| With the national economic development,the domestic hydropower further development,the main market moves to the southwest and west,followed by the size of the high CCRD from 200m to 300m level across,The stress environment of the rockfill particles of the 300m-grade the high CCRD has changed greatly compared with the 200m the high CCRD,the deformation characteristics exhibited by the high confining pressure are more complicated.Due to the presence of the scale effect,at present,the rockfill constitutive model and the method of measuring the parameters of indoor test can not effectively simulate the stress and deformation of dam,stress-deformation characteristics and trends can not be effectively predicted during the process of dam construction or later operation,making the construction of the face rockfill dam safety,economy can not be effectively protected,parameter inversion is an effective means to guide the dam construction and deformation prediction based on the measured data.The existing method of inversion of rockfill parameters is mainly single target method,the use of an objective function for the whole dam or a section is not considered for the regional and global effects of the rockfill material,all materials are integrated together,the mapping relationship is complex,the inversion effect is poor and the result is unstable.In this paper,the idea of multi-objective is introduced,the dam is divided into several sub-intervals,and the inversion objective function is established in each sub-interval,which considers the regional impact of the material,and based on the non-dominated multi-objective genetic algorithm,which takes into account the overall impact of the material,a multi-objective parameter inversion method for rockfill dam based on NSGA-Ⅱ and RBF neural network is proposed.The main contents of this paper are as follows:1)The basic principle of NSGA-Ⅱ and RBF neural network is introduced,and the lap technology for inversion engineering of rockfill dam parameters,a multi-objective parameter inversion platform for high rockfill dam based on NSGA-Ⅱ algorithm and RBF neural network,and systematically expounded the various aspects of the inversion platform.2)Summarizes the principles and typical forms of rockfill dam division,and put forward the basic principles,methods and advantages of sub-interval division,take a typical material partitioning model as an example,the standard dam model was established to study the regional sensitivity of materials and Summarize the rules,the regional effect of rockfill material on deformation is verified,and an inverse sub-partition scheme for this partition pattern is proposed.3)The basic principle of E-B constitutive model and rheological constitutive model of rockfill body and the finite element method of nine parameter rheological model are expounded.through the sensitivity study of the mechanical parameters of the transient and rheological constitutive models,the most suitable inversion parameters are selected.By arranging and analyzing the deformation detection data of Shuibuya concrete face rockfill dam,the most suitable parameter inversion section and measuring point are selected.4)In this paper,the transient and rheological parameters of the main rockfill and secondary rockfill zones are analyzed by the above method,and compared with the single target inversion method.The multi-objective parameter inversion method is not only consistent with the measured value in the deformation and deformation trend,but also the inversion calculation value of all the monitoring points agrees well with the measured values,and the stability of the calculated results is better,the overall deformation and stress Distribution laws are in line with reality,but the results of inversion of single target parameters are only close to the measured values in the local region and the calculated results are unstable,the inversion results of the multi-objective parameters are obviously superior to the single-objective parameters inversion,and the multi-objective parameter inversion method has the rationality and the superiority. |
