Wavelet Particle Swarm Optimization Algorithm For Damage Identification Of Frame Structures

In practical engineering,the service life of the general structure can be up to fifty years or more,in the life cycle,due to the coupling effect of various uncertainties,the cumulative damage is inevitable.If the structure is not detected and maintained in time,structural damage can be accumulated to a certain extent,which will result in the loss of its function,and even lead to catastrophic events.Because of this,how to determine the location and damage degree of structure have great theoretical significance and engineering application value.Wavelet analysis is considered to be an internationally recognized time-frequency analysis tool,it can observe the whole face of the signal and can analyze the various details of the signal,very suitable for identifying the subtle difference between the positive and negative signals.Therefore,wavelet analysis can be used to identify the damage location of the structure.Particle swarm optimization algorithm is a kind of iterative optimization method,which originated from the method of birds looking for food.It is through the unity and cooperation among the particles to find the global optimum in the complex space.Because the particle swarm optimization algorithm has the advantages of simple,easy to implement,strong global search ability,fast convergence speed and less adjustable parameters,it has good effect on the damage identification of the structure.In this paper,the concept of wavelet particle swarm optimization(PSO)is proposed for the first time by combining the two methods.In the process of structural damage identification by wavelet particle swarm optimization,the structural model of damage must be established by finite element analysis software,and the structural modal parameters can be obtained,then the wavelet coefficients can be obtained by choosing DB wavelet to do the wavelet transform of the modal parameters,the mutation position in the wavelet coefficient graph corresponds to the damage location of the frame structure.Before calculating the degree of structural damage,In this paper,the fitness function of particle swarm optimization(PSO)is established based on the frequency and mode shapes of the structure,and a particle swarm optimization program is programmed,in the process of calculating the damage degree of the structure,the particle swarm optimization is used to search the minimum fitness function.When the particle swarm optimization algorithm is used to optimize the calculation of the required solution several times,the output of the program is the damage degree of the structure.To verify the effectiveness of the proposed method,wavelet particle swarm optimization(PSO)and wavelet neural network(ANN)are adopted to identify the damage of a story frame structure.In the two methods,the damage location of the structure is determined by wavelet analysis.Then the damage degree of the frame is calculated by particle swarm optimization(PSO)and neural network algorithm respectively.After comparing the identification data of the two methods,it is found that the wavelet particle swarm optimization algorithm is more accurate in identifying the degree of structural damage.This paper use wavelet particle swarm optimization algorithm to identify multiple damage cases across the two layer frame structure,two span two story frame structure,and the numerical simulation results show the accuracy of this method.It shows that this method has reference value for damage identification of Engineering structures.