Prediction of airblast levels at Anglogold Iduapriem mine using empirical models

Abstract

The study used empirical predictor models to predict airblast (vibration) levels at AngloGold Iduapriem mine in Tarkwa, Western Region of Ghana to address the complaints of Teberebie and Domeabra citizens on the effects of blasting activities on their structures and environment. The predictor empirical models adopted for the study were USBM, National Association of Australian State, McKenzine’s, Ollofson’s, Holmberg and Persson’s predictors. The Root Mean Squared Error (RMSE), Coefficient of determination (R2) and the standard error equation (s) performance indices were used to determine and compare the accuracy, precision and the associated level of deviation of each predictor from the measured data. Measured airblast levels, cooperating charges (kg) and the distances from the blast face to the monitoring point data were obtained from the mining department of the mine. The site constants of AngloGold Iduapriem mine according to the USBM Predictor are H = 141.48 and ß = 0.0568 respectively. The performance indices (R2, RMSE and s) for the USBM Predictor have the values of 0.9913, 0.2729 and 9.60 while that of McKenzine’s predictor model has the values of 0.9886, 0.2982 and 11.41 respectively. The study, therefore, concluded that the USBM predictor is the best empirical model to predict the airblast level at Anglogold Iduapriem Mine.