Development of hybrid intelligent models for predicting TBM penetration rate in hard rock condition

One of the main concerns in geotechnical engineering is slope stability prediction during the earthquake. In this study, two intelligent systems namely artificial neural network (ANN) and particle swarm optimization (PSO)---ANN models were developed to predict factor of safety (FOS) of homogeneous slopes. Geostudio program based on limit equilibrium method was utilized to obtain 699 FOS values with different conditions. The most influential factors on FOS such as slope height, gradient, cohesion, friction angle and peak ground acceleration were considered as model inputs in the present study. A series of sensitivity analyses were performed in modeling procedures of both intelligent systems. All 699 datasets were randomly selected to 5 different datasets based on training and testing. Considering some model performance indices, i.e., root mean square error, coefficient of determination (R2) and value account for (VAF) and using simple ranking method, the best ANN and PSO---ANN models were selected. It was found that the PSO---ANN technique can predict FOS with higher performance capacities compared to ANN. R2 values of testing datasets equal to 0.915 and 0.986 for ANN and PSO---ANN techniques, respectively, suggest the superiority of the PSO---ANN technique.

Prediction of seismic slope stability through combination of particle swarm optimization and neural network

In the present study, we applied artificial neural network (ANN) optimized with particle swarm optimization (PSO) for the problem of landslide susceptibility mapping (LSM) prediction. Many studies have revealed that the ANN-based techniques are reliable methods for estimating the LSM. However, most ANN training models facing with major problems such as slow degree of learning system as well as being trapped in their local minima. Optimization algorithms (OA) such as PSO can improve performance results of ANN. Existing applications of PSO model to ANN training have not been used in area of landslide mapping, neither assess the optimal architecture of networks nor the influential factors affecting this problem. Hence, the present study focused on the application of a hybrid PSO-based ANN model (PSO-ANN) to the prediction of landslide susceptibility hazardous mapping. To prepare training and testing datasets for the ANN and PSO-ANN network models, large data collection (i.e., a database consists 168970 training datasets and 42243 testing datasets) were provided from an area of Layleh valley, located in Kermanshah, west of Iran. All the variables of PSO algorithm (e.g., in addition to the network parameter and network weights) were optimized to achieve the most reliable maps of landslide susceptibility. The input dataset includes elevation, slope aspect, slope degree, curvature, soil type, lithology, distance to road, distance to river, distance to fault, land use, stream power index (SPI) and topographic wetness index (TWI), where the output was taken landslide susceptibility value. The predicted results (e.g., from ANN, PSO-ANN) for both of datasets (e.g., training and testing) of the models were assessed based on two statistical indices namely, coefficient of determination (R2) and root-mean-squared error (RMSE). In this study, to evaluate the ability of all methods, color intensity rating (CER) based on the result of above indices was developed. Apart from CER, the total ranking system was also used to rank the obtained statistical indexes. As a result, both models presented good performance, however, according to the introduced ranking system, the PSO-ANN model could perform a better performance compared to ANN. According to R2 and RMSE values of (0.9717 and 0.1040) and (0.99131 and 0.0366) were found for training dataset and values of (0.9733 and 0.111) and (0.9899 and 0.0389) obtained for testing dataset, respectively, for the ANN and PSO-ANN approximation models, it can be resulted that PSO-ANN model showed higher reliability in estimating the LSM compared to the ANN.

Modification of landslide susceptibility mapping using optimized PSO-ANN technique

Feasibility of indirect determination of blast induced ground vibration based on support vector machine

This paper presents a new hybrid artificial neural network (ANN) optimized by imperialist competitive algorithm (ICA) to predict peak particle velocity (PPV) resulting from quarry blasting. For this purpose, 95 blasting works were precisely monitored in a granite quarry site in Malaysia and PPV values were accurately recorded in each operation. Furthermore, the most influential parameters on PPV were measured and used to train the ICA-ANN model. Considering the measured data from the granite quarry site, a new empirical equation was developed to predict PPV. For comparison, a pre-developed ANN model was developed for PPV prediction. The results demonstrated that the proposed ICA-ANN model is able to predict blasting-induced PPV better than other presented techniques.

Ground vibration prediction in quarry blasting through an artificial neural network optimized by imperialist competitive algorithm

Blasting is a major component of the construction and mining industries in terms of rock fragmentation and concrete demolition. Blast designers are constantly concerned about flyrock and ground vibration induced by blasting as adverse and unintended effects of explosive usage on the surrounding areas. In recent years, several researches have been done to predict flyrock and ground vibration by means of conventional backpropagation (BP) artificial neural network (ANN). However, the convergence rate of the BP-ANN is relatively slow and solutions can be trapped at local minima. Since particle swarm optimization (PSO) is a robust global search algorithm, it can be used to improve ANNs' performance. In this study, a novel approach of incorporating PSO algorithm with ANN has been proposed to eliminate the limitation of the BP-ANN. This approach was applied to simulate the flyrock distance and peak particle velocity (PPV) induced by blasting. PSO parameters and optimal network architecture were determined using sensitivity analysis and trial and error method, respectively. Finally, a model was selected, and the proposed model was trained and tested using 44 datasets obtained from three granite quarry sites in Malaysia. Each dataset involved ten inputs, including the most influential parameters on flyrock distance and PPV, and two outputs. The results indicate that the proposed method is able to predict flyrock distance and PPV induced by blasting with a high degree of accuracy. Sensitivity analysis was also conducted to determine the influence of each parameter on flyrock distance and PPV. The results show that the powder factor and charge per delay are the most effective parameters on flyrock distance, whereas sub-drilling and charge per delay are the most effective parameters on PPV.

