Research On Structural Damage Identification And Signal Denoising Based On Wavelet-firefly Algorithm

Civil engineering structures in service,affected by various factors,will produce structural damage.With the continuous accumulation of damage,safety accidents may occur when the accumulation of damage reaches a critical value.Identification of structural damage requires mastery of the measured response signals of the structure.In the process of obtaining the measured response signals of the structure,the signals are usually interfered by environmental noise,which affects the recognition effect of the structural response signals and reduces the credibility of damage detection.Therefore,the research on denoising method of structural response signals has important theoretical significance and engineering application value.The main research contents of this paper are as follows:Firstly,based on the theory of wavelet analysis and firefly algorithm,a wavelet-firefly algorithm(WGSO)for structural damage identification is established,which can effectively identify the location and extent of structural damage.It not only reduces the solving dimension of the algorithm,but also makes up for the disadvantage that the wavelet analysis is difficult to identify the damage degree.Based on the principle of wavelet singularity,the method uses wavelet transform to the structural modal data,and then determines the damage location of the structure according to the modulus maximum points in the coefficient graph.On the basis of the known damage location,the objective function was established based on the natural frequency and mode shapes before and after the damage,and the firefly algorithm(GSO)was programmed with Matlab to optimize the inversion of the objective function,and the changes of the physical parameters before and after the damage were obtained,so as to identify the damage degree.Through numerical simulation of fixed beams under different damage conditions,the identification accuracy and running time of genetic algorithm(GA)are compared and analyzed.The results show that the firefly algorithm has higher stability and convergence speed.Secondly,based on the measured vibration signals of damaged beams,this paper studies the denoising problem of structural damage identification,proposes a denoising method based on the wavelet threshold function,and gives the comprehensive evaluation function of denoising.First,the signal is decomposed by multilayer wavelet,and the signal component of each layer is quantified by threshold value,and then the processed signal component is reconstructed to obtain the de-noised signal.Using MATLAB to simulate the vibration signal denoising process,through the analysis under the different layer number of wavelet basis and decomposition of the vibration signal denoising effect,it is concluded that the vibration signal denoising of optimal wavelet base and optimal decomposition layers,and compared the traditional thresholding function and adaptive thresholding function denoising effect,verify the validity of the proposed denoising method in this paper.Finally,this paper completes the damage identification test of the fixed beam,and studies the denoising and damage identification of the measured response signals of the structure.The experimental results show that the wavelet-firefly algorithm can effectively identify the damage location and damage degree of the structure,and the wavelet threshold denoising method can effectively improve the accuracy of structural damage identification.