Search

In this paper, an attempt has been made to implement various robust techniques to predict rock fragmentation due to blasting in open pit mines using effective parameters. As rock fragmentation prediction is very complex and complicated, and due to that various artificial intelligence-based techniques, such as artificial neural network (ANN), classification and regression tree and support vector machines were selected for the modeling. To validate and compare the prediction results, conventional multivariate regression analysis was also utilized on the same data sets. Since accuracy and generality of the modeling is dependent on the number of inputs, it was tried to collect enough required information from four different open pit mines of Iran. According to the obtained results, it was revealed that ANN with a determination coefficient of 0.986 is the most precise method of modeling as compared to the other applied techniques. Also, based on the performed sensitivity analysis, it was observed that the most prevailing parameters on the rock fragmentation are rock quality designation, Schmidt hardness value, mean in-situ block size and the minimum effective ones are hole diameter, burden and spacing. The advantage of back propagation neural network technique for using in this study compared to other soft computing methods is that they are able to describe complex and nonlinear multivariable problems in a transparent way. Furthermore, ANN can be used as a first approach, where much knowledge about the influencing parameters are missing.

Blast-induced ground vibration is considered as one of the most hazardous phenomena of mine blasting, which can even cause casualties and severe damages to the adjacent properties. Measuring peak particle velocity (PPV) is helpful to know the actual vibration level but prediction of blast vibration prior to the blast is a tedious job due to involvement of blast design, explosive and rock parameters. Nowadays, efficient application of intelligent systems has been approved in different branches of science and technology. In this paper, a gene expression programming (GEP) model was developed to predict PPV using various blasting patterns as model inputs, which showed a high level of accuracy for the implemented model. Also, to optimize blast pattern attaining minimum ground vibration during blasting operation, the developed functional GEP model was taken as objective function for grasshopper optimization algorithm (GOA). Construction of GOA model was performed using a trial and error mechanism to find out the best possible pertinent GOA parameters. Finally, it was observed that utilizing GOA technique, PPV can be reduced by 67% with optimized blast parameters including burden of 3.21 m, spacing of 3.75 m, and charge per delay of 225 kg. A sensitivity analysis was also performed to understand the influence of each input parameters on the blast vibrations.

Blasting operations create significant problems to residential and other structures located in the close proximity of the mines. Blast vibration is one of the most crucial nuisances of blasting, which should be accurately estimated to minimize its effect. In this paper, an attempt has been made to apply various models to predict ground vibrations due to mine blasting. To fulfill this aim, 112 blast operations were precisely measured and collected in one the limestone mines of Iran. These blast operation data were utilized to construct the artificial neural network (ANN) model to predict the peak particle velocity (PPV). The input parameters used in this study were burden, spacing, maximum charge per delay, distance from blast face to monitoring point and rock quality designation and output parameter was the PPV. The conventional empirical predictors and multivariate regression analysis were also performed on the same data sets to study the PPV. Accordingly, it was observed that the ANN model is more accurate as compared to the other employed predictors. Moreover, it was also revealed that the most influential parameters on the ground vibration are distance from the blast and maximum charge per delay, whereas the least effective parameters are burden, spacing and rock quality designation. Finally, in order to minimize PPV, the developed ANN model was used as an objective function for imperialist competitive algorithm (ICA). Eventually, it was found that the ICA algorithm is able to decrease PPV up to 59% by considering burden of 2.9 m, spacing of 4.4 m and charge per delay of 627 Kg.

Iron is one of the most applicable metals in the world. The global price of iron ore is determined based on demand and supply. There are numerous parameters (e.g., price of steel, steel production, oil price, gold price, interest rate, inflation rate, iron production, and aluminum price) affecting the global iron ore price. Considering the high number of effective parameters and existence of complex relationship among them, artificial intelligence-based approaches can be employed to predict iron ore price. In this paper, a new intelligence system namely group method of data handling (GMDH) was developed and introduced to predict the price of iron ore. For comparison purposes, four other techniques i.e., autoregressive integrated moving average (ARIMA), support vector regression (SVR), artificial neural network (ANN), and classification and regression tree (CART) were developed for prediction of monthly iron ore price. Then, using testing datasets, the developed models were validated and their performance capacities were compared. The results showed that performance prediction of the GMDH model is significantly better than other predictive models based on four performance indices i.e., root mean square error, variance account for (VAF), mean absolute error, and mean absolute percentage error. Results of VAF (97.89%, 90.81%, 80.95%, 55.02%, and 23.87% for GMDH, SVR, ANN, CART, and ARIMA models, respectively) revealed that the GMDH technique is able to predict iron ore price with higher degree of accuracy compared to the other techniques.

In the civil and mining projects, blasting operation is important from technical and economical point of view. There are several parameters which affect the result of operation such as desired fragmentation and undesired phenomena, e.g., ground vibration, fly rock, etc. From these parameters, rock mass characterizations can be considered as more influential as compared to the blasting pattern. In other words, it can be said that pattern specifications should primarily be designed according to the rock mass properties to reach the main objective of the operation, i.e., rock fragmentation. Complex nature of the problem needs to implement robust approaches such as artificial intelligence-based techniques. In this paper, an attempt has been made to develop some models by which the impact of each and every parameter influencing the result of blasting operation can be evaluated. For this research work, 432 datasets from 14 mines situated in the different parts of the world has been collected. In developing of the models, 19 parameters such as uniaxial compressive strength, tensile strength, brittleness, Point Load Index, Young's modulus, Poisson's ratio, rock quality designation, cohesion, friction angle, burden, spacing and stemming were incorporated. Regression analysis, decision tree and artificial neural network methods were employed for developing the models for predicting fragmentation. Determination coefficient (R2) for artificial neural network modeling, multivariate linear regression and decision tree was computed 0.98, 0.83 and 0.45, respectively, showing accuracy of network modeling over the other applied methods. In addition, it was revealed that the most influential parameters on fragmentation are Point Load Index, uniaxial compressive strength, Poisson's ratio, cohesion and rock quality designation, respectively, and the least effective ones are stemming, spacing and hole diameter, respectively.

Flyrock, the propelled rock fragments beyond a specific limit, can be considered as one of the most crucial and hazardous events in the open pit blasting operations. Involvement of various effective parameters has made the problem so complicated, and the available empirical methods are not proficient to predict the flyrock. To achieve more accurate results, employment of new approaches, such as artificial neural network (ANN) can be very helpful. In this paper, an attempt has been made to apply the ANN method to predict the flyrock in the blasting operations of Sungun copper mine, Iran. Number of ANN models was tried using various permutation and combinations, and it was observed that a model trained with back-propagation algorithm having 9-5-2-1 architecture is the best optimum. Flyrock were also com- puted from various available empirical models suggested by Lundborg. Statistical modeling has also been done to com- pare the prediction capability of ANN over other methods. Comparison of the results showed absolute superiority of the ANN modeling over the empirical as well as statistical models. Sensitivity analysis was also performed to identify the most influential inputs on the output results. It was observed that powder factor, hole diameter, stemming and charge per delay are the most effective parameters on the flyrock.