Monitoring Model Of Concrete Dam Deformation Safety Based On Copula Theory

Deformation is the most important effect quantity that characterizes the change of the working behavior of concrete dams.Analyzing and monitoring it is one of the most important research contents in the field of dam engineering.The preprocessing of deformation information,the optimization of the components that affect the deformation,the construction of single-point and multi-point deformation prediction models are all hot spots and difficulties in the dam engineering discipline.In this paper,comprehensive use of mathematical and mechanical methods,dam engineering and statistical theory and knowledge,computer technology,in-depth study of the monitoring information characterizing the deformation behavior of concrete dams,the establishment of concrete dam deformation prediction models.The main research contents are as follows:(1)According to the operation monitoring results of a concrete gravity dam for many years,the measurement points of typical dam sections are selected,and the environmental measurement process line and deformation monitoring process process line are drawn,and a qualitative analysis is carried out on a concrete gravity dam.The analysis carries out characteristic value analysis and comparison and comparison of the environmental quantity and deformation monitoring quantity,so as to have a qualitative understanding of its changing law and a preliminary judgment as to whether it is abnormal.(2)Analyze and compare the advantages and disadvantages of the Copula function applied to the selection of dam deformation monitoring influence factors,and propose to establish a Copula-RF-based dam deformation monitoring model and compare it with the least square method.Based on the monitoring data in the actual project,Copula function is used to perform nonlinear correlation test on the influence factor set,and the optimal factor set is selected.The random forest(RF)method is used to predict the deformation of the dam,and the accuracy of the modified model is verified through engineering examples.(3)The MVRVM multi-output correlation vector machine multi-point deformation monitoring model is proposed.Based on the monitoring data of a concrete gravity dam,the data is also preprocessed by gross error removal and normalization,and the Copula function is used to factor s Choice.The good function approximation ability of the correlation vector machine is used for regression prediction of multiple measurement points.The example verification shows that the accuracy of the model can reach the accuracy level of the one-sided point model.