Study On Thin Plate Structure Damage Identification Based On Wavelet-Differential Evolution Algorithm

With the progress of society and science and technology,some complex structures keep emerging,which inject power into human survival and industrial development.However,due to the adverse factors in construction and use,the structure will inevitably be damaged during the service life.When the damage accumulates gradually,the service life of the structure will be reduced,and the serious damage will lead to the deformation and collapse of the structure,endangering the safety of human life and property.Therefore,it is very necessary to carry out damage identification and health and safety monitoring of the structure.In this paper,the damage problem of thin plate is studied,and a two-stage damage identification method based on wavelet-differential evolution algorithm is proposed.This method can not only overcome the problem that the wavelet transform is difficult to identify the damage degree,but also reduce the problem that the dimension caused by the differential evolution algorithm is too high and the calculation work is too heavy.In this paper,the theory is verified by experiments.The damage position and damage degree of steel sheet were studied.The modal parameters of the steel sheet were collected by the acceleration sensor in the test,and the collected modal information was imported into the computer for sorting and sorting,and the sorted data was imported into MATLAB for processing.Finally,the damage location and damage degree information were obtained through software analysis.This paper analyzes and summarizes the problems in the test,finds out the reasons for the problems,and makes improvements for the future test.Secondly,in order to further study the impact of damage degree and boundary conditions on the identification effect,the damage problem of thin plate was numerically simulated in this paper.The displacement mode and angular mode of the thin plate are obtained by establishing the finite element model.In the process of location identification,the effects of different damage degrees and boundary conditions on damage location identification were compared by setting different working conditions,and the effects of cell size division on damage location identification were discussed.Finally,the identification of single damage degree and multiple damage degree is simulated numerically.In this paper,the objective function is constructed based on the natural frequency and displacement mode of thin plate and solved by differential evolution algorithm.In order to further study the advantage of the differential evolution algorithm,this paper chooses the particle swarm optimization algorithm and the differential evolution algorithm to compare the recognition results.Under the same working condition,the performance of the algorithm is evaluated by running time and identification accuracy.The results show that the two algorithms can meet the requirements in recognition accuracy,but the differential evolution algorithm USES less time in the operation process of the algorithm,and the running time of each algorithm fluctuates less,indicating that the differential evolution algorithm has better convergence speed and stability.