The Research And Application Of Genetic Algorithm In Concrete Arch Dam Safety Monitoring Model

Due to various and complicated factors influencing dam safety, it is urgent to establishdam safety monitoring model and reflect the relationship between cause size and effect sizein dam safety working state. At present, there are three main monitoring models used forstudying the relationship between dam deformation and each influence factors, which arestatistical model, deterministic model and hybrid model. As the high nonlinearity betweenthe influence factors and the dam deformation, the traditional statistical models areineffective in many cases. In order to improve the model precision, it is extremelynecessary to take further study on the dam safety monitoring model.In this dissertation, Stepwise Regression Method and Genetic Algorithm are adoptedto optimize the statistic model respectively. The predictor of conventional statistical modelin concrete arch dam is optimized by the stepwise regression analysis method, and then thedeformation monitoring model in concrete arch dam is established. Also the main factors tohorizontal displacement in various elevation of each dam section are obtained. Thedisplacement monitoring model of concrete arch dam is divided into three componentsincluding hydraulic pressure component, temperature component, and aging component.On basis of those above, the deformation law of dam under each component was studied.Significant factors are obtained by Stepwise Regression Method, then GeneticAlgorithm is introduced to the displacement of concrete arch dam safety monitoring model.Through controlling the fitness function, improving genetic operators and programmingGenetic Algorithm to optimize searching operations and establish monitoring model ofgenetic algorithms.In this dissertation, the author takes the Geheyan gravity arch dam as an example.Through analysis of the displacement monitoring dates of the Geheyan gravity arch dam,the displacement monitoring model are established respectively based on StepwiseRegression Method and Genetic Algorithm. In case of the same predictor number,compared with the stepwise regression displacement monitoring model, the geneticalgorithm displacement monitoring model is better in fitting precision and prediction accuracy.According to the displacement monitoring model by genetic algorithm, the horizontaldisplacement monitoring index of Geheyan gravity arch dam was drafted by confidenceinterval, which can be served as a reference date in dam safety management.