Blasting-induced flyrock and ground vibration prediction through an expert artificial neural network based on particle swarm optimization

Blasting operation is an inseparable operation of the rock fragmentation process in the surface mines and tunneling projects. Ground vibration is one of the most undesirable effects induced by blasting operation which can cause damage to the surrounding residents and structures. So, the ability to make precise predictions of ground vibration is very important to reduce the environmental side effects caused by mine blasting. The aim of this paper is to develop a simple, accurate, and applicable model based on particle swarm optimization (PSO) approach for predicting the ground vibration induced by blasting operations in Shur River dam region, Iran. In this regard, two forms of PSO models, linear and power, were developed. For this work, a database including 80 data sets was collected, and the values of the maximum charge weight used per delay (W), distance between blast-point and monitoring station (D) and peak particle velocity (PPV) were measured. To develop the PSO models, PPV was used as output parameter, while W and D were used as input parameters. To check the performance of the proposed PSO models, multiple linear regression (MLR) model and United States Bureau of Mines (USBM) equation were also developed. Accuracy of models established was evaluated using statistical criteria, i.e., coefficient of correlation (R2) and root mean square error (RMSE), variance absolute relative error (VARE) and Nash & Sutcliffe (NS). Finally, it was found that the PSO power form provided more accurate predictions in comparison with PSO linear form, MLR and USBM models.

Prediction of blast-produced ground vibration using particle swarm optimization

Rock mass is blasted to break it into smaller pieces such as in most surface mining, quarrying operation, dimensional stone mining and some civil engineering application. Flyrock is one of the most hazardous side effects of blasting operation in surface mining. This phenomenon can be considered as the main cause of casualties and damages. The aim of this study is to compare the actual distance of flyrock with the prediction suggested by empirical methods and by using Artificial Neural Network. In addition, this study is also aimed to investigate the most significant input parameters that affecting the flyrock. During this study, flyrock projections for 16 granitic boulders were monitored at Ulu Tiram-quarry site. Blasting parameters such as amount of explosive used, burden, stemming, hole depth, hole angle and hole diameter were carefully measured and recorded. By using these data and applying MATLAB (Matrix Laboratory) program (neural network toolbox), the flyrock distances were predicted for similar condition. The result shows that the coefficient of correlation between the actual and the predicted flyrock distance based on empirical methods is insignificant that is around 0.2. However the result revealed that the coefficient of correlation for overall analysis of flyrock distance is 0.92 based on ANN method. Based on Max-Min method powder factor, stemming and charge length are the most significant parameters in controlling the flyrock distance. This study found that ANN method produced a more accurate prediction than the empirical methods in assessing the actual flyrock projection.

Prediction of flyrock in boulder blasting by using artificial neural network

Blast-induced ground vibration is one of the most important environmental impacts of blasting operations because it may cause severe damage to structures and plants in nearby environment. Estimation of ground vibration levels induced by blasting has vital importance for restricting the environmental effects of blasting operations. This study is aimed to compare the ground vibrations predicted from empirical formula and analytical program with the real data. Several predictor equations have been proposed by various researchers to predict ground vibration prior to blasting, but these are site specific and not generally applicable beyond the specific conditions. To evaluate and calculate the blast-induced ground vibration by incorporating blast design and rock strength, artificial neural networks (ANN) was used. In this study, 12 experiments based on blasting parameters, for modeling in MATLAB software and neural network systems were evaluated to predict the Peak Particle Velocity (PPV) and frequency. It was found that in this study, USBM is one of the most accurate empirical formula for the prediction. The advantage of the ANN compared to other empirical relations used for prediction of PPV is the fact that there was no limitation in the number of input parameters. The correlation coefficients for overall analysis for velocity and frequency are 0.997 and 0.989 respectively. The average relative error obtained from ANN estimation was 0.01 % for velocity and 3.96 % for frequency which are negligible when compared with the those predicted by empirical relationships. This study found that ANN method produced more accurate prediction than the empirical formula.

Seyed Ahmad Noorani

Papers

Blasting-induced flyrock and ground vibration prediction through an expert artificial neural network based on particle swarm optimization

Prediction of blast-produced ground vibration using particle swarm optimization

Prediction of flyrock in boulder blasting by using artificial neural network

Simulation of blasting induced ground vibration by using artificial neural network