关键词: 粒子群算法, 支持向量机, 围岩松动圈, 仿真预测

Abstract: A new method using Support Vector Machine（SVM） which is optimized by Modified Particle Swarm Optimization（MPSO） to predict loosening zones around roadway is proposed, avoiding shortcomings of current determining methods. Compression factor is introduced to standard PSO. Effective balance between global search and local optimization is achieved. MPSO is applied to optimizing the SVM parameters C and g, and MPSO-SVM regression prediction model is established. The model is applied to predicting loosening zones around roadway. The prediction result is compared with PSO-SVM, GA（Genetic Algorithm）-SVM, GSM（Grid Search Method）-SVM, BP neural network. The results show that the model has better generalization performance and higher prediction accuracy. It can effectively predict loosening zones around roadway.

Key words: Particle Swarm Optimization（PSO）, Support Vector Machine（SVM）, loosening zones around roadway, simulation and forecast