The Study And Application Of EV Model On Deformation Monitoring Of Dams

It is an important means to analyze the observation material to mastery the working state of dams, the safety degree of dams also can be judged according to the observation material. So it is one of the important tasks to make timely, reasonable and effective analysis for dam observation data.Regression model has developed theory as a basic method in analyzing the observation material of dams. The method has many merits, for instance, it is simple, not demands complicated calculation, and its predicting period is comparative long. The most important of all, the regression can explain the relations in all variables. But the regression doesn't consider the errors in all variables; hence the result can be anamorphic. According to regression's shortage, this paper introduced a new model, namely EV (Errors-in-Variables) model. This paper studied the basic theories of EV model and the means for parameter estimation, and explored the application in analyzing the observation material of dams. The main contents of this paper are as follows:1. The characteristics and basic theories of EV model is studied compared with the regression. And this paper discussed the species of EV models and the means for parameter estimation.2. Estimation of nonlinear EV models is studied. Presented a general iterative formula for parameter estimation based on the maximum-likelihood estimation, and presented an easier arithmetic so-called HEIV arithmetic for a kind of especial equations whose parameters can be separated form variables.3. Explored the application of nonlinear EV model in observation datum of Qingtongxia dam, and established material nonlinear EV model. Solved the errors covariance matrix problem, and mended the model. Furthermore, analyzed the precision of the model, and discussed its capability in forecasting distortion.4. Presented the statistics diagnostics of EV models, and discussed several diagnostic measures. At length, this paper explored the application of EV model's statistics diagnostics in the displacement datum of Qingtongxia dam.