SSA-MLP模型在岩质边坡稳定性预测中的应用  

Application of SSA MLP model to stability prediction of rock slope

作　　者：侯克鹏[1,2] 包广拓 孙华芬 HOU Kepeng;BAO Guangtuo;SUN Huafen(School of Land and Resources Engineering,Kunming University of Science and Technology,Kunming 650093,China;Yunnan Key Laboratory of Sino-German Blue Mining and Utilization of Special Underground Space,Kunming 650093,China)

机构地区：[1]昆明理工大学国土资源工程学院,昆明650093 [2]云南省中德蓝色矿山与特殊地下空间开发利用重点实验室,昆明650093

出　　处：《安全与环境学报》2024年第5期1795-1803,共9页Journal of Safety and Environment

基　　金：云南省科技厅项目(202101AT070094)。

摘　　要：岩质边坡的力学参数量化及稳定性分析对岩质边坡灾害的防治具有重要意义。Hoek-Brown(H B)准则是一种用于确定岩体力学参数的经典方法,能反映出边坡岩体变形和位移的非线性破坏特征。在此基础上,首先,提出一种麻雀搜索算法(Sparrow Search Algorithm,SSA)改进多层感知器(Multi-Layer Perceptron,MLP)的神经网络模型,并用于边坡稳定性预测、指标敏感性分析及参数反演。其次,将收集的1085组岩质边坡的几何参数和H B准则参数等作为输入变量,极限平衡理论Bishop法求解的安全系数作为输出变量,对SSA MLP模型进行训练学习和性能评估。最后,将该模型运用于25个边坡实例,验证模型的有效性。结果显示,该模型收敛速度快、精度高,为边坡稳定性分析和参数量化提供了一种新思路。In this paper,a new method of rock slope stability prediction and parameter inversion is proposed by combining the rock strength criterion,swarm intelligence algorithm,and artificial neural network theory.Hoek-Brown(H B)strength criterion is a classical method used to determine the mechanical parameters of rock mass,which can reflect the nonlinear failure characteristics of rock mass deformation and displacement,and has good applicability in the stability analysis of rock slopes.Therefore,based on the parameters of the H B criterion,the index system of rock slope stability prediction is constructed.Then,according to the complex nonlinear characteristics of slope engineering stability problem,a prediction model of Multi-Layer Perceptron(MLP)neural network improved by Sparrow Search Algorithm(SSA)is proposed for the prediction of safety factor and parameter inversion of rock slope.The parameter optimization function of the sparrow search algorithm is used to optimize the connection weights and thresholds of the multi-layer perceptron neural network,and the SSA MLP neural network model with higher prediction accuracy is obtained.Besides,according to the geometric parameters and H B criterion parameters of 1085 sets of rock slopes collected as input variables,and the slope safety factor solved by the Bishop method based on the limit equilibrium theory as output variables,the training learning and performance evaluation of the SSA MLP model are carried out,and compared with other network models.It is analyzed that the model had high feasibility.In addition,the sensitivity analysis between the safety factor and the characteristic index is carried out by using the SSA MLP model and the Kendall correlation coefficient.Finally,the model is applied to 25 slope cases,and the parameter inversion of the disturbance coefficient(D)and the geological strength index(G SI)in engineering cases is carried out to further verify the effectiveness of the model The results show that the model has fast convergence speed and high precisio

关 键 词：安全工程 边坡稳定性 HOEK-BROWN准则 多层感知器(MLP)神经网络 麻雀搜索算法 参数反演 

分 类 号：X43[环境科学与工程—灾害防治]