Experimental method for modal parameter identification of civil engineering structures based on improved wavelet transform

Abstract

With the continuous acceleration of the global construction industry, many structural infrastructure structures in China have been put into use for decades. They are very prone to damage due to fatigue. The modal parameter identification of civil engineering construction can evaluate the safety status of infrastructure structures. In view of this, an identification system of civil engineering structure modal parameters is proposed based on improved wavelet transform. In the process, the mode shape was chosen as the method of wavelet transform. The data was discretized by selecting the actual data of a high-speed railway station combined with sensors and wavelet transform. Finally, the correct identification of the modal parameters of civil engineering structures is realized. The data shows that under normal conditions where there is only white noise interference, the waveform of the structure is relatively stable, and the amplitude fluctuation is in the [-2,3] interval. At the same time, the average amplitude of the structure is in the [2.2, −1.5] interval under normal conditions. In addition, the positive and negative extreme points are 3.7 and −2.3, respectively. This indicates that the structure amplitude fluctuation is in a dynamic and stable state under normal circumstances. The optimized wavelet transform method identifies a total of four orders in the first six natural frequencies. The minimum error is 0.11%, the maximum error is 1.50%, and the average error of the first four natural frequencies is 0.578%. In addition, based on the comparison of theoretical and identification values of longitudinal vibration shapes, the proposed method can successfully detect abnormal values at the 10th and 18th nodes. From the above results, it shows that the wavelet transform method has high accuracy and small error in frequency identification. It meets the requirements of identifying the natural frequency parameters of the structure. From the calculation, the method proposed in the experiment can realize the real-time health detection of the structure, which is of great significance.