pre-split holes. The UAV-based monitoring was done in three different stages, including pre-blasting, blasting and post-blasting. In the pre-blasting stage, pit walls were mapped to collect structural data to predict in situ block size distribution and to develop as-built pit wall digital elevation models (DEM) to assess blast-induced damage. This was followed by mapping the production blasthole patterns implemented in the mine to investigate drillhole alignment. To monitor the blasting process, a high-speed camera was mounted on the UAV to investigate blast initiation, sequencing, misfired holes and stemming ejection. In the post-blast stage, the blasted rock pile (muck pile) was monitored to estimate fragmentation and assess muck pile configuration, heave and throw. The collected aerial data provide detailed information and high spatial and temporal resolution on the quality of the blasting process and significant opportunities for process improvement. The current challenges with regards to the application of UAVs for blasting process monitoring are discussed, and recommendations for obtaining the most value out of an UAV application are provided.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The current techniques used for monitoring the blasting process in open pit mines are manual, intermittent and inefficient and can expose technical manpower to hazardous conditions. This study presents the application of unmanned aerial vehicle (UAV) systems for monitoring and improving the blasting process in open pit mines. Field experiments were conducted in different open pit mines to assess rock fragmentation, blast-induced damage on final pit walls, blast dynamics and the accuracy of blastholes including production and pre-split holes. The UAV-based monitoring was done in three different stages, including pre-blasting, blasting and post-blasting. In the pre-blasting stage, pit walls were mapped to collect structural data to predict in situ block size distribution and to develop as-built pit wall digital elevation models (DEM) to assess blast-induced damage. This was followed by mapping the production blasthole patterns implemented in the mine to investigate drillhole alignment. To monitor the blasting process, a high-speed camera was mounted on the UAV to investigate blast initiation, sequencing, misfired holes and stemming ejection. In the post-blast stage, the blasted rock pile (muck pile) was monitored to estimate fragmentation and assess muck pile configuration, heave and throw. The collected aerial data provide detailed information and high spatial and temporal resolution on the quality of the blasting process and significant opportunities for process improvement. The current challenges with regards to the application of UAVs for blasting process monitoring are discussed, and recommendations for obtaining the most value out of an UAV application are provided.
Medinac, Filip; Bamford, Thomas; Hart, Matthew; Kowalczyk, Michal; Esmaeili, Kamran
Haul road monitoring in open pit mines using unmanned aerial vehicles Journal Article
In: Mining, Metallurgy & Exploration (Springer), vol. 37, pp. 1877-1883, 2020.
@article{Medinac2020,
title = {Haul road monitoring in open pit mines using unmanned aerial vehicles},
author = {Filip Medinac and Thomas Bamford and Matthew Hart and Michal Kowalczyk and Kamran Esmaeili},
doi = {DOI: 10.1007/s42461-020-00291-w},
year = {2020},
date = {2020-08-17},
urldate = {2020-08-17},
journal = {Mining, Metallurgy & Exploration (Springer)},
volume = {37},
pages = {1877-1883},
abstract = {Road quality has a significant impact on mining operations and is therefore identified as an area for improvement. Roads require constant monitoring because they are prone to defects caused by the daily wear and tear from heavy machinery and rough weather conditions. Insufficient or inadequate vigilance over the design and maintenance of haul roads can have detrimental consequences, negatively impacting productivity, costs and safety. In addition, maintenance issues that arise due to poor road conditions can increase costs and require countless man-hours that could be used more effectively. Conversely, good road conditions can improve safety and equipment efficiency, lower fuel consumption, increase tire life and reduce maintenance requirements. To this end, continuous haul-road monitoring and optimization efforts are required to improve operational efficiency at mine sites.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Road quality has a significant impact on mining operations and is therefore identified as an area for improvement. Roads require constant monitoring because they are prone to defects caused by the daily wear and tear from heavy machinery and rough weather conditions. Insufficient or inadequate vigilance over the design and maintenance of haul roads can have detrimental consequences, negatively impacting productivity, costs and safety. In addition, maintenance issues that arise due to poor road conditions can increase costs and require countless man-hours that could be used more effectively. Conversely, good road conditions can improve safety and equipment efficiency, lower fuel consumption, increase tire life and reduce maintenance requirements. To this end, continuous haul-road monitoring and optimization efforts are required to improve operational efficiency at mine sites.
Arasteh, Hossein; Saeedi, Golamreza; Ebrahimi, Mohammad Ali; Esmaeili, Kamran
A New Model for Calculation of the Plastic Compression Index and Porosity and Permeability of Gob Materials in Longwall Mining Journal Article
In: Geotechnical and Geological Engineering, 2020.
@article{GOB,
title = {A New Model for Calculation of the Plastic Compression Index and Porosity and Permeability of Gob Materials in Longwall Mining},
author = {Hossein Arasteh and Golamreza Saeedi and Mohammad Ali Ebrahimi and Kamran Esmaeili },
doi = {https://doi.org/10.1007/s10706-020-01444-w},
year = {2020},
date = {2020-07-06},
urldate = {2020-07-06},
journal = {Geotechnical and Geological Engineering},
abstract = {In coal longwall mining the accuracy of gob methane drainage system design, control of methane leakage into the longwall stope, and control of spontaneous combustion potential all depend on the porosity and permeability of the gob material. However, longwall mining conditions do not allow for any direct measurements of the gob material characteristics. Therefore, the simulation of gob material as a granular medium contributes to the ability to predict gob porosity. Many features of a granular medium, like gob material, are directly and indirectly affected by its particle size distribution. Fractal models are commonly used for porosity prediction which allows for a better understanding of how porosity can change with stress states based on the fragmentation size distribution of gob material. The plastic compression index (PCI), which describes the degree of compression of the broken material in gob, can be calculated using the physical properties of the broken material. However, using the existing fractal models, it is difficult to determine this parameter because of the multiplicity of parameters. In this research, a new fractal model was developed to facilitate the calculation of the PCI parameter and to increase the accuracy and validity of the results. Using data from the Tabas coal mine, the porosity-stress and permeability-stress curves were investigated for the gob material. The porosity results from this study agreed with those of Fan's gob compression model in a comparison. Although the obtained permeability results were within the range of previous studies, Berg's permeability model provided a better tool for estimation of the gob permeability.
