基于优化Bagging-LSSVM模型的冲击地压预测

发布时间：2019-02-13 03:51

【摘要】：为准确预测冲击地压危险性,提出一种优化Bagging算法动态集成的最小二乘支持向量机(LSSVM)的预测模型。在设计和优化Bagging-LSSVM模型流程的基础上,引入经典分类数据集,验证模型的可行性,并通过试验得出实现模型最优分类条件下的基分类模型数的最小值。综合考虑冲击地压的主要影响因素,确定其评判指标;以重庆砚石台煤矿的35组实测数据为试验样本,利用核主成分分析(KPCA)消除指标间的相关性,对比分析样本数据处理前后应用模型的预测效果;比较优化Bagging-LSSVM模型、优化Bagging-SVM模型和LSSVN模型预测冲击地压危险性的准确率。结果表明:经KPCA处理后的样本相较于原始样本,其应用于优化Bagging-LSSVM模型的预测准确率更高,耗时更少;且优化Bagging-LSSVM模型预测冲击地压危险性的准确率高于其他模型。
[Abstract]:In order to accurately predict the risk of rock burst, a prediction model of (LSSVM) based on least squares support vector machine (LS-SVM), which optimizes the dynamic integration of Bagging algorithm, is proposed. Based on the design and optimization of Bagging-LSSVM model flow, the classical classification data set is introduced to verify the feasibility of the model, and the minimum value of the number of base classification models under the optimal classification conditions is obtained through experiments. Considering the main influencing factors of impact ground pressure, the evaluation index is determined. Taking 35 groups of measured data of Yanshitai coal mine as experimental samples, the correlation between (KPCA) elimination indexes was analyzed by using kernel principal component analysis, and the prediction results of the model before and after sample data processing were compared and analyzed. Compare the accuracy of optimized Bagging-LSSVM model, optimized Bagging-SVM model and LSSVN model in predicting rock burst hazard. The results show that compared with the original samples, the sample processed by KPCA has higher prediction accuracy and less time consuming when applied to optimize the Bagging-LSSVM model, and the accuracy of the optimized Bagging-LSSVM model is higher than that of other models.
【作者单位】：辽宁工程技术大学系统工程研究所;辽宁工程技术大学工商管理学院;
【基金】：国家自然科学基金资助(71371091) 辽宁省高等学校杰出青年学者成长计划项目(LJQ2012027)
【分类号】：TD324

【相似文献】