Search

After each blasting operation in surface mines, undesirable environmental impacts, such as air-blast (AB) and ground vibration, are inevitable. Therefore, minimizing and controlling these impacts are crucial in order to reduce environmental problems. This study presents new and practical advanced machine learning methods for AB prediction using 62 datasets gathered from four quarry sites in Malaysia. The developed models were constructed based on the boosted regression tree (BRT) and least-squares support vector machine (LSSVM), improved with three metaheuristic algorithms: the gray wolf optimizer (GWO), genetic algorithm (GA), and artificial bee colony (ABC). Six hybrid models, namely BRT–GA, BRT–ABC, BRT–GWO, LSSVM–GA, LSSVM–ABC, and LSSVM–GWO models, were developed and their performances were evaluated using metrics such as R-squared correlation and other methods like the Taylor diagram and quantile–quantile plots. To provide a better assessment of the models' performances, the dataset were categorized into training and testing parts. The results demonstrated that, among the six hybrid models, the LSSVM–GWO model provided the highest efficiency in the testing part while the BRT–GWO model had the best accuracy but the performance of BRT–GWO was the best in the training part. In other words, the BRT hybrid models had the best performance in training, and LSSVM hybrid models in the testing part. The results indicate the effectiveness of combining GWO with LSSVM and BRT models to predict AB. [ABSTRACT FROM AUTHOR]

Rainfall is crucial for the development and management of water resources. Six hybrid soft computing models, including multilayer perceptron (MLP)–Henry gas solubility optimization (HGSO), MLP–bat algorithm (MLP–BA), MLP–particle swarm optimization (MLP–PSO), radial basis neural network function (RBFNN)–HGSO, RBFNN–PSO, and RBFGNN–BA, were used in this study to forecast monthly rainfall at two stations in Malaysia (Sara and Banding). Different statistical measures (mean absolute error (MAE) and Nash–Sutcliffe efficiency (NSE) and percentage of BIAS (PBIAS)) and a Taylor diagram were used to assess the models' performance. The results indicated that the MLP–HGSO performed better than the other models in forecasting rainfall at both stations. In addition, transition matrices were computed for each station and year based on the conditional probability of rainfall or absence of rainfall on a given month. The values of MAE for testing processes for the MLP–HGSO, MLP–PSO, MLP–BA, RBFNN–HGSO, RBFNN–BA, and RBFNN–PSO at the first station were 0.712, 0.755, 0.765, 0.717, 0.865, and 0.891, while the corresponding NSE and PBIAS values were 0.90–0.23, 0.83–0.29, 0.85–0.25, 0.87–0.27, 0.81–0.31, and 0.80–0.35, respectively. For the second station, the values of MAE were found 0.711, 0.743, 0.742, 0.719, 0.863 and 0.890 for the MLP–HGSO, MLP–PSO, MLP–BA, RBFNN–HGSO, RBFNN–BA, and RBFNN–PSO during testing processes and the corresponding NSE–PBIAS values were 0.92–0.22, 0.85–0.28, 0.89–0.26, 0.91–0.25, 0.83–0.31, 0.82–0.32, respectively. Based on the outputs of the MLP–HGSO, the highest rainfall was recorded in 2012 with a probability of 0.72, while the lowest rainfall was recorded in 2006 with a probability of 0.52 at the Sara Station. In addition, the results indicated that the MLP–HGSO performed better than the other models within the Banding Station. According to the findings, the hybrid MLP–HGSO was selected as an effective rainfall prediction model. [ABSTRACT FROM AUTHOR]

Climate change has increased drought frequency globally, which harms the environment, agriculture, and water resources. This study explores the capacity of a hybrid model based on the integration of extreme learning machine (ELM) with a novel meta-heuristic algorithm called Beluga Whale Optimization (BWO) for drought forecasting within tropical region. The forecasting adopted for the standardized streamflow index (SSI) for several time scales (SSI-1, SSI-3, SS1-6, SSI-9, SSI-12, and SSI-24) over Selangor state, Malaysia. The drought calculation was based on 58 years of stream flow data (1961 to 2018) obtained from two hydrological stations, S3615412 and S3414421. The ELM-BWO model was validated with standalone models including Gradient boosting regression (GBR) and classical ELM in addition to the hybridization of ELM with genetic algorithm (GA). The inputs were nominated by using partial autocorrelation function (PACF) at 5% significance level. The research finding reveals that the ELM-BWO model used for the first in hydrological drought forecasting, outperforms other models and provides the narrowest uncertainty bounds. The proposed ELM-BWO model accuracy was demonstrated by the decreased metric of root mean square error (RMSE) values across different drought timescales (e.g., SSI-9 by 12.65% to 78.74%). In addition, the model obtains the highest correlation coefficient (0.9777 and 0.9944) and Wilmot index values (0.9882 and 0.9971) for both stations. The ELM-BWO model can be used to form a dependable expert intelligent system for anticipating hydrological drought at multiple time scales, making decisions about appropriate measures to deal with hydrological drought at the studied stations, and supporting sustainable water resource management. [ABSTRACT FROM AUTHOR]