},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In coal longwall mining the accuracy of gob methane drainage system design, control of methane leakage into the longwall stope, and control of spontaneous combustion potential all depend on the porosity and permeability of the gob material. However, longwall mining conditions do not allow for any direct measurements of the gob material characteristics. Therefore, the simulation of gob material as a granular medium contributes to the ability to predict gob porosity. Many features of a granular medium, like gob material, are directly and indirectly affected by its particle size distribution. Fractal models are commonly used for porosity prediction which allows for a better understanding of how porosity can change with stress states based on the fragmentation size distribution of gob material. The plastic compression index (PCI), which describes the degree of compression of the broken material in gob, can be calculated using the physical properties of the broken material. However, using the existing fractal models, it is difficult to determine this parameter because of the multiplicity of parameters. In this research, a new fractal model was developed to facilitate the calculation of the PCI parameter and to increase the accuracy and validity of the results. Using data from the Tabas coal mine, the porosity-stress and permeability-stress curves were investigated for the gob material. The porosity results from this study agreed with those of Fan's gob compression model in a comparison. Although the obtained permeability results were within the range of previous studies, Berg's permeability model provided a better tool for estimation of the gob permeability.
Cudjoe, Francisca; Esmaeili, Kamran
A 3D Spatial Model of Slope Mass Rating to Assess Potential Risk of Pit Wall Failure Conference
54th U.S. Rock Mechanics/Geomechanics Symposium, 28 June - 1 July 2020 American Rock Mechanics Association, 2020.
@conference{Cudjoe2020,
title = {A 3D Spatial Model of Slope Mass Rating to Assess Potential Risk of Pit Wall Failure},
author = {Francisca Cudjoe and Kamran Esmaeili},
url = {https://www.onepetro.org/conference-paper/ARMA-2020-1314},
year = {2020},
date = {2020-06-29},
urldate = {2020-06-29},
publisher = {American Rock Mechanics Association},
organization = {54th U.S. Rock Mechanics/Geomechanics Symposium, 28 June - 1 July 2020},
abstract = {In this study, a stochastic modelling approach is used to evaluate the potential risk of failure for an open pit mine in Quebec, Canada. More than 200 km of geotechnical borehole data drilled and logged in the pit area is used to develop 3D block models of Rock Mass Rating (RMR) using stochastic Sequential Gaussian Simulation (SGS) method. The pit design was excavated into the 3D RMR block models and the models were sectorized into zones based on the relative orientation of the pit walls. The 3D RMR block models were then converted to the block models of Slope Mass Rating (SMR) by applying joint orientation factors, and excavation method factor to the RMR blocks of each zone. Unlike the RMR block model, the 3D SMR model allows incorporating the influence of joint orientation and pit wall excavation method on the pit slope stability analysis. The method enables probabilistic analysis of geotechnical risks by identifying potential weak zones on the pit walls. The analysis could serve as a guide to make appropriate plans for mitigating the geotechnical risks.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
In this study, a stochastic modelling approach is used to evaluate the potential risk of failure for an open pit mine in Quebec, Canada. More than 200 km of geotechnical borehole data drilled and logged in the pit area is used to develop 3D block models of Rock Mass Rating (RMR) using stochastic Sequential Gaussian Simulation (SGS) method. The pit design was excavated into the 3D RMR block models and the models were sectorized into zones based on the relative orientation of the pit walls. The 3D RMR block models were then converted to the block models of Slope Mass Rating (SMR) by applying joint orientation factors, and excavation method factor to the RMR blocks of each zone. Unlike the RMR block model, the 3D SMR model allows incorporating the influence of joint orientation and pit wall excavation method on the pit slope stability analysis. The method enables probabilistic analysis of geotechnical risks by identifying potential weak zones on the pit walls. The analysis could serve as a guide to make appropriate plans for mitigating the geotechnical risks.
Medinac, Filip; Esmaeili, Kamran
A Stochastic-Deterministic Discrete Fracture Network Model for Pit Slope Stability Analysis Using UAV Collected Structural Data Conference
54th U.S. Rock Mechanics/Geomechanics Symposium, 28 June - 1 July, 2020 American Rock Mechanics Association, 2020.
@conference{Medinac2020b,
title = {A Stochastic-Deterministic Discrete Fracture Network Model for Pit Slope Stability Analysis Using UAV Collected Structural Data},
author = {Filip Medinac and Kamran Esmaeili},
year = {2020},
date = {2020-06-29},
urldate = {2020-06-29},
publisher = {American Rock Mechanics Association},
organization = {54th U.S. Rock Mechanics/Geomechanics Symposium, 28 June - 1 July, 2020},
abstract = {Existing data acquisition techniques for structural data collection are manual, time-consuming, and can expose technical manpower to hazardous conditions. Advances in Unmanned Aerial Vehicles (UAVs) technology allows collecting photogrammetry data of pit slopes. The aerial approach is fast, on demand and can improve the spatial and temporal resolution of the collected structural data. The collected data can be used to generate digital elevation models (DEMs) and point clouds to assess the slope configuration and to perform virtual mapping of the pit wall to collect detailed structural data.
This study presents the application of UAV technology to collect structural data from a pit wall, in Nevada, USA. The aerial photogrammetry method is used to generate a point cloud of the pit slope for virtual structural mapping. The structural mapping data is then integrated into the surveyed pit slope geometry to generate a conditioned Discrete Fracture Network (DFN) model. The discontinuities mapped on the slope surface are replicated in the DFN model, while behind the pit wall, a constrained stochastic model is used to describe the structural complexity of the rock mass. This combined stochastic-deterministic DFN model is used to conduct a kinematic stability analysis of the pit slope. The results are compared to the field observations of pit slope failure.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Existing data acquisition techniques for structural data collection are manual, time-consuming, and can expose technical manpower to hazardous conditions. Advances in Unmanned Aerial Vehicles (UAVs) technology allows collecting photogrammetry data of pit slopes. The aerial approach is fast, on demand and can improve the spatial and temporal resolution of the collected structural data. The collected data can be used to generate digital elevation models (DEMs) and point clouds to assess the slope configuration and to perform virtual mapping of the pit wall to collect detailed structural data.
This study presents the application of UAV technology to collect structural data from a pit wall, in Nevada, USA. The aerial photogrammetry method is used to generate a point cloud of the pit slope for virtual structural mapping. The structural mapping data is then integrated into the surveyed pit slope geometry to generate a conditioned Discrete Fracture Network (DFN) model. The discontinuities mapped on the slope surface are replicated in the DFN model, while behind the pit wall, a constrained stochastic model is used to describe the structural complexity of the rock mass. This combined stochastic-deterministic DFN model is used to conduct a kinematic stability analysis of the pit slope. The results are compared to the field observations of pit slope failure.
This study presents the application of UAV technology to collect structural data from a pit wall, in Nevada, USA. The aerial photogrammetry method is used to generate a point cloud of the pit slope for virtual structural mapping. The structural mapping data is then integrated into the surveyed pit slope geometry to generate a conditioned Discrete Fracture Network (DFN) model. The discontinuities mapped on the slope surface are replicated in the DFN model, while behind the pit wall, a constrained stochastic model is used to describe the structural complexity of the rock mass. This combined stochastic-deterministic DFN model is used to conduct a kinematic stability analysis of the pit slope. The results are compared to the field observations of pit slope failure.
