The Health Monitoring Research Of Single Tower Cable-stayed Bridge With Steel Box On Wavelet Method

Wavelet analysis is known as the "signal microscope", its multi-scale analysis and singular point detection and wavelet packet node energy based on the wavelet transform provide a convenient and efficient tool for the real-time monitoring signal processing. From the current research, this paper firstly analyzes the bridge health monitoring problems and gives a brief introduction on the background of the long-term health monitoring system, then introduces the wavelet analysis theory, and studies the application of wavelet method systematically in the long-term health monitoring of bridges based on the combination of program background and numerical simulation, next expounds the multi-scale analysis and threshold de-noising analysis of measured signal based on wavelet transform theory and applies the wavelet packet energy theory and wavelet singularity detection theory on the damage detection and localization of the single tower cable-stayed bridge with steel box, finally gives the improvement suggestions on the problems existing in the theoretical research applied to the engineering practice. The main contents are as follows:1)This paper firstly combines the wavelet multi-scale analysis and threshold de-noising theory and applies the different characteristics of the true signal and the noise signal on the different scales wavelet transform to achieve the threshold de-noising monitoring signal, then contrasts the amplitude of the measured signal and the de-noising signal, and meanwhile the amplitude of the measured signal and the de-noising signal after using the fast Fourier transform, finally compares the performance results of measured signal on different thresholds rules and discusses the basic rules of the selected threshold condition. When we conduct the multi-scale analysis through the wavelet analysis theory, we only need to import the collected acceleration signal acquisition to the MATLAB software, and prepare the corresponding program that is able to express the damage characteristics. As the result, this position at some scales time will appear the extreme point.2)This paper firstly combines the wavelet decomposition and the wavelet packet energy theory and applies to achieve the damage identification of bridge structures, then uses the Midas Civil software to establish the finite element model of Tonghai Road cable-stayed bridge and begins the time-history analysis, next extracts the acceleration signal of main beam nodes when the bridge suffers a single injury and multiple injuries, finally begins the wavelet packet energy analysis and gets the wavelet packet node energy change values of the intact beam and the cable forces when the bridge encounters a variety of conditions. By comparing the percentage of band energy of the wavelet packet, we can conclude that the wavelet packet energy change values have a high sensitivity on the small cable force changes of the structure. Because this method requires only the acceleration time-history signal of the damage cases, so it is very beneficial to apply on the real-time online monitoring of bridge structures, and can provide important reference suggestion for the timely repair some damaged bridges.3)This paper firstly combines the wavelet transform and the curvature mode theory and applies to locate bridge structures damage, then uses the Midas Civil software to establish the finite element model of Tonghai Road cable-stayed bridge and begins the mode analysis, next extracts the first-order and the second-order displacement mode of the main beam nodes, finally gets its curvature mode by using the central difference formula and begins the wavelet transform for the different curvature mode. By analyzing the first-order and the second-order curvature mode graphics of the stiffness reduction on the main beam for each condition, we can obviously find the injury location. Because this method can not only well identify the minor damage of the main beam element, but also can locate the stay-cables damage by the curvature mode signal from the main beam element, so it is very beneficial to promote the wavelet singularity theory applying on the long-term health monitoring of a variety of different types bridges.