Exploration
Ashraf Ismael; Abdelrahem Khalefa Embaby; Faissal Ali; Hussin Farag; Sayed Gomaa; Mohamed Elwageeh; Bahaa Mousa
Abstract
The mineral resource estimation process necessitates a precise prediction of the grade based on limited drilling data. Grade is crucial factor in the selection of various mining projects for investment and development. When stationary requirements are not met, geo-statistical approaches for reserve estimation ...
Read More
The mineral resource estimation process necessitates a precise prediction of the grade based on limited drilling data. Grade is crucial factor in the selection of various mining projects for investment and development. When stationary requirements are not met, geo-statistical approaches for reserve estimation are challenging to apply. Artificial Neural Networks (ANNs) are a better alternative to geo-statistical techniques since they take less processing time to create and apply. For forecasting the iron ore grade at El-Gezera region in El- Baharya Oasis, Western Desert of Egypt, a novel Artificial Neural Network (ANN) model, geo-statistical methods (Variograms and Ordinary kriging), and Triangulation Irregular Network (TIN) were employed in this study. The geo-statistical models and TIN technique revealed a distinct distribution of iron ore elements in the studied area. Initially, the tan sigmoid and logistic sigmoid functions at various numbers of neurons were compared to choose the best ANN model of one and two hidden layers using the Levenberg-Marquardt pure-linear output function. The presented ANN model estimates the iron ore as a function of the grades of Cl%, SiO2%, and MnO% with a correlation factor of 0.94. The proposed ANN model can be applied to any other dataset within the range with acceptable accuracy.
Exploration
Bashir Shokouh Saljoughi; Ardeshir Hezarkhani
Abstract
The porphyry Cu-mineralization potential area studied in this article is located in the southern section of the Central Iranian volcano–sedimentary complex, contains large number of mineral deposits, and occurrences that are currently facing a shortage of resources. Therefore, prospecting potential ...
Read More
The porphyry Cu-mineralization potential area studied in this article is located in the southern section of the Central Iranian volcano–sedimentary complex, contains large number of mineral deposits, and occurrences that are currently facing a shortage of resources. Therefore, prospecting potential areas in the deeper and peripheral spaces has become a high priority in this region. Different direct and indirect methods try to predict promising areas for future explorations that most of them are very time-consuming and costly. The main goal of mineral prospecting is applying a transparent and robust approach for identifying high potential areas to be explored further in the future. This study presents the procedure taken to create two different Cu-mineralization prospectivity maps. This study aims to investigate the results of applying the ANN technique, and to compare them with the outputs of applying GEP method. The geo-datasets employed for creating evidential maps of porphyry Cu mineralization include solid geology map, alteration map, faults, dykes, airborne total magnetic intensity, airborne gamma-ray spectrometry data (U, Th, K and total count), and known Cu occurrences. Based on this study, the ANN technique (10 neurons in the hidden layer and LM learning algorithm) is a better predictor of Cu mineralization compared to the GEP method. The results obtained from the P-A plot showed that the ANN model indicates that 80% (vs. 70% for GEP) of the identified copper occurrences are projected to be present in only 20% (vs. 30% for GEP) of the surveyed area. The ANN technique due to capabilities such as classification, pattern matching, optimization, and prediction is useful in identifying anomalies associated with the Cu mineralization.
Rock Mechanics
Alireza Afradi; Arash Ebrahimabadi; Mansour Hedayatzadeh
Abstract
Tunnel Boring Machines (TBMs) are extensively used to excavate underground spaces in civil and tunneling projects. An accurate evaluation of their penetration rate is the key factor for the TBM performance prediction. In this study, artificial intelligence methods are used to predict the TBM penetration ...
