Prediction of Blast-Induced Flyrock in Opencast Mines Using ANN and ANFIS

Abstract

The aim of present study is prediction of blast-induced flyrock distance in opencast limestone mines using artificial intelligence techniques such as artificial neural network (ANN) and adaptive neuro fuzzy inference system (ANFIS). Blast design and geotechnical variables such as linear charge concentration, burden, stemming length, specific charge, unconfined compressive strength, and rock quality designation have been selected as independent variables and flyrock distance has been used as dependent variable. Blasts required for the study purpose have been conducted in four limestone mines in India. Out of one hundred and twenty-five (125) blasts, dataset of one hundred blasts have been used for training, testing and validation of the ANN and ANFIS based prediction model. Twenty-five (25) data have been used for evaluation of the trained ANN and ANFIS models. In order to know the relationship among the independent and dependent variables, multi-variable regression analysis (MVRA) has also been performed. The performance indices such as root mean square error (RMSE), mean absolute error (MAE) and coefficient of determination (R2) have been evaluated for ANN, ANFIS and MVRA. RMSE as well as MAE have been found lower and R2 has been found higher in case of ANFIS in comparison of ANN and MVRA. ANFIS has been found a superior predictive technique in comparison to ANN and MVRA. Sensitivity analysis has also been performed using ANFIS to assess the impact of independent variables on flyrock distance.