Research On Health Diagnosis Method Of Concrete Dam Based On Improved D-S Evidence Theory

There are nearly 100,000 reservoir dams under construction or built in China,and the rate of reservoir construction has entered a slow period.Most of the reservoirs have been put into use for a long time and are in different degrees of disease and danger,and the health condition of reservoirs is closely related to the lives and properties of downstream residents,so it is very necessary to conduct health diagnosis of reservoir dams.D-S evidence theory is a reliable information fusion algorithm that can fuse many indicators and make a comprehensive judgment with high confidence.However,in the face of high conflict of evidence bodies,the synthesis of evidence theory is prone to a series of paradoxical problems.In view of this,this thesis relies on engineering practice to carry out research on the health diagnosis method of concrete dams based on improved D-S evidence theory.The main research work and results are as follows:(1)Based on Long Gang Reservoir Dam,the influencing factors of concrete dam health were analyzed,and a combined qualitative and quantitative concrete dam health diagnostic index system was constructed.The subjective and objective weights of the primary and secondary indicators in the diagnostic index system were determined by using hierarchical analysis and entropy weighting method,and the combination weights were optimally solved as the optimal interval based on the extreme difference maximization method,and a combination weighting model was established to quantitatively evaluate the importance of the diagnostic indicators of the concrete dam,which provided the basis for the subsequent fusion diagnosis.(2)Based on the cloud model theory,the numerical parameters of quantitative indicators of concrete dams: expectation,entropy and super entropy were obtained by normal cloud,and the health state affiliation of indicators was obtained by constructing the normal cloud diagram of each indicator;for qualitative indicators,the numerical parameters of cloud model were obtained by scoring each indicator with expert evaluation,and the health state affiliation of qualitative indicators was obtained by floating cloud.(3)The index health state affiliation obtained from the cloud model is transformed into a basic probability assignment,the evidence credibility is determined based on the inter-evidence similarity coefficient,the synthesis rule is improved by combining the index weights to eliminate the conflict problem in the fusion process,and the health state of the concrete dam is determined by fusion diagnosis of the optimized evidence body using the improved synthesis rule.A concrete dam health diagnosis method based on the improved D-S evidence theory is proposed to solve the problem of randomness and ambiguity of multi-indicator fusion and improve the accuracy of concrete dam health diagnosis.(4)Based on the theories of empowerment method,cloud model and evidence theory to achieve dam health diagnosis,the theoretical part is programmed in Visual Studio development environment using desktop development software Winform to build an intelligent system for dam health diagnosis.The system has the functions of customizing the dam health diagnosis system,calculating the health state affiliation of quantitative index cloud model and index fusion diagnosis,etc.It can show most of the theoretical parts of health diagnosis through the system.