Prediction of Rock Fragmentation Due to Blasting in Sarcheshmeh Copper Mine Using Artificial Neural Networks

Abstract

The main objective in production blasting is to achieve a proper fragmentation. In this paper, rock fragmentation the Sarcheshmeh copper mine has been predicted by developing a model using artificial neural network. To construct the model, parameters such as burden to spacing ratio, hole-diameter, stemming, total charge-per-delay and point load index have been considered as input parameters. A model with architecture 9-8-5-1 trained by back propagation method was found to be optimum. To compare performance of the neural network, statistical method was also applied. Determination coefficient (R2) and root mean square error were calculated for both the models, which show absolute superiority of neural network over traditional statistical method.