Test Study On Wavelet Ridge Extraction And Modal Parameter Identification Based On Improved Crazy Climber Algorithm

In recent years,with the continuous development of civil engineering structural design theory and technology,the function,scale and shape of the structure have become more diverse and complex.Therefore,the health monitoring and safety assessment of engineering structures are attracting more and more attention.Modal parameter identification is a key research issue in structural health monitoring.In this paper,based on continuous wavelet transform,the wavelet ridge extraction and application problems in structural modal parameter identification are studied.The amplitude and phase information contained in the wavelet ridge line is an important factor in the identification of structural modal parameters.Whether the extracted ridge line accurately affects the accuracy of parameter identification.Crazy Climber algorithm is currently used more extraction algorithm,but the operation rate is slower,which is not conducive to practical application.This paper improves based on this deficiency,and applies the improved ridge line extracted by Crazy Climber algorithm to wavelet theory on modal parameter identification.In this paper,the problem of multi-ridge extraction in continuous wavelet transform structure parameter identification is studied in depth.The main research contents are as follows:(1)The principle and implementation process of continuous wavelet transform are expounded.The theoretical research on signal wavelet ridge concept and wavelet ridge extraction method is carried out.The phase method,modulus maxima method and Crazy Climber algorithm in the ridge line extraction method are compared.(2)Aiming at the problem of long time and low efficiency of the Crazy Climber algorithm,an improved Crazy Climber ridge extraction algorithm for normalizing the cooling is proposed.Firstly,using the rate of temperature change to affect the speed of each point gathering near the ridge line,the simulated normalizing cooling algorithm is used to track the wavelet ridge points in the time-frequency plane,and each random point is quickly gathered on the ridge line.Then,it is applied to the single-degree-of-freedom and multi-degree-of-freedom structure system.By comparing the computational efficiency and recognition accuracy of the improved Crazy Climber algorithm and the traditional Crazy Climber algorithm,the results show that the improved algorithm is basically unchanged.Under the premise,the operation rate is greatly improved.Finally,several factors affecting accuracy,such as mother wavelet parameter selection,metric threshold setting,and ridge line length setting,are discussed,and the principle of parameter setting is proposed to improve the accuracy of wavelet ridge extraction.(3)Combining the improved Crazy Climber algorithm with the wavelet theory modal parameter identification,the improved algorithm can accurately identify the correctness of the modal parameters of the structural system.Firstly,the modal parameter identification process formula based on the improved Crazy Climber algorithm is derived.The logarithm of the wavelet coefficient of the wavelet ridge and the phase information are fitted to the curve,and the system modal parameters are calculated from the slope of the curve.Then,the multi-degree-of-freedom vibration system is numerically simulated to verify the correctness of the improved algorithm.Finally,the multi-layer steel frame model is used to test the modal parameter identification test.The measured response data is output,and the modal parameters obtained by finite element simulation,improved Crazy Climber algorithm and SSI method are compared.The test results show that the improvement is improved.The natural frequency identification result of the algorithm is better than the SSI method.