Medinac, Filip; Esmaeili, Kamran
Integrating unmanned aerial vehicle photogrammetry in design compliance audits and structural modelling of pit walls Conference
International Symposium on Slope Stability in Open Pit Mining and Civil Engineering, 12-14 May 2020, Perth, Australia. ACG, 2020.
@conference{SlopeStability2020,
title = {Integrating unmanned aerial vehicle photogrammetry in design compliance audits and structural modelling of pit walls},
author = {Filip Medinac and Kamran Esmaeili},
year = {2020},
date = {2020-05-12},
urldate = {2020-05-12},
publisher = {ACG},
organization = {International Symposium on Slope Stability in Open Pit Mining and Civil Engineering, 12-14 May 2020, Perth, Australia.},
abstract = {Existing field data collection methods for pit slope assessment are manual, time consuming, and can expose technical manpower in hazardous conditions. Advances in unmanned aerial vehicles (UAVs) technology allows collecting photogrammetry data of pit slopes. This aerial approach is fast, on demand and can improve the spatial and temporal resolution of the collected data. The collected data can be used to generate digital elevation models (DEMs) and point clouds to assess the bench face angle and catch benches. Furthermore, virtual mapping can be used to collect detailed structural data.
This study presents the application of UAV technology to collect data at a pit wall, in Nevada, USA. A DEM is generated to conduct a design compliance audit of the pit slope. The aerial photogrammetry data is used to generate a point cloud of the slope for virtual structural mapping. The structural mapping data is integrated with the surveyed pit slope geometry to generate a conditioned Discrete Fracture Network (DFN) model. The discontinuities mapped on the slope surface are replicated in the DFN model, while behind the wall, a constrained stochastic model is used to describe the structural complexity of the rock mass. This combined deterministic-stochastic DFN model is used to conduct a kinematic stability analysis of the pit slope. The results are compared to the field observations of slope failure.
},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Existing field data collection methods for pit slope assessment are manual, time consuming, and can expose technical manpower in hazardous conditions. Advances in unmanned aerial vehicles (UAVs) technology allows collecting photogrammetry data of pit slopes. This aerial approach is fast, on demand and can improve the spatial and temporal resolution of the collected data. The collected data can be used to generate digital elevation models (DEMs) and point clouds to assess the bench face angle and catch benches. Furthermore, virtual mapping can be used to collect detailed structural data.
This study presents the application of UAV technology to collect data at a pit wall, in Nevada, USA. A DEM is generated to conduct a design compliance audit of the pit slope. The aerial photogrammetry data is used to generate a point cloud of the slope for virtual structural mapping. The structural mapping data is integrated with the surveyed pit slope geometry to generate a conditioned Discrete Fracture Network (DFN) model. The discontinuities mapped on the slope surface are replicated in the DFN model, while behind the wall, a constrained stochastic model is used to describe the structural complexity of the rock mass. This combined deterministic-stochastic DFN model is used to conduct a kinematic stability analysis of the pit slope. The results are compared to the field observations of slope failure.
This study presents the application of UAV technology to collect data at a pit wall, in Nevada, USA. A DEM is generated to conduct a design compliance audit of the pit slope. The aerial photogrammetry data is used to generate a point cloud of the slope for virtual structural mapping. The structural mapping data is integrated with the surveyed pit slope geometry to generate a conditioned Discrete Fracture Network (DFN) model. The discontinuities mapped on the slope surface are replicated in the DFN model, while behind the wall, a constrained stochastic model is used to describe the structural complexity of the rock mass. This combined deterministic-stochastic DFN model is used to conduct a kinematic stability analysis of the pit slope. The results are compared to the field observations of slope failure.
Tang, Mingliang; Esmaeili, Kamran
Mapping Surface Moisture Distribution of Heap Leach Pad using Unmanned Aerial Vehicle Conference
MineXchange 2020 SME Annual Conference & Expo, Phoenix, AZ, USA SME, 2020.
@conference{SME2020,
title = {Mapping Surface Moisture Distribution of Heap Leach Pad using Unmanned Aerial Vehicle},
author = {Mingliang Tang and Kamran Esmaeili},
year = {2020},
date = {2020-02-24},
urldate = {2020-02-24},
publisher = {SME},
organization = {MineXchange 2020 SME Annual Conference & Expo, Phoenix, AZ, USA},
abstract = {Understanding of the spatial and temporal variations of surface moisture content over a heap leach pad (HLP) is essential for leaching production and to achieve a high ore recovery. The current practice of leach pad monitoring and data collection remains highly manual which is labor-intensive and exposes technical staffs to hazardous material (e.g. cyanide solution). To address these challenges, we propose using unmanned aerial vehicles (UAVs) equipped with thermal imaging sensors to remotely obtain image data with high temporal and spatial resolutions for monitoring surface moisture distribution over HLPs. A field study was
conducted over a sprinkler-irrigated HLP at El Gallo Mine, and the acquired data were used to derive an empirical relationship between the surface moisture content and the remotely sensed surface temperature using linear regression. Moreover, the data were used to generate moisture distribution maps over the entire HLP area. In-situ samples were taken manually during the field experiments to measure the ground-truth material moisture at selected sampling locations. The results show a good agreement between the remote sensing method and the measured ground-truth samples.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Understanding of the spatial and temporal variations of surface moisture content over a heap leach pad (HLP) is essential for leaching production and to achieve a high ore recovery. The current practice of leach pad monitoring and data collection remains highly manual which is labor-intensive and exposes technical staffs to hazardous material (e.g. cyanide solution). To address these challenges, we propose using unmanned aerial vehicles (UAVs) equipped with thermal imaging sensors to remotely obtain image data with high temporal and spatial resolutions for monitoring surface moisture distribution over HLPs. A field study was
conducted over a sprinkler-irrigated HLP at El Gallo Mine, and the acquired data were used to derive an empirical relationship between the surface moisture content and the remotely sensed surface temperature using linear regression. Moreover, the data were used to generate moisture distribution maps over the entire HLP area. In-situ samples were taken manually during the field experiments to measure the ground-truth material moisture at selected sampling locations. The results show a good agreement between the remote sensing method and the measured ground-truth samples.
conducted over a sprinkler-irrigated HLP at El Gallo Mine, and the acquired data were used to derive an empirical relationship between the surface moisture content and the remotely sensed surface temperature using linear regression. Moreover, the data were used to generate moisture distribution maps over the entire HLP area. In-situ samples were taken manually during the field experiments to measure the ground-truth material moisture at selected sampling locations. The results show a good agreement between the remote sensing method and the measured ground-truth samples.