Read More
Tunnel Boring Machines (TBMs) are extensively used to excavate underground spaces in civil and tunneling projects. An accurate evaluation of their penetration rate is the key factor for the TBM performance prediction. In this study, artificial intelligence methods are used to predict the TBM penetration rate in excavation operations in the Kerman tunnel and the Gavoshan water conveyance tunnels. The aim of this paper is to show the application of the Multivariate Linear Regression (MLR), Artificial Neural Network (ANN), and Support Vector Machine (SVM) for the TBM penetration rate prediction. The penetration rate parameter is considered as a dependent variable, and the Rock Quality Designation (RQD), Brazilian Tensile Strength (BTS), Uniaxial Compressive Strength (UCS), Density (D), Joint Angle (JA), Joint Spacing (JS), and Poisson's Ratio are considered as independent variables. The obtained results by the several proposed methods indicated a high accuracy between the predicted and measured penetration rates, but the support vector machine yields more precise and realistic outcomes.
Leila Nikakhtar; Shokroallah Zare; Hossein Mirzaei
Abstract
Surface settlement induced by tunneling is one of the most crucial problems in urban environments. Hence, accurate prediction of soil geotechnical properties is an important prerequisite in the minimization of it. In this research work, the amount of surface settlement is predicted using three-dimensional ...
Read More
Surface settlement induced by tunneling is one of the most crucial problems in urban environments. Hence, accurate prediction of soil geotechnical properties is an important prerequisite in the minimization of it. In this research work, the amount of surface settlement is predicted using three-dimensional numerical simulation in the finite difference method and Artificial Neural Network (ANN). In order to determine the real geotechnical properties of soil layers around the tunnel; back-analysis is carried out using the optimization algorithm and monitoring data. Among the different optimization methods, genetic algorithm (GA) and particle swarm optimization (PSO) are selected, and their performance is compared. The results obtained show that the artificial neural network has a high ability with the amounts of R=0.99, RMSE=0.0117, and MSE= 0.000138 in predicting the surface settlement obtained from 150 simulations from randomly generated data. Comparing the results of back-analysis using the optimization algorithm, the genetic algorithm shows less error than the particle swarm algorithm in different initial populations. In all cases of analysis, the calculation time for both algorithms lasts about 5 minutes, which indicates the applicability of both algorithms in optimizing the parameters in mechanized tunneling in a short time.
Blessing Olamide Taiwo
Abstract
Assessment of blast results is a significant approach for the improvement of mining operations. The different procedures for investigating rock fragmentation have their limitations, causing different variation prediction errors. Thus every technique is site-explicit, and applicable for a few explicit ...
Read More
Assessment of blast results is a significant approach for the improvement of mining operations. The different procedures for investigating rock fragmentation have their limitations, causing different variation prediction errors. Thus every technique is site-explicit, and applicable for a few explicit purposes. This work evaluates the existing empirical blast fragmentation model predictions in the case study of small-scale dolomite quarries. An attempt is made to compare the prediction accuracy of the modified Kuz-Ram model, Lawal 2021 model, and Kuznetsov-Cunningham-Ouchterlony (KCO) model with the WipFrag© analysis result and proposed artificial neural network (ANN) models. The prediction error analysis of the current models and that of the new proposed ANN models is evaluated using the three model assessment indices. The assessment indices uncover that the KCO model when compared to the modified Kuz-Ram model has the least error for most blast round percentage passing size predicted. However, the proposed artificial neural network models show high prediction exactness in predicting blast fragment mean size than the existing empirical models. Therefore, the proposed ANN models can be used to improve the productivity of small-scale dolomite blasting operation results for practical purposes.
J. Shakeri; H. Amini Khoshalan; H. Dehghani; M. Bascompta; K. Onyelowe
Abstract
In this research work, a comprehensive study is conducted to predict flyrock as a typical and undesirable phenomenon occurring during the blasting operation in open-pit mining. Despite the availability of several empirical methods for predicting the flyrock distance, the complexity of flyrock analysis ...
