Rock Burst Intensity Classification Prediction Model Based on a Bayesian Hyperparameter Optimization Support Vector Machine

Rock burst disasters occurring in underground high-stress rock mass mining and excavation engineering seriously threaten the safety of workers hinders progress construction. classification prediction is basis reducing even eliminating hazards. Currently, most mainstream discriminant models for grade are based on small samples. Comprehensive selection according to many pieces literature, maximum tangential stress surrounding uniaxial compressive strength ratio coefficient (stress state parameter), tensile (brittleness modulus), elastic energy index used as a grading evaluation collection different construction instances 114 groups extensive sample data regions world, which carry out training study. The representativeness accuracy were verified by indicator variance analysis Spearman correlation hypothesis test. Intelligent Identification System (IRIS) an optimizable SVM model was established using set After cross-validation training, can reach 95.6%, significantly better than traditional 71.9%. classify predict intensity 10 typical projects at home abroad. results consistent with actual intensity, other existing models. application examples shows that processing method good agreement effectively provide reference area.