Intelligent Prediction Of Residual Life Of Diversion Channel Slope Based On Multivariate Wiener Process

The water diversion project is a livelihood project to meet the needs of people’s daily life and achieve the sustainable utilization of water resource.As an important part of the water diversion project,the stability of the channel slope is directly related to the water benefit of the diversion project.Through monitoring,we can master the related characteristics of landslide,and timely life prediction can ensure the stability of the slope.Therefore,monitoring and predicting the health state of the slope is an inevitable choice to deal with geological disasters before they happen.With the passage of time,under the comprehensive influence of a variety of factors,the channel slope will be damaged at the level of material and structure.which will lead to the degradation of slope performance,stability,and even landslide in serious cases,resulting in unpredictable losses.The residual life prediction method based on performance degradation can predict the residual life by modeling and analyzing the performance index representing the health state.This method has gradually become a research focus,however,there are not many applications in water diversion project at present,which is the starting point of this paper.The main research contents of this paper are as follows:Considering that only a single performance index can not fully reflect the running state of the slope,this paper considers to use multiple performance indexes to comprehensively reflect the health state of the slope.The binary Wiener process is used for analysis,and determines the reasonable threshold according to the safety factor corresponding to the creep characteristics of the slope.Considering the correlation between the two performance indexes and choosing the appropriate Copula function to describe,the joint probability density function of slope life is established to predict its remaining life.Compared to the unary Wiener process,the binary Wiener process can take into account more monitoring information and reflects the health state of slope more accurately and comprehensively.Taking the Xintan landslide as an example,the effectiveness of the proposed method was verified and applied to the slope of a diversion channel in service.The results show that the prediction is consistent with the practice,and the proposed method can provide theoretical and technical support for the safe operation of the slope.In order to obtain the displacement trend of the channel slope in the future,to better grasp the changes of the remaining life in the future time,according to the operating characteristics of the channel slope,the influencing factors are selected to construct the displacement prediction model of the PSO-SVM.The displacement prediction analysis of the creeping diversion channel slope is carried out,and the displacement prediction value is substituted into the remaining life prediction model of the binary Wiener process to construct the remaining life intelligent pre-analysis model.Taking the slope of an in-service diversion channel as an example,the intelligent pre-analysis of remaining service life is carried out.The results show that this method can provide a scientific basis for managers to formulate reasonable preventive maintenance strategies.