Research On Fault Prediction And Health Evaluation Of Reservoir Bank Slope Based On Safety Monitoring

Bank slope stability is one of the important factors that affect reservoir safety.The integrity and stability of the bank slope are related to the operation of power generation,shipping and irrigation in the reservoir area.In order to ensure the long-term and safe operation of the reservoir area,it is of great significance to the prediction of the failure of the bank slope,the estimation of the remaining life and the evaluation of the health state.The reservoir slope is affected by the disturbance of the reservoir water level fluctuation,rainfall,groundwater level,earthquake and human activities,and its safety status is relatively complex.Once the bank slope is unstable,the direct and secondary disasters will be serious.In this paper,the advanced fault prediction and health management technology are introduced into the bank slope,and the PHM technical framework is set up to realize the failure prediction of the bank slope,the estimation of the remaining life and the evaluation of the health state,which lays the foundation for the health management of the bank slope and the self-reliance maintenance.This article first introduced the characteristics of the safety state of reservoir bank slopes in detail,and analyzed the importance of safety monitoring for bank slopes,according to the structural design method of PHM system,combined with the characteristics of reservoir slope,building the PHM system of the bank slope,analyzing the function of each module;Secondly,the development of fault prediction technology and the commonly used model methods are emphatically introduced.The steps of displacement prediction using BP neural network are introduced in detail.In order to realize the prediction of bank slope fault,the displacement deformation rate and deformation acceleration are selected as failure basis,and the prediction of the displacement is combined with the displacement prediction value to predict the failure of the bank slope;in order to reflect the state of bank slope from many angles,the prediction module for remaining life of bank slope is constructed,expounding the concept of remaining life and the main method of remaining life estimation,considering the randomness of the degradation of the bank slope under the influence of the external environment,this paper focuses on the Wiener process and the process of the degradation modeling,and defines the life of the bank slope by using the concept of inverse Gauss distribution in the first time.The probability density function of the life of the bank slope is estimated by the maximum likelihood method,and the mean residual life is estimated by using the concept of mean residual life;in order to consider the status of reservoir slope under various factors,the fuzzy comprehensive evaluation method is applied to evaluate the health status of the bank slope,and the evaluation factor set and evaluation level of the bank slope are determined.Using fuzzy statistics method,membership function and analytic hierarchy process to determine the degree of membership and the weight of each assessment factor corresponding to the health level,and evaluate the historical status of the bank slope and the health status in the future period.Through the research on the failure prediction of the bank slope,the remaining life and the health evaluation technology,the feasibility of implementing the reservoir bank slope PHM system was verified,and combined with the actual situation,the frame and main functions of the bank slope PHM system were realized.And it provides the basis for the ultimate realization of the health management of the bank slope.