Inversion Method of Seepage Parameters of Core Rockfill Dam Based on LHS-ISSA-SVR and Its Application
Efficiently and reasonably determining the seepage parameters of dams is the key basis for scientifically an-alyzing the seepage behaviors of dams.In order to address the complex multiple seepage parameters inversion problem of core rockfill dams,this paper proposed a novel method for seepage parameters inversion in core rockfill dams based on the LHS-ISSA-SVR fusion surrogate model.Based on the saturated-unsaturated seepage theory,a finite element seepage model of a core rockfill dam was established.The Latin Hypercube Sampling method was used to obtain the permeability coefficients combination samples with better homogeneity,which is substituted into the finite element model to simulate the water head at the monitoring points.The mapping relationship between the water head and permeability coefficients was accurately constructed by Support Vector Regression(SVR).The Sparrow Search Algorithm(SSA)was effectively improved by fusing the Good Point Set,Levy Flight and Cauchy Inverse Learning Strategies.Both the optimization of SVR model parameters and the intelligent inversion of seepage parameters were carried out based on the improved spar-row search algorithm.The application of engineering examples shows that the relative error between the inversion value and the measured value of water head at each monitoring point is within 2.5%,and the potential distribution law of seep-age field in the core rockfill dam is reasonable,indicating the accuracy and reliability of the inversion results for seepage parameters.This method provides a new way to accurately invert seepage parameters of core rockfill dams in long-term service.
NETL
NSTL
万方数据
