Prediction and Assessment of Rock Burst Using Various Meta-heuristic Approaches

One of the utmost severe mining catastrophes in underground hard rock mines is rock burst phenomena. It can lead to damage to mine openings and equipment as well as trigger accidents or even threat to life as well. Due to this, a number of researchers are forced to study some easy-to-use alternative methods to predict the rock burst occurrence. Nevertheless, due to the extremely multifaceted relation between mechanical, geological and geometric factors of the mines, the conventional prediction methods are not able to produce accurate results. With the expansion of machine learning methods, a revolution in the rock burst occurrence has become imaginable. In present study, three machine learning methods, namely XGBoost, decision tree and support vector machine, are utilized to predict the occurrence of rock burst in various underground projects. A total of 134 rock burst events were gathered together from various published literatures comprising maximum tangential stress (MTS), elastic energy index (EEI), uniaxial compressive strength and uniaxial tensile stress (UTS) that have been used to develop various machine learning models. The performance of machine learning methods is evaluated based on the accuracy, sensitivity and specificity of the rock burst prediction.