Rockburst Prediction Based on the KPCA-APSO-SVM Model and Its Engineering Application

The progress of construction and safe production in mining, water conservancy, tunnels, other types deep underground engineering is seriously affected by rockburst disasters. This makes it essential to accurately predict intensity. In this paper, the ratio maximum tangential stress surrounding rock uniaxial compressive strength (σθ/σc), tensile (σc/σt), elastic energy index (Wet) were chosen as input indices, rockbursts graded level I (none rockburst), II (light III (medium IV (strong rockburst). A total 104 groups samples, collected widely from around world, divided into a training set (84 samples) test (20 samples). Based on kernel principal component analysis (KPCA), adaptive particle swarm optimization (APSO) algorithm, support vector machine (SVM), KPCA-APSO-SVM model was established. proposed showed satisfactory classification performance: prediction accuracies 98.81% 95%, respectively. addition, trained applied five cases compared with BP neural network model, SVM APSO-SVM model. comparative results show that has higher accuracy; such, provides new reliable method for prediction.