摘要
将萤火虫算法应用于最小二乘支持向量机的核函数及其参数的优化,建立了萤火虫算法-最小二乘支持向量机反分析模型.结合溧阳抽水蓄能电站基本地质条件和监测数据,建立地下厂房块体离散元数值模型,反演获得围岩力学参数.研究结果表明:两个侧压力系数反演值分别为0.493、0.633,三类岩体弹性模量分别为11.82、5.48、2.33 GPa;将反演结果代入数值模型,最终得到的多点位移值与现场监测的位移值的最大误差为11.11%,满足精度要求;利用反演参数进行正演分析,基于地下洞室群开挖过程中的变形协同、应力分布、塑性区探讨了洞室群围岩稳定性.研究成果可为地下洞室稳定性控制提供参考.
Based on the optimization of kernel function and parameters of firefly algorithm applied to least squares support vector machine,a model of firefly algorithm-least squares support vector machine was established.A discrete element numerical model of underground powerhouse block was established in combination with the basic engineering geological conditions and monitoring data of Liyang pumped storage power station.By inputting the displacement value monitored in the field into the model,the mechanical parameters of surrounding rock are obtained by inversion.The two side pressure coefficients are 0.493 and 0.633 respectively,and the three types of elastic modulus are 11.82,5.48 and 2.33 GPa respectively.By substituting the inversion results into the numerical model,the maximum error between the obtained multi-point displacement value and the displacement value monitored in the field is 11.11%,meeting the accuracy requirements.Based on the analysis of deformation coordination,stress distribution and plastic zone distribution during the excavation of Liyang Power station underground cavern group,forward analysis of the surrounding rock stability is carried out by using inversion parameters.
作者
王秀菊
张金龙
戴薇
石崇
WANG Xiuju;ZHANG Jinlong;DAI Wei;SHI Chong(Nanjing Institute of Communications Technology,Nanjing 211188,China;Institute of Geotechnical Engineering,Hohai University,Nanjing 210098,China;South China Survey and Design Institute Co.Ltd.,Power China,Changsha 410014,China)
出处
《河南科学》
2020年第8期1249-1257,共9页
Henan Science
基金
国家自然科学基金面上资助项目(51679071,41831278)。
关键词
参数反演
萤火虫算法
最小二乘支持向量机方法
块体离散元方法
parameter inversion
firefly algorithm
least squares support vector machine method
block discrete element method
