BPF-WT combined filtering method for indirect identