山岭隧道围岩参数SSA-KELM模型智能反演分析

Analysis on Intelligent Inversion of SSA-KELM Model for Surrounding Rock Parameters of Mountain Tunnel

为确保隧道围岩力学参数取值的可靠性,提出一种基于FSSA-KELM模型的新型隧道围岩参数智能反演模型(SSA-KELM)。将智能模型应用于福宜高速阳宗隧道左线某区间隧道围岩力学参数反演过程;在FLAC^(3D)数值模拟软件中进行隧道变形实测值正演,计算出围岩位移值,并与现场实测值进行对比分析。结果表明:SSA算法优化了KELM模型的正则化参数C和核函数参数,使模型计算精度大幅提高;SSA-KELM模型在隧道围岩变形方面计算结果相对误差均小于4%,远低于ELM模型和BP模型,说明SSA-KELM模型反演结果更精确,模型实用性更强。

In order to ensure that the mechanical parameters of tunnel surrounding rock are reasonable and reliable,a new intelligent inversion model of tunnel surrounding rock parameters based on FSSA-KELM model is proposed.The intelligent model is applied to the inversion of the mechanical parameters of the surrounding rock in the left line section of Yangzong Tunnel on Fude-Yiliang Expressway.The displacement value of the surrounding rock is calculated forward in FLAC^(3D) numerical simulation software based on the measured value of the tunnel deformation on site,and is compared with the measured value on site.The results show that SSA algorithm optimizes the regularization parameter C and kernel function of the KELM model,significantly improving the computational accuracy of the model.The relative error of the SSA-KELM model in calculating the deformation of tunnel surrounding rock is less than 4%,which is much lower than that of the ELM model and BP model,indicating that the SSA-KELM model has more accurate inversion results and stronger practicality.