Study On Comprehensive Diagnosis Of Behavior Of Inservice Concrete Gravity Dam Based On Analysis Of Observation Data

With the construction of nearly 100 thousand dams in China and the increase of the age of old dams in service life,the main focus of dam construction field has gradually changed from "engineering construction" to "safety management" concept.More and more scholars pay attention to the study of safe operation management and health diagnosis theory of high dam and large reservoir.The research on dam safety monitoring theory is not only of great significance to ensure the stable and safe operation of dam,but also has a broad prospect of application and development.This paper aims at the problems existing in the current safety monitoring model and health diagnosis system of concrete dam.In view of the existing problems in the safety monitoring model and health diagnosis system of concrete dam,this paper starts from the construction of the safety evaluation system of concrete dam,takes the deformation and seepage of the operation monitoring data as the analysis object,using Principal Component Analysis(PCA),Support Vector Machine(SVM),Analytic Hierarchy Process(AHP),Grey Wolf Optimization(GWO),Information Entropy and Cloud Model and other mathematical theories and bionic intelligent algorithms to improve the existing methods.The high prediction accuracy and scientific diagnosis result are realized,which provides a new way for the safe operation management of concrete dam.The main research contents of this paper are as follows:(1)According to the principle of concrete dam evaluation system,a safety evaluation system of concrete dam operation physical characteristics with multimeasuring points and multi-monitoring indexes is constructed.Based on the fuzzy comprehensive diagnosis theory,the weight of each evaluation index is established by the combination of analytic hierarchy process and improved entropy weight method,and the evaluation set of the health condition of the four-level concrete dam between the third and fifth levels is proposed.(2)The contribution degree of each influencing factor to the behavior evolution of concrete dam is discussed,and the principal component analysis method is used to pretreat the monitoring data of independent variables with dimension reduction.Using the monitoring data after dimension-reduction,a GWO-SVM monitoring model for the deformation and seepage of concrete dam is established.The fitting and forecasting accuracy of the model are improved,and the rationality and feasibility of the model are verified.(3)According to the safety evaluation system of concrete dam operation characteristics and the way of weighting constructed above,the zoning standard and quantification method of different evaluation indexes of concrete dam operation are given.Considering the fuzziness and randomness in the evaluation process,a fuzzy comprehensive cloud diagnosis model of an inservice concrete gravity dam based on feedback control analysis of observation data is established.The result of diagnosis is consistent with the engineering analysis report.