论文部分内容阅读
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。
In order to accurately predict the risk of rock burst, a prediction model of least squares support vector machine (LSSVM) is proposed to optimize the dynamic integration of Bagging algorithm. Based on the design and optimization of the Bagging-LSSVM model, the classical classification data set is introduced to verify the feasibility of the model. The minimum number of base classification models under the optimal classification condition is obtained through experiments. Considering the main influencing factors of rock burst and determining its evaluation index, 35 groups of measured data of Chongqing Yanshi Tai Coal Mine were used as test samples, KPCA was used to eliminate the correlation between the indexes, and comparative analysis of sample data processing The prediction results of the model before and after the application are compared. The Bagging-LSSVM model is optimized and the Bagging-SVM model and the LSSVN model are optimized to predict the accuracy of the rock burst risk. The results show that compared with the original sample, the KPCA-treated samples are more accurate and less time-consuming than the Bagging-LSSVM model, and the Bagging-LSSVM model is more accurate in predicting the risk of rock burst Higher than other models.