Read More
In this research work, a comprehensive study is conducted to predict flyrock as a typical and undesirable phenomenon occurring during the blasting operation in open-pit mining. Despite the availability of several empirical methods for predicting the flyrock distance, the complexity of flyrock analysis has resulted in the low performance of these models. Therefore, the statistical and robust artificial intelligence techniques are applied for flyrock prediction in the Sungun copper mine in Iran. For this purpose, the linear multivariate regression (LMR), imperialist competitive algorithm (ICA), adaptive neuro-fuzzy inference system (ANFIS), and artificial neural network (ANN) methods are applied to predict flyrock with effective parameters including the blasthole diameter, stemming, burden, powder factor, and maximum charge per delay. According to the attained results, the ANN model with the structure of 5-8-1, Levenberg-Marquardt as the learning algorithm, and log-sigmoid (logsig) as the transfer functions are selected as the optimal network with the RMSE and R2 values of 5.04 m and 95.6% to predict flyrock, respectively. Also it can be concluded that the ICA technique has a relatively high capability in predicting flyrock, with the LMR and ANFIS models placed in the next. Finally, the sensitivity analysis reveal that the powder factor and blasthole diameters have the most importance on the flyrock distance in the present work.
Mine Economic and Management
R. Bastami; A. Aghajani Bazzazi; H. Hamidian Shoormasti; K. Ahangari
Abstract
The use of blasting cost (BC) prediction to achieve optimal fragmentation is necessary in order to control the adverse consequences of blasting such as fly rock, ground vibration, and air blast in open-pit mines. In this research work, BC is predicted through collecting 146 blasting data from six limestone ...
Read More
The use of blasting cost (BC) prediction to achieve optimal fragmentation is necessary in order to control the adverse consequences of blasting such as fly rock, ground vibration, and air blast in open-pit mines. In this research work, BC is predicted through collecting 146 blasting data from six limestone mines in Iran using the artificial neural networks (ANNs), gene expression programming (GEP), linear multivariate regression (LMR), and non-linear multivariate regression (NLMR) models. In all models, the ANFO value, number of detonators, Emolite value, hole number, hole length, hole diameter, burden, spacing, stemming, sub-drilling, specific gravity of rock, hardness, and uniaxial compressive strength are used as the input parameters. The ANN model results in the test stage indicating a higher correlation coefficient (0.954) and a lower root mean square error (973) compared to the other models. In addition, it has a better conformity with the real blasting costs in comparison with the other models. Although the ANNs method is regarded as one of the intelligent and powerful techniques in parameter prediction, its most important fault is its inability to provide mathematical equations for engineering operations. In contrast, the GEP model exhibits a reliable output by presenting a mathematical equation for BC prediction with a correlation coefficient of 0.933 and a root mean square error of 1088. Based on the sensitivity analysis, the spacing and ANFO values have the maximum and minimum effects on the BC function, respectively. The number of detonators, Emolite value, hole number, specific gravity, hardness, and rock uniaxial compressive strength have a positive correlation with BC, while the ANFO value, hole length, hole diameter, burden, spacing, stemming, and sub-drilling have a negative correlation with BC.
B. Shokouh Saljoughi; A. Hezarkhani; E. Farahbakhsh
Abstract
The study area, located in the southern section of the Central Iranian volcano–sedimentary complex, contains a large number of mineral deposits and occurrences which is currently facing a shortage of resources. Therefore, the prospecting potential areas in the deeper and peripheral spaces has become ...
Read More
The study area, located in the southern section of the Central Iranian volcano–sedimentary complex, contains a large number of mineral deposits and occurrences which is currently facing a shortage of resources. Therefore, the prospecting potential areas in the deeper and peripheral spaces has become a high priority in this region. Different direct and indirect methods try to predict promising areas for future explorations, most of which are very time-consuming and costly. The main goal of mineral prospecting is applying a transparent and robust approach for identifying high potential areas to be explored further in the future. This work presents the procedure taken to create two different Cu-mineralization prospectivity maps. The first map is created using a knowledge-driven fuzzy technique and the second one by a data-driven Artificial Neural Network (ANN) approach. In this study aim is to investigate the results of applying the ANN technique and to compare them with the outputs of applying the fuzzy logic method. The geo-datasets employed for creating evidential maps of porphyry Cu mineralization include the solid geology map, alteration map, faults, dykes, airborne total magnetic intensity, airborne gamma-ray spectrometry data (U, Th, K and total count), and known Cu occurrences. Based on this study, the ANN technique is a better predictor of Cu mineralization compared to the fuzzy logic method. The ANN technique, due to capabilities such as classification, pattern matching, optimization, and prediction, is useful in identifying the anomalies associated with the Cu mineralization.
F. Razavi Rad; F. Mohammad Torab; A. Abdollahzadeh
Abstract
Considering the importance of Cd and U as pollutants of the environment, this study aims to predict the concentrations of these elements in a stream sediment from the Eshtehard region in Iran by means of a developed artificial neural network (ANN) model. The forward selection (FS) method is used to select ...
Read More
Considering the importance of Cd and U as pollutants of the environment, this study aims to predict the concentrations of these elements in a stream sediment from the Eshtehard region in Iran by means of a developed artificial neural network (ANN) model. The forward selection (FS) method is used to select the input variables and develop hybrid models by ANN. From 45 input candidates, 13 and 14 variables are selected using the FS method for Cadmium and Uranium, respectively. Considering the correlation coefficient (R2) values, both the ANN and FS-ANN models are acceptable for estimation of the Cd and U concentrations. However, the FS-ANN model is superior because the R2 values for estimation of Cd and U by the FS-AAN model is higher than those for estimation of these elements by the ANN model. It is also shown that the FS-ANN model is preferred in estimating the Cd and U population due to reduction in the calculation time as a consequence of having less input variables.
S. Bahrami; F. Doulati Ardejani
Abstract
In this study, a hybrid intelligent model has been designed to predict groundwater inflow to a mine pit during its advance. Novel hybrid method coupling artificial neural network (ANN) with genetic algorithm (GA) called ANN-GA, was utilised. Ratios of pit depth to aquifer thickness, pit bottom radius ...
Read More
In this study, a hybrid intelligent model has been designed to predict groundwater inflow to a mine pit during its advance. Novel hybrid method coupling artificial neural network (ANN) with genetic algorithm (GA) called ANN-GA, was utilised. Ratios of pit depth to aquifer thickness, pit bottom radius to its top radius, inverse of pit advance time and the hydraulic head (HH) in the observation wells to the distance of observation wells from the centre of pit were used as inputs to the network. An ANN-GA with 4-5-3-1 arrangement was found capable to predict the groundwater inflow to mine pit. The accuracy and reliability of model was verified by field data. Predicted results were very close to the field data. The correlation coefficient (R) value was 0.998 for training set, and in testing stage it was 0.99.
Saeed Alishahi; Ahmad Darban; Mahmood Abdollahi
Abstract
Since a high toxicity of cyanide which use as a reagent in the gold processing plant, thiosulfate has been recognized as a environmental friendly reagent for leaching of gold from ore. After gold leaching process it's important for recovery of gold from solution using adsorption or extraction methods, ...
Read More
Since a high toxicity of cyanide which use as a reagent in the gold processing plant, thiosulfate has been recognized as a environmental friendly reagent for leaching of gold from ore. After gold leaching process it's important for recovery of gold from solution using adsorption or extraction methods, One of these methods is activated carbon.The loading of gold from industrial thiosulfate solution that obtained from Zarshuran gold plant-Takab-Iran, onto activated carbon have been investigated. The affecting variables of the adsorption of gold on the carbon included, temperature, concentration of gold, size of activated carbon, pH and the ratio of amount of activated carbon to the volume of solution. The results have shown that at low concentration of gold, effective loading can be achieved at pH 10.5. The size of activated carbon has a significant effect on the loading of gold. In this research the recovery of gold on activated carbon has been predicted using artificial neural network. For this purpose temperature, pH, the proportion of solution volume to weight of activated carbon, gold concentration and time of adsorption were taken as input parameters, whereas, the recovery of gold on activated carbon from thiosulfate solution was considered as an output parameter. The network with LMBP algorithm with two hidden layer were used and the topology 5-4-13-1 showed the best ability for prediction.Moreover sensitive analyze were indicated parameters pH and temperature have substantial influence on adsorption.