Christoffersen, Peter; Esmaeili, Kamran; Rivard, Benoit; Feng, Jilu; Osinski, Gordon
Developing spectral ore-waste discrimination methods: A case study at the El Gallo Silver Deposit, Mexico Conference
AME Roundup, Vancouver, Canada, 2020.
@conference{Peter,
title = {Developing spectral ore-waste discrimination methods: A case study at the El Gallo Silver Deposit, Mexico},
author = {Peter Christoffersen and Kamran Esmaeili and Benoit Rivard and Jilu Feng and Gordon Osinski},
year = {2020},
date = {2020-01-22},
urldate = {2020-01-22},
organization = {AME Roundup, Vancouver, Canada,},
abstract = {Near real time differentiation of ore from waste in muck piles offers increased efficiency in mine production via mass sorting. Visible and near-inferred spectroscopy offers a powerful tool for detecting and identifying alteration mineralogy, indicative or associated with mineralization in several deposit types. To date, most methods using this technique require a laboratory setting or collecting field spectra on a point by point basis. This project evaluates spectral methods for discriminating between ore and waste rock, and identifies data reduction techniques which enable the rapid discrimination of areas of ore and waste rock. To test these methods, ~16m of identified ore and waste drill core samples from the El Gallo Silver deposit, an epithermal deposit in Sinaloa State of Mexico were scanned using hyperspectral visible and near infrared detectors. To “ground-truth” the spectral interpretations, petrographic and geochemical analysis were conducted to verify silver content, and the composition of alteration mineral assemblages in the drill core. Whole silver contents in the rock samples ranged between <1ppm to 3630 ppm Ag; while the alteration mineral assemblage was identified to consist of propylitic alteration, calcite veining, and quartz stockwork. The spectral images were transformed using principle component analysis (PCA). When compared to the geochemical and petrographic data, certain principle components were found to best describe the ore and waste material. In this mineralogical setting, PCA is effective at discriminating between the ore and waste material. However, more complex mineral systems, or finer differentiations between ore types remains a challenge.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Near real time differentiation of ore from waste in muck piles offers increased efficiency in mine production via mass sorting. Visible and near-inferred spectroscopy offers a powerful tool for detecting and identifying alteration mineralogy, indicative or associated with mineralization in several deposit types. To date, most methods using this technique require a laboratory setting or collecting field spectra on a point by point basis. This project evaluates spectral methods for discriminating between ore and waste rock, and identifies data reduction techniques which enable the rapid discrimination of areas of ore and waste rock. To test these methods, ~16m of identified ore and waste drill core samples from the El Gallo Silver deposit, an epithermal deposit in Sinaloa State of Mexico were scanned using hyperspectral visible and near infrared detectors. To “ground-truth” the spectral interpretations, petrographic and geochemical analysis were conducted to verify silver content, and the composition of alteration mineral assemblages in the drill core. Whole silver contents in the rock samples ranged between <1ppm to 3630 ppm Ag; while the alteration mineral assemblage was identified to consist of propylitic alteration, calcite veining, and quartz stockwork. The spectral images were transformed using principle component analysis (PCA). When compared to the geochemical and petrographic data, certain principle components were found to best describe the ore and waste material. In this mineralogical setting, PCA is effective at discriminating between the ore and waste material. However, more complex mineral systems, or finer differentiations between ore types remains a challenge.
2019
Haghgouei, Hadi; Kargar, Ali Reza; Amini, Medi; Esmaieli, Kamran
An analytical solution for analysis of toppling-slumping failure in rock slopes Journal Article
In: Engineering Geology, vol. 265, 2019.
@article{Haghgouei2019,
title = {An analytical solution for analysis of toppling-slumping failure in rock slopes},
author = {Hadi Haghgouei and Ali Reza Kargar and Medi Amini and Kamran Esmaieli},
doi = {10.1016/j.enggeo.2019.105396 },
year = {2019},
date = {2019-11-18},
urldate = {2019-11-18},
journal = {Engineering Geology},
volume = {265},
abstract = {When a slope consists of a horizontal weak rock or soil overlaid by some vertical strong slender rock columns, the slope is prone to a secondary toppling failure which is known as “toppling-slumping” failure. In the slope, rock columns impose pressure on the lower continuous weak rock or soil mass which leads to a differential settlement at the base of each rock column or a circular shear failure in the continuous mass. If the differential settlements reach a special threshold, rock column overturn and toppling-slumping will happen. However, the circular shear failure of the soil or weak rock mass which is a well-recognized failure mechanism will occur when shear stress in the slope exceeds its shear strength. In this paper, an attempt has been made to better understanding of the “toppling-slumping” failure mechanism through developing an analytical solution. For this purpose, first an Airy stress function is acquired for the lower continuous soil mass and its stress distribution and displacement are accordingly determined analytically due to the overlaid rock columns loads. Then, differential settlements at the base of all rock columns and shear stresses in the soil or weak rock mass are computed and compared with the toppling and shear failure criteria to assess slope failure. Finally, several design-charts are developed that determine the stability of a slope against toppling-slumping failure mechanism based on geometric characteristics of the slope and shear strength of its constituent material. Also, a Mathematica package code was developed to evaluate the stress distribution within the slope, and the safety factor. The results show that the vertical stress distribution plays a significant role in the settlement of the rock columns. Results also manifested that at a certain distance from the slope crest, the set of rock columns will act like a single column on a half-plane. The shear or toppling failure of the slope not only depends on the material's strength but also on the geometry and the normalized column distance.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
When a slope consists of a horizontal weak rock or soil overlaid by some vertical strong slender rock columns, the slope is prone to a secondary toppling failure which is known as “toppling-slumping” failure. In the slope, rock columns impose pressure on the lower continuous weak rock or soil mass which leads to a differential settlement at the base of each rock column or a circular shear failure in the continuous mass. If the differential settlements reach a special threshold, rock column overturn and toppling-slumping will happen. However, the circular shear failure of the soil or weak rock mass which is a well-recognized failure mechanism will occur when shear stress in the slope exceeds its shear strength. In this paper, an attempt has been made to better understanding of the “toppling-slumping” failure mechanism through developing an analytical solution. For this purpose, first an Airy stress function is acquired for the lower continuous soil mass and its stress distribution and displacement are accordingly determined analytically due to the overlaid rock columns loads. Then, differential settlements at the base of all rock columns and shear stresses in the soil or weak rock mass are computed and compared with the toppling and shear failure criteria to assess slope failure. Finally, several design-charts are developed that determine the stability of a slope against toppling-slumping failure mechanism based on geometric characteristics of the slope and shear strength of its constituent material. Also, a Mathematica package code was developed to evaluate the stress distribution within the slope, and the safety factor. The results show that the vertical stress distribution plays a significant role in the settlement of the rock columns. Results also manifested that at a certain distance from the slope crest, the set of rock columns will act like a single column on a half-plane. The shear or toppling failure of the slope not only depends on the material's strength but also on the geometry and the normalized column distance.
Esmaieli, Kamran
Improving the quality and quantity of geotechnical core logging data Conference
14th International Congress of Rock Mechanics Brazil, 2019.
@conference{Esmaeili2019,
title = {Improving the quality and quantity of geotechnical core logging data },
author = {Kamran Esmaieli},
year = {2019},
date = {2019-09-18},
urldate = {2019-09-18},
address = {Brazil},
organization = {14th International Congress of Rock Mechanics},
abstract = {Conventional geotechnical core logging method which is commonly used for field data col-lection is subjective, inconsistent and time consuming. This can increase the uncertainty in mine planning and design which can lead in higher risk of mining operation. This paper presents a portable multi-sensor core logging machine that has been designed and devel-oped by the author. Different sensors can be installed on the machine including: a digital camera, an Equotip Leeb hardness tester, an ultrasonic pulse velocity, a portable X-Ray fluorescent analyzer and a handheld 3D scanner. The collected data allows crating a high-resolution digital log of core samples. This can be beneficial for reducing logging incon-sistency and providing better resolution for characterizing rock structural, physical and mechanical properties. The high-resolution multivariate core logging data can be used for descriptive and predictive data analytics to better understand the inherent variability and complexity of rock mass geotechnical properties.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Conventional geotechnical core logging method which is commonly used for field data col-lection is subjective, inconsistent and time consuming. This can increase the uncertainty in mine planning and design which can lead in higher risk of mining operation. This paper presents a portable multi-sensor core logging machine that has been designed and devel-oped by the author. Different sensors can be installed on the machine including: a digital camera, an Equotip Leeb hardness tester, an ultrasonic pulse velocity, a portable X-Ray fluorescent analyzer and a handheld 3D scanner. The collected data allows crating a high-resolution digital log of core samples. This can be beneficial for reducing logging incon-sistency and providing better resolution for characterizing rock structural, physical and mechanical properties. The high-resolution multivariate core logging data can be used for descriptive and predictive data analytics to better understand the inherent variability and complexity of rock mass geotechnical properties.
Ohadi, Behrouz; Sun, Xi; Esmaieli, Kamran; Consens, Mariano
Predicting Blast-induced Outcomes Using Random Forest Models of Multi-year Blasting Data from an Open Pit Mine Journal Article
In: Bulletin of Engineering Geology and the Environment, 2019.
@article{Blasting,
title = {Predicting Blast-induced Outcomes Using Random Forest Models of Multi-year Blasting Data from an Open Pit Mine },
author = {Behrouz Ohadi and Xi Sun and Kamran Esmaieli and Mariano Consens},
doi = {10.1007/s10064-019-01566-3},
year = {2019},
date = {2019-06-21},
urldate = {2019-06-21},
journal = {Bulletin of Engineering Geology and the Environment},
abstract = {Rock fragmentation and movement are two important outcomes of the blasting process in open pit mines. They are influenced by blasting design parameters, as well as by the physical and geomechanical characteristics of the rock mass. This paper presents the results of analyzing multi-year blasting data from an open pit mine in Canada and proposes a predictive model for the blast-induced outcomes that incorporates both rock mass properties and blasting parameters. The analysis employed the Decision Tree (DT) and Random Forest (RF) models to determine influential parameters, confirming that the blast-induced fragmentation and movement are influenced by rock mass characteristics (i.e. intact rock strength and RQD), as well as by blasting design parameters. The Decision Tree model facilitates the visualization of geomechanical and blasting design parameters influencing the blast-induced outcomes. The robust Random Forest model provides prediction of blast-induced outcomes. The Decision Tree and Random Forest models make it possible to determine blasting design parameters that could be modified to achieve better blast-induced outcomes based on the rock mass conditions at the mine site. },
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Rock fragmentation and movement are two important outcomes of the blasting process in open pit mines. They are influenced by blasting design parameters, as well as by the physical and geomechanical characteristics of the rock mass. This paper presents the results of analyzing multi-year blasting data from an open pit mine in Canada and proposes a predictive model for the blast-induced outcomes that incorporates both rock mass properties and blasting parameters. The analysis employed the Decision Tree (DT) and Random Forest (RF) models to determine influential parameters, confirming that the blast-induced fragmentation and movement are influenced by rock mass characteristics (i.e. intact rock strength and RQD), as well as by blasting design parameters. The Decision Tree model facilitates the visualization of geomechanical and blasting design parameters influencing the blast-induced outcomes. The robust Random Forest model provides prediction of blast-induced outcomes. The Decision Tree and Random Forest models make it possible to determine blasting design parameters that could be modified to achieve better blast-induced outcomes based on the rock mass conditions at the mine site.
Bhuiyan, Mahadi; Esmaieli, Kamran; Ordóñez-Calderón, Juan Carlos
In: Minerals, vol. 9, no. 302, 2019.
@article{DataAnalytics,
title = {Application of Data Analytics Techniques to Establish Geometallurgical Relationships to Bond Work Index at the Paracutu Mine, Minas Gerais, Brazil},
author = {Mahadi Bhuiyan and Kamran Esmaieli and Juan Carlos Ordóñez-Calderón},
doi = {doi:10.3390/min9050302},
year = {2019},
date = {2019-05-16},
urldate = {2019-05-16},
journal = {Minerals},
volume = {9},
number = {302},
abstract = {Analysis of geometallurgical data is essential to building geometallurgical models that capture physical variability in the orebody and can be used for the optimization of mine planning and the prediction of milling circuit performance. However, multivariate complexity and compositional data constraints can make this analysis challenging. This study applies unsupervised and supervised learning to establish relationships between the Bond ball mill work index (BWI) and geomechanical, geophysical and geochemical variables for the Paracatu gold orebody. The regolith and fresh rock geometallurgical domains are established from two cluster sets resulting from K-means clustering of the first three principal component (PC) scores of isometric log-ratio (ilr) coordinates of geochemical data and standardized BWI, geomechanical and geophysical data. The first PC is attributed to weathering and reveals a strong relationship between BWI and rock strength and fracture intensity in the regolith. Random forest (RF) classification of BWI in the fresh rock identifies the greater importance of geochemical ilr balances relative to geomechanical and geophysical variables.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Analysis of geometallurgical data is essential to building geometallurgical models that capture physical variability in the orebody and can be used for the optimization of mine planning and the prediction of milling circuit performance. However, multivariate complexity and compositional data constraints can make this analysis challenging. This study applies unsupervised and supervised learning to establish relationships between the Bond ball mill work index (BWI) and geomechanical, geophysical and geochemical variables for the Paracatu gold orebody. The regolith and fresh rock geometallurgical domains are established from two cluster sets resulting from K-means clustering of the first three principal component (PC) scores of isometric log-ratio (ilr) coordinates of geochemical data and standardized BWI, geomechanical and geophysical data. The first PC is attributed to weathering and reveals a strong relationship between BWI and rock strength and fracture intensity in the regolith. Random forest (RF) classification of BWI in the fresh rock identifies the greater importance of geochemical ilr balances relative to geomechanical and geophysical variables.
2018
Amini, Mehdi; Sarfaraz, Hassan; Esmaieli, Kamran
Stability analysis of slopes with a potential of slide-head-toppling failure Journal Article
In: Int. J. Rock Mech. Min. Sci., vol. 112, pp. 108-121, 2018.
@article{SlopeStability,
title = {Stability analysis of slopes with a potential of slide-head-toppling failure},
author = {Mehdi Amini and Hassan Sarfaraz and Kamran Esmaieli},
doi = {https://doi.org/10.1016/j.ijrmms.2018.09.008},
year = {2018},
date = {2018-10-25},
urldate = {2018-10-25},
journal = {Int. J. Rock Mech. Min. Sci.},
volume = {112},
pages = {108-121},
abstract = {Toppling is a mode of slope instability that may occur in a wide range of layered or blocky rock masses. If this instability is caused by some natural or man-made external factors, it is termed a secondary toppling failure. One of the important modes of toppling instability is the slide-head-toppling failure. When a layer of soil or weak rock is overlaid by a blocky or layered rock mass, the combination of a toppling failure in the rock at the top and a circular sliding in the soil or weak rock in the bottom leads to this hybrid failure. The mechanism of this hybrid failure is extremely complicated and it may occur in some special geological conditions in civil engineering and mining projects. In this paper, the mechanisms of toppling failures, in general, and the mechanism of slide-head-toppling failure, in particular, are presented and described. Then, the outcomes of seven physical models, which were constructed with base friction materials and conducted under static loading conditions by a tilting table apparatus, are summarized to investigate this hybrid failure mechanism. Subsequently, a new step by step theoretical approach is proposed for assessment of this failure mode and a typical example is analyzed to better demonstrate the results. Finally, the results of the physical modeling are compared with the outcomes of the proposed theoretical method and an existing analytical approach. This comparison shows that there are acceptable agreements between the physical and analytical results. Hence, both proposed and existing analytical approaches can be used for the analysis of the slide-head-toppling slope failure.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Toppling is a mode of slope instability that may occur in a wide range of layered or blocky rock masses. If this instability is caused by some natural or man-made external factors, it is termed a secondary toppling failure. One of the important modes of toppling instability is the slide-head-toppling failure. When a layer of soil or weak rock is overlaid by a blocky or layered rock mass, the combination of a toppling failure in the rock at the top and a circular sliding in the soil or weak rock in the bottom leads to this hybrid failure. The mechanism of this hybrid failure is extremely complicated and it may occur in some special geological conditions in civil engineering and mining projects. In this paper, the mechanisms of toppling failures, in general, and the mechanism of slide-head-toppling failure, in particular, are presented and described. Then, the outcomes of seven physical models, which were constructed with base friction materials and conducted under static loading conditions by a tilting table apparatus, are summarized to investigate this hybrid failure mechanism. Subsequently, a new step by step theoretical approach is proposed for assessment of this failure mode and a typical example is analyzed to better demonstrate the results. Finally, the results of the physical modeling are compared with the outcomes of the proposed theoretical method and an existing analytical approach. This comparison shows that there are acceptable agreements between the physical and analytical results. Hence, both proposed and existing analytical approaches can be used for the analysis of the slide-head-toppling slope failure.
Bhuiyan, Mahadi; Esmaieli, Kamran
Investigating geometallurgical relationships by principal component analysis of compositional and non-compositional data Conference
Geometallurgy Conference 2018, August 6-8, Cape Town, South Africa, The South African Institute of Mining and Metallurgy 2018.
@conference{Geomet2018,
title = {Investigating geometallurgical relationships by principal component analysis of compositional and non-compositional data},
author = {Mahadi Bhuiyan and Kamran Esmaieli},
year = {2018},
date = {2018-08-06},
urldate = {2018-08-06},
booktitle = {Geometallurgy Conference 2018, August 6-8, Cape Town, South Africa},
organization = {The South African Institute of Mining and Metallurgy},
abstract = {Multivariate analysis of orebody data is an important first step in delineating spatial geometallurgical domains. Orebody data may include compositional, e.g., multi-element geochemistry, and non-compositional data, e.g., Bond Work index (BWI). Principal component analysis (PCA) is a statistical procedure that is useful for studying geometallurgical relationships. However, combining compositional and non-compositional data for PCA requires special techniques for the treatment of compositional data due to its relative nature. This study presents a PCA analysis using a data-set of integrated compositional and non-compositional variables obtained from different drilling campaigns at Paracatu mine, Brazil. PCA results show relationships between the BWI and geometallurgical variables, which vary spatially according to orebody characteristics and indicates different controls on grindability.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Multivariate analysis of orebody data is an important first step in delineating spatial geometallurgical domains. Orebody data may include compositional, e.g., multi-element geochemistry, and non-compositional data, e.g., Bond Work index (BWI). Principal component analysis (PCA) is a statistical procedure that is useful for studying geometallurgical relationships. However, combining compositional and non-compositional data for PCA requires special techniques for the treatment of compositional data due to its relative nature. This study presents a PCA analysis using a data-set of integrated compositional and non-compositional variables obtained from different drilling campaigns at Paracatu mine, Brazil. PCA results show relationships between the BWI and geometallurgical variables, which vary spatially according to orebody characteristics and indicates different controls on grindability.
Siamaki, Ali; Esmaieli, Kamran; Mohanty, Bibhu
Effects of Blast Induced Wave Propagation on the Degradation of Jointed Rock Masses Conference
2nd International Discrete Fracture Network Engineering Conference, June 20-22 Seattle, USA, 2018.
@conference{Siamaki2018c,
title = {Effects of Blast Induced Wave Propagation on the Degradation of Jointed Rock Masses},
author = {Ali Siamaki and Kamran Esmaieli and Bibhu Mohanty },
year = {2018},
date = {2018-06-26},
urldate = {2018-06-26},
address = { Seattle, USA},
organization = {2nd International Discrete Fracture Network Engineering Conference, June 20-22},
abstract = {In open pit mines, repetitive blast-induced ground vibration can increase the risk of pit wall instability due to strain accumulation along discontinuities. Presence of discrete fractures or fracture networks in a rock mass can influence the propagation of blast-induced shock waves in the rock and consequently the degradation of shear strength of the jointed rock mass. Quantification of blast-induced rock mass degradation is essential for prediction of potential risk of pit slope failure. This paper presents the results of a series of numerical experiments that examine the effects of ground vibration from a single row blast on a jointed rock mass, simulated using the Particle Flow Code (PFC2D). Discrete Fracture Network (DFN) was used to generate two joint sets in a rock block of 4 m x 8 m. Two different scenarios were considered: a) two orthogonal joint sets (one horizontal and one vertical set), b) two inclined joint sets. For each joint orientation scenario, five different fracture intensities (P 21 ) were generated, varying between 0.4 and 5 m -1 . Recorded blast vibration history from a quarry was applied to the 2D jointed rock mass samples and wave propagation was monitored along the rock blocks. Results show that the first fractures along the wave propagation path have experienced more degradation (damage) in the form of micro-crack generation along the fractures. Damage was only developed along the horizontal joint set for the rock mass model with orthogonal sets, whereas in the rock mass with two inclined joint sets, damage was accumulated on both joint sets. Rock mass degradation increases as the fracture intensity increase.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
In open pit mines, repetitive blast-induced ground vibration can increase the risk of pit wall instability due to strain accumulation along discontinuities. Presence of discrete fractures or fracture networks in a rock mass can influence the propagation of blast-induced shock waves in the rock and consequently the degradation of shear strength of the jointed rock mass. Quantification of blast-induced rock mass degradation is essential for prediction of potential risk of pit slope failure. This paper presents the results of a series of numerical experiments that examine the effects of ground vibration from a single row blast on a jointed rock mass, simulated using the Particle Flow Code (PFC2D). Discrete Fracture Network (DFN) was used to generate two joint sets in a rock block of 4 m x 8 m. Two different scenarios were considered: a) two orthogonal joint sets (one horizontal and one vertical set), b) two inclined joint sets. For each joint orientation scenario, five different fracture intensities (P 21 ) were generated, varying between 0.4 and 5 m -1 . Recorded blast vibration history from a quarry was applied to the 2D jointed rock mass samples and wave propagation was monitored along the rock blocks. Results show that the first fractures along the wave propagation path have experienced more degradation (damage) in the form of micro-crack generation along the fractures. Damage was only developed along the horizontal joint set for the rock mass model with orthogonal sets, whereas in the rock mass with two inclined joint sets, damage was accumulated on both joint sets. Rock mass degradation increases as the fracture intensity increase.
Siamaki, Ali; Esmaieli, Kamran; Mohanty, Bibhu
Effects of the Blast Vibration Frequency on the Shear Strength of the Joints and Pit Slope Stability Conference
Slope Stability 2018 Symposium, April 10-13 Sevilla, Spain., 2018.
@conference{Siamaki2018d,
title = {Effects of the Blast Vibration Frequency on the Shear Strength of the Joints and Pit Slope Stability },
author = {Ali Siamaki and Kamran Esmaieli and Bibhu Mohanty},
url = {https://omre-researchgroup.com/wp-content/uploads/2018/06/Paper-Slope-Stability-2018-Siamaki-et-al-revised.pdf},
year = {2018},
date = {2018-06-26},
urldate = {2018-06-26},
address = {Sevilla, Spain.},
organization = {Slope Stability 2018 Symposium, April 10-13},
abstract = {Blast induced vibrations are the blasting drawbacks which may cause instability in the adjacent and nearby structures. These vibrations could degrade the shear strength of the major discontinuities in the rock mass. Therefore, the bearing capacity of the rock mass may be reduced by the repetitive blasts which lead to the slope instabilities. In this paper, the recorded vibration histories of a single row blast with double deck boreholes in a quarry was considered. The blast histories were recorded at two stations, one on the same bench as the blasting (station S1) and another one on the upper bench (station S2). The results showed peak particle velocities at station S1 were influenced by the upper decks, while the bottom decks have more influence on the recorded peak particle velocities in S2. The dominant wave frequency at station S1 was 121 Hz, while it was 48 Hz at station S2. The recorded vibration loads at each station were applied to a stiff rock block of 2 m height ́ 1 m width, with a single inclined persistent joint, simulated by 2D Particle Flow Code (PFC2D). The vibration loads, recorded at the two stations, were separately applied to the jointed rock block for consecutive times to model the effect of repeated loading due to separate blasts. The numerical experiments allowed investigation of the interaction between blast-induced stress waves and the joint surface and quantification of the progressive accumulation of damage inflicted along the joint. The vibration-induced damage was equated with the number of micro-cracks generated along the joint surface. The results show that the frequency content has a significant impact on the degradation of joint shear strength in a rock mass. Higher micro-crack generation rate (degradation rate) was recorded along the joint surface
when subjected to the lower vibration wave frequency, recorded at station S2. The results show that the numerical method can be used as a diagnostic and predictive tool for characterization of blast-induced damage in open pit mines, when validated by careful blast vibration and slope monitoring programs.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Blast induced vibrations are the blasting drawbacks which may cause instability in the adjacent and nearby structures. These vibrations could degrade the shear strength of the major discontinuities in the rock mass. Therefore, the bearing capacity of the rock mass may be reduced by the repetitive blasts which lead to the slope instabilities. In this paper, the recorded vibration histories of a single row blast with double deck boreholes in a quarry was considered. The blast histories were recorded at two stations, one on the same bench as the blasting (station S1) and another one on the upper bench (station S2). The results showed peak particle velocities at station S1 were influenced by the upper decks, while the bottom decks have more influence on the recorded peak particle velocities in S2. The dominant wave frequency at station S1 was 121 Hz, while it was 48 Hz at station S2. The recorded vibration loads at each station were applied to a stiff rock block of 2 m height ́ 1 m width, with a single inclined persistent joint, simulated by 2D Particle Flow Code (PFC2D). The vibration loads, recorded at the two stations, were separately applied to the jointed rock block for consecutive times to model the effect of repeated loading due to separate blasts. The numerical experiments allowed investigation of the interaction between blast-induced stress waves and the joint surface and quantification of the progressive accumulation of damage inflicted along the joint. The vibration-induced damage was equated with the number of micro-cracks generated along the joint surface. The results show that the frequency content has a significant impact on the degradation of joint shear strength in a rock mass. Higher micro-crack generation rate (degradation rate) was recorded along the joint surface
when subjected to the lower vibration wave frequency, recorded at station S2. The results show that the numerical method can be used as a diagnostic and predictive tool for characterization of blast-induced damage in open pit mines, when validated by careful blast vibration and slope monitoring programs.
when subjected to the lower vibration wave frequency, recorded at station S2. The results show that the numerical method can be used as a diagnostic and predictive tool for characterization of blast-induced damage in open pit mines, when validated by careful blast vibration and slope monitoring programs.
Wang, Jianxiu; Yin, Yao; Esmaieli, Kamran
Numerical Simulations of Rock Blasting Damage Based on Laboratory Scale Experiments Journal Article
In: Journal of Geophysics and Engineering, pp. 39, 2018.
