改进支持向量机在水工结构健康监测中的应用水工结构工程专业论文.docxVIP

The The Application of the Improved Support Vector Machine in the Health Monitoring Hydraulic Structure Author：Cheng Xiyang Major：Hydraulic Structure Engineering Supervisor：Qie Zhihong Abstract With the increase of the running time of hydraulic structures，it is inevitable that aging， disease and even accidents happen．Therefore，the health problems of hydraulic structures have been increasingly concerned．In order to reduce the operational risks of hydraulic structure，the health monitoring of hydraulic structures must be implemented．The important research contents in the field of hydraulic structure health monitoring the deterioration diagnosis evaluation of hydraulic structure and the structural parameter inversion analysis，which also hotspots in the field of hydraulic structure．In recent years，more and more artificial intelligence technology is applied to hydraulic structure health monitoring．When using the Support Vector Machine(SVM)for machine learning， parameter selection depends the accuracy of the classification and prediction of the Support Vector Machine(SVM)．There are traditional methods and swarms of intelligence algorithm previous researches about the parameter selection of Support Vector Machine． Aimed at the shortcomings of the previous selection method，a newly developed group of intelligent algorithm，Glowworm Swarm Optimization Algorithm is introduced．In this paper，GSOA—SVM is used to evaluate the degradation of hydraulic structures and the inversion analysis of structural parameters．Therefore，the main contents of this paper summarized follows： (1)The basic principles and algorithm processes of Support Vector Machine and Glowworm Swarm Optimization Algorithm introduced。The parameters which influenced the optimization performance of SVM were also analyzed．The Glowworm Swarm Optimization Algorithm is adopted tO optimize the parameter，which effectively enhance the optimized ability of Support Vector Machine algorithm．Hence，the improved vector machme modeli