露天煤矿边坡稳定性预测的PSO-LSSVM模型  

PSO-LSSV Mmodel for slope stability prediction of open pit coal mine

作　　者：温廷新[1] 张波[1] 

机构地区：[1]辽宁工程技术大学系统工程研究所,辽宁葫芦岛125105

出　　处：《有色金属（矿山部分）》2014年第1期51-56,共6页NONFERROUS METALS(Mining Section）

基　　金：辽宁省教育厅创新团队基金(LT2010048);山东省自然科学基金(ZR2010FL012)

摘　　要：针对边坡工程稳定性预测的复杂性,将粒子群算法和最小二乘支持向量机结合,使用粒子群优化算法寻找最小二乘支持向量机的最优参数,选取七项因素(岩石重度、黏聚力、内摩擦角、边坡角、边坡高度、孔隙水压力和振动系数)作为边坡稳定性的影响因素,建立PSO-LSSVM的边坡稳定性预测模型。利用矿山实测30组边坡稳定性数据进行学习训练,另用12组数据进行测试,同时与LSSVM测试数据进行比较,验证了PSO-LSSVM模型在矿山边坡稳定性预测中有较高的准确度。Aimed at the complexity of coal mine slope engineering stability prediction, the particle swarm optimization algorithm and Least Squares Support Vector Machine （LSSVM） are combined, and the particle swarm optimization algorithm is used to find the optimal parameters of LSSVM. Seven factors are chosen as the influences of slope stability, including rock density, cohesive force, internal friction angle, slope angle, slope height, pore water pressure and vibration coefficient. The PSO-LSSVM model is established for the prediction of mine slope stability. 30 groups real mine slope stability data are used for training, and other 12 groups are used for testing. In the mean- time, they are compared with the LSSVM test data. The result shows that the PSO-LSSVM model has a high accu- racy in mine slope stability prediction.

关 键 词：粒子群算法 最小二乘支持向量机 边坡稳定性 

分 类 号：TD824.71[矿业工程—煤矿开采]