@article{Wang2018,
title = {Numerical Simulations of Rock Blasting Damage Based on Laboratory Scale Experiments},
author = {Jianxiu Wang and Yao Yin and Kamran Esmaieli },
url = {https://omre-researchgroup.com/wp-content/uploads/2018/06/Wangetal_2018_J._Geophys._Eng._10.1088_1742-2140_aacf17.pdf},
doi = {10.1088/1742-2140/aacf17},
year = {2018},
date = {2018-06-26},
urldate = {2018-06-26},
journal = {Journal of Geophysics and Engineering},
pages = {39},
abstract = {In order to study the damage induced by rock blasting, a numerical simulation method based on Johnson-Holmquist-Ⅱ(JH-2) damage model combined with Arbitrary Lagrange Euler (ALE) method is proposed. The process of dynamic breakage and damage evolution of Barre Granite is reproduced using explicit hydrocode, ANSYS/LS-DYNA, based on the prototype experiments in laboratory. The results show that both of the crack patterns and measured pressures are in agreement with the results from the lab-scale experiments. The attenuation curves of the pressure and particle peak velocity (PPV) along the radial direction are respectively determined, and the corresponding theoretical formulas are summarized together with the most suitable attenuation exponent (α ). In addition, comparisons of blasting tests separately carried out using DEM-SPH hybrid method and ALE/JH-2 method demonstrates similar crack patterns formed both in intact rock disks and jointed rock disks. In the jointed rock disk, the pressure sharply declines when crossing the joint surface, while the PPV close to the joint increases before going across the joint surface, representing the "weak transmission-strong reflection" effect of the joint surface. Different yield stresses of joint properties are further studied, which indicate that the joint with lower magnitude of yield stress can prevent more transmissions of waves crossing the joint surface. In future studies, the ALE method combined with JH-2 damage model can be applied to larger scale rock engineering problems to optimize the blasting design.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In order to study the damage induced by rock blasting, a numerical simulation method based on Johnson-Holmquist-Ⅱ(JH-2) damage model combined with Arbitrary Lagrange Euler (ALE) method is proposed. The process of dynamic breakage and damage evolution of Barre Granite is reproduced using explicit hydrocode, ANSYS/LS-DYNA, based on the prototype experiments in laboratory. The results show that both of the crack patterns and measured pressures are in agreement with the results from the lab-scale experiments. The attenuation curves of the pressure and particle peak velocity (PPV) along the radial direction are respectively determined, and the corresponding theoretical formulas are summarized together with the most suitable attenuation exponent (α ). In addition, comparisons of blasting tests separately carried out using DEM-SPH hybrid method and ALE/JH-2 method demonstrates similar crack patterns formed both in intact rock disks and jointed rock disks. In the jointed rock disk, the pressure sharply declines when crossing the joint surface, while the PPV close to the joint increases before going across the joint surface, representing the "weak transmission-strong reflection" effect of the joint surface. Different yield stresses of joint properties are further studied, which indicate that the joint with lower magnitude of yield stress can prevent more transmissions of waves crossing the joint surface. In future studies, the ALE method combined with JH-2 damage model can be applied to larger scale rock engineering problems to optimize the blasting design.
Medinac, Filip; Bamford, Thomas; Esmaieli, Kamran; Schoellig, Angela
Pre- and Post-Blast Rock Block Size Analysis Using UAV-based Data and Discrete Fracture Network Conference
2nd International Discrete Fracture Network Engineering Conference, June 20-22 Seattle, USA., 2018.
@conference{Medinac2018,
title = {Pre- and Post-Blast Rock Block Size Analysis Using UAV-based Data and Discrete Fracture Network},
author = {Filip Medinac and Thomas Bamford and Kamran Esmaieli and Angela Schoellig},
year = {2018},
date = {2018-06-20},
urldate = {2018-06-20},
address = {Seattle, USA.},
organization = {2nd International Discrete Fracture Network Engineering Conference, June 20-22},
abstract = {Drilling and blasting is one of the key processes in open pit mining, required to reduce in-situ rock block size to rock
fragments that can be handled by mine equipment. It is a significant cost driver of any mining operation which can influence the
downstream mining processes. In-situ rock block size influences the muck pile size distribution after blast, and the amount of drilling
and explosive required to achieve a desired distribution. Thus, continuous measurement of pre- and post-blast rock block size
distribution is essential for the optimization of the rock fragmentation process.
This paper presents the results of a case study in an open pit mine where an Unmanned Aerial Vehicle (UAV) was used for mapping
of the pit walls before blast. Pit wall mapping and aerial data was used as input to generate a 3D Discrete Fracture Network (DFN)
model of the rock mass and to estimate the in-situ block size distribution. Data collected by the UAV was also used to estimate the
post-blast rock fragment size distribution. The knowledge of in-situ and blasted rock size distributions can be related to assess blast
performance. This knowledge will provide feedback to production engineers to adjust the blast design.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Drilling and blasting is one of the key processes in open pit mining, required to reduce in-situ rock block size to rock
fragments that can be handled by mine equipment. It is a significant cost driver of any mining operation which can influence the
downstream mining processes. In-situ rock block size influences the muck pile size distribution after blast, and the amount of drilling
and explosive required to achieve a desired distribution. Thus, continuous measurement of pre- and post-blast rock block size
distribution is essential for the optimization of the rock fragmentation process.
This paper presents the results of a case study in an open pit mine where an Unmanned Aerial Vehicle (UAV) was used for mapping
of the pit walls before blast. Pit wall mapping and aerial data was used as input to generate a 3D Discrete Fracture Network (DFN)
model of the rock mass and to estimate the in-situ block size distribution. Data collected by the UAV was also used to estimate the
post-blast rock fragment size distribution. The knowledge of in-situ and blasted rock size distributions can be related to assess blast
performance. This knowledge will provide feedback to production engineers to adjust the blast design.
fragments that can be handled by mine equipment. It is a significant cost driver of any mining operation which can influence the
downstream mining processes. In-situ rock block size influences the muck pile size distribution after blast, and the amount of drilling
and explosive required to achieve a desired distribution. Thus, continuous measurement of pre- and post-blast rock block size
distribution is essential for the optimization of the rock fragmentation process.
This paper presents the results of a case study in an open pit mine where an Unmanned Aerial Vehicle (UAV) was used for mapping
of the pit walls before blast. Pit wall mapping and aerial data was used as input to generate a 3D Discrete Fracture Network (DFN)
model of the rock mass and to estimate the in-situ block size distribution. Data collected by the UAV was also used to estimate the
post-blast rock fragment size distribution. The knowledge of in-situ and blasted rock size distributions can be related to assess blast
performance. This knowledge will provide feedback to production engineers to adjust the blast design.