The efficiency of tunnel boring machines (TBMs) in underground projects has great significance for the mining and tunneling industries, demanding a reliable estimation of the TBM's performance in different geotechnical conditions. The current research work attempted to suggest an optimal predictor model of TBM performance as a reliable alternative to experimental and numerical techniques. To achieve this target, three data-mining techniques, namely neural network (NN), gene expression programming (GEP), and multivariate adaptive regression splines (MARS), were employed for modelling the TBM performance, and then the most robust predictive model was optimized via a metaheuristic search method known as whale optimization algorithm (WOA). For the modelling purpose, an experimental database was compiled by performing a field assessment program in a tunneling project in Malaysia and then conducting laboratory testing on the derived rock specimens. Based on the measured experimental data, the six most influential parameters were identified and served as model inputs to predict penetration rate (PR). In order to indicate the capability of the developed GEP and NN models, a stepwise linear regression model, i.e., MARS, was designed for PR prediction as well. The predictive capacity of the constructed models was quantified using a series of statistical indices, i.e., root mean squared error (RMSE), determination coefficient (R2) and variance account for (VAF). Based on the computed indices for testing records, both the proposed GEP and NN models (with RMSE values of 0.1882 and 0.2120 and R2 values of 0.9058 and 0.8735, respectively) yielded more accurate predictive results than the MARS model with RMSE of 0.2553 and R2 of 0.8346. Hence, by achieving the most robust performance compared to the rest, GEP-based model can provide a new practical equation with a high level of accuracy. In other part of this study, the six input parameters of the GEP model and its equation were, respectively, defined as decision variables and objective function for the WOA technique to find the optimum values of PR. As a consequence of optimizing the GEP equation, the maximum value of PR rose from 3.75 m/h to 4.022 m/h, equivalent to an increase of 7.25% in PR value. The findings of this study verified the applicability of the proposed hybrid GEP and WOA approach in the site investigation phase of tunneling projects constructed by TBMs. [ABSTRACT FROM AUTHOR]

Prediction of ground vibration induced by blasting operations is a crucial challenge to engineers working in surface mines. This study aims to assess the efficiency of two advanced machine learning models in predicting ground vibrations in a granite quarry located in Malaysia. To this end, two intelligent models were proposed by hybridizing the relevance vector regression (RVR) with the grey wolf optimization (GWO) (which formed the RVR-GWO model) and with the bat-inspired algorithm (BA) (which formed the RVR-BA model). To the best of our knowledge, this is the first attempt to predict ground vibration using the RVR-GWO and RVR-BA models. The afore-mentioned models were developed and tested using 95 datasets. Then, the performance of the developed models was statistically checked through four comparative experiments using, among others, mean square error (MSE) and correlation coefficient (R). The results indicated the superiority of the RVR-GWO model over the RVR-BA model in terms of prediction precision. The RVR-GWO model with R of 0.915 and MSE = 7.920 predicted the ground vibration better than the RVR-BA model with R of 0.867 and MSE = 8.551. Accordingly, it was concluded that applying the GWO algorithm to RVR can result in high accuracy in the prediction of blast-induced ground vibration. [ABSTRACT FROM AUTHOR]

For binary cycle power plants, the energy of solar may be utilized as a heating source. Malaysia has a tropical climate, which receives 4500 kWh/m2 of solar irradiation on a daily basis and roughly 12 h of sunshine every day. Thus, the potential of binary power cycle plants in Malaysia by utilizing the energy of solar as a source of heat is very promising. However, to date, no feasibility study on binary power cycle generation by using the energy of solar as a source of heat in Malaysia has been conducted. Therefore, in this work, a techno-economic analysis on the proposed binary power cycle generation in Malaysia is conducted using commercially available software, HOMER and MATLAB Simulink. The analysis is conducted to assess the proposed binary power generation implementation in Malaysia. Particle Swarm Optimization (PSO), Artificial Bee Colony (ABC) and Teaching Learning Based Optimization (TLBO) are three separate metaheuristic algorithms that are exploited so as to determine the optimum cost of economic of the project. From the results obtained in this work, TLBO gives the most optimized economic cost compared to PSO and ABC. Also, it is found that binary power cycle generation by using energy of solar as a source of heat is feasible in Malaysia. [ABSTRACT FROM AUTHOR]

The advance rate (AR) of a tunnel boring machine (TBM) in hard rock condition is a key parameter for the successful accomplishment of a tunneling project, and the proper and reliable prediction of this parameter can lead to minimizing the risks associated to high capital costs and scheduling for such projects. This research aims at optimizing the hyper-parameters of the support vector machine (SVM) technique through the use of three optimization algorithms, namely, gray wolf optimization (GWO), whale optimization algorithm (WOA) and moth flame optimization (MFO), in forecasting TBM AR. In fact, the role of these optimization techniques is to optimize the hyperparameters 'C' and 'gamma' of the SVM model to get higher performance prediction. To develop the hybrid SVM-based models, 1,286 sample sets of data collected from a water transfer tunnel in Malaysia comprising seven input variables, i.e., rock mass rating, uniaxial compressive strength, Brazilian tensile strength, rock quality designation, weathering zone, thrust force and revolution per minute, and one output variable, i.e., TBM AR, were considered and used. Several GWO-SVM, WOA-SVM and MFO-SVM models were constructed to predict TBM AR considering their effective parameters. The accuracy levels of the proposed models were assessed using four statistical indices, i.e., the coefficient of determination (R2), root mean squared error (RMSE), mean absolute error (MAE), and variance accounted for (VAF). Modeling results revealed that the MFO algorithm can capture better hyper-parameters of the SVM model in predicting TBM AR among all three hybrid models. R2 of (0.9623 and 0.9724), RMSE of (0.1269 and 0.1155), and VAF of (96.24 and 97.34%), respectively, for training and test stages of the MFO-SVM model confirmed that this hybrid SVM model is a powerful and applicable technique addressing problems related to TBM performance with a high level of accuracy. [ABSTRACT FROM AUTHOR]