S. Alamdari; M.H. Basiri; A. Mousavi; A. Soofastaei
Abstract
The haul trucks consume a significant energy source in open-pit mines, where diesel fuel is widely used as the main energy source. Improving the haul truck fuel consumption can considerably decrease the operating cost of mining, and more importantly, reduce the pollutants and greenhouse gas emissions. ...
Read More
The haul trucks consume a significant energy source in open-pit mines, where diesel fuel is widely used as the main energy source. Improving the haul truck fuel consumption can considerably decrease the operating cost of mining, and more importantly, reduce the pollutants and greenhouse gas emissions. This work aims to model and evaluate the diesel fuel consumption of the mining haul trucks. The machine learning techniques including multiple linear regression, random forest, artificial neural network, support vector machine, and kernel nearest neighbor are implemented and investigated in order to predict the haul truck fuel consumption based on the independent variables such as the payload, total resistance, and actual speed. The prediction models are built on the actual dataset collected from an Iron ore open-pit mine located in the Yazd province, Iran. In order to evaluate the goodness of the predicted models, the coefficient of determination, mean square error, and mean absolute error are investigated. The results obtained demonstrate that the artificial neural network has the highest accuracy compared to the other models (coefficient of determination = 0.903, mean square error = 489.173, and mean absolute error = 13.440). In contrast, the multiple linear regression exhibits the worst result in all statistical metrics. Finally, a sensitivity analysis is used to evaluate the significance of the independent variables.
M. Kamran
Abstract
Drillability is one of the significant issues in rock engineering. The drilling rate index (DRI) is an important tool in analyzing the drillability of rocks. Several efforts have been made by the researchers to correlate and evaluate DRI of rocks. The ensemble learning methods including the decision ...
Read More
Drillability is one of the significant issues in rock engineering. The drilling rate index (DRI) is an important tool in analyzing the drillability of rocks. Several efforts have been made by the researchers to correlate and evaluate DRI of rocks. The ensemble learning methods including the decision tree (DT), adaptive boosting (AdaBoost), and random forest (RF) are employed in this research work in order to predict DRI of rocks. A drillability database with four parameters is compiled in this work. A relationship between the input parameters and DRI is established using the simple regression analysis. In order to train the model, different mechanical properties of rocks incorporating the uniaxial compressive strength (UCS), Brazilian tensile strength (BTS), brittleness test (S20), and sievers’ J-miniature drill value (Sj) are taken as the input variables. The original DRI database is randomly divided into the training and test sets with an 80/20 sampling method. Various algorithms are developed, and consequently, several approaches are followed in order to predict DRI of the rock samples. The model performance has revealed that RF predicts DRI with a high accuracy rate. Besides, the Monte Carlo simulations exhibit that this approach is more reliable in predicting the probability distribution of DRI. Therefore, the proposed model can be practiced for the stability risk management and the investigative design of DRI.
Rock Mechanics
Sh. Bacha; Z. Mu Long; A. Javed; Sh. Al Faisal
Abstract
Rock burst is the most attractive and hot research area in geomechanics, mining, and civil engineering due to the increasing depth of mines and construction of deep underground structures. It has also been a severe problem in ground control measures in the last few decades. Many studies have been done ...
Read More
Rock burst is the most attractive and hot research area in geomechanics, mining, and civil engineering due to the increasing depth of mines and construction of deep underground structures. It has also been a severe problem in ground control measures in the last few decades. Many studies have been done by different researchers in order to minimize the hazards of rock burst and to provide a safe mining/working environment. It is important to review the current advancement of rock burst prediction and its preventive measures. This paper reviews the experimental progress of rock burst warning, prediction, control measures, and potential damage measures. Different effective methods of rock burst prediction and control are also described.
Rock Mechanics
H. Fattahi; N. Zandy Ilghani
Abstract
Horizontal directional drilling is usually used in drilling engineering. In a variety of conditions, it is necessary to predict the torque required for performing the drilling operation. Nevertheless, there is presently not a convenient method available to accomplish this task. In order to overcome this ...
Read More
Horizontal directional drilling is usually used in drilling engineering. In a variety of conditions, it is necessary to predict the torque required for performing the drilling operation. Nevertheless, there is presently not a convenient method available to accomplish this task. In order to overcome this difficulty, the current work aims at predicting the required rotational torque (RT) to operate horizontal directional drilling on the 7 effective parameters including the length of drill string in the borehole (L), axial force on the cutter/bit (P), total angular change of the borehole (KL), radius for the ith reaming operation (Di), rotational speed (rotation per minute) of the bit (N), mud flow rate (W), and mud viscosity (V). In this paper, we propose an approach based on the model selection criteria such as various statistical performance indices mean squared error (MSE), variance account for (VAF), root mean squared error (RMSE), squared correlation coefficient (R2), and mean absolute percentage error (MAPE) to select the most appropriate model among a set of 20 candidate ones to estimate RT, given a set of observed data. Once the most appropriate model is selected, a Bayesian framework is employed to develop the predictive distributions of RT, and to update them with new project-specific data that significantly reduce the associated predictive uncertainty. Overall, the results obtained indicate that the proposed RT model possesses a satisfactory predictive performance.
M. Mohseni; M. Ataei
Abstract
In this work, the time series modeling was used to predict the Tazareh coal mine risks. For this purpose, initially, a monthly analysis of the risk constituents including frequency index and incidence severity index was performed. Next, a monthly time series diagram related to each one of these indices ...
Read More
In this work, the time series modeling was used to predict the Tazareh coal mine risks. For this purpose, initially, a monthly analysis of the risk constituents including frequency index and incidence severity index was performed. Next, a monthly time series diagram related to each one of these indices was for a nine year period of time from 2005 to 2013. After extrusion of the trend, seasonality, and remainder constituents of the time series modeling, the final time series model of the indices was determined with high precision. The precision level of the resulting model was evaluated using the root mean square error (RMSE) method. The values obtained for the severity index and accident frequency index were 0.001 and 6.400, respectively. Evaluation of the seasonal time series constituent of the frequency index showed that, yearly, most number of accidents occurred in April, and the least one took place in January. Additionally, evaluation of the seasonal time series constituent of the severity index showed that, every year, the severest accidents occurred in October, and the least ones happened in January. Using the final model, a monthly prediction of indices was performed for a four year period of time from 2014 to 2017. Subsequently, using the known mean work hours in the mine, predictions of the number of accidents and the number of work days lost within a similar time period were made. The prediction results showed that in the future, the number of accidents and the number of work days lost would have a down-going trend such that for similar months, annually, an average 22% decrease in the number of accidents and an average 24% decrease in the number of work days lost are expected.
F. Khorram; H. Memarian; B. Tokhmechi; H. Soltanian-zadeh
Abstract
In this study based on image analysis, an ore grade estimation model was developed. The study was performed at a limestone mine in central Iran. The samples were collected from different parts of the mine and crushed in size from 2.58 cm down to 15 cm. The images of the samples were taken in appropriate ...
Read More
In this study based on image analysis, an ore grade estimation model was developed. The study was performed at a limestone mine in central Iran. The samples were collected from different parts of the mine and crushed in size from 2.58 cm down to 15 cm. The images of the samples were taken in appropriate environment and processed. A total of 76 features were extracted from the identified rock samples in all images. Neural network used as an intelligent tool for ore grade estimation and the features of every image were combined with weighted average method. In order to feature dimensional decrease, principal component analysis method was used. Six principal components, which were extracted from the feature vectors, captured 90.661% of the total feature variance. Components were used as the input to neural network and four grade attributes of limestone (CaCO3, Al2O3, Fe2O3 and MgCO3) were used as the output. The root of mean squared error between the observed values and the model estimated values for the test data set are 6.378, 4.847, 0.1513 and 0.0284, the R2 values are 0.7852, 0.8663, 0.7591and 0.8094 for the mentioned chemical composition respectively. The magnitude of R2 indicates the correlation between actual and estimated data. Therefore, it can be inferred that the model can successfully estimate the limestone chemical compositions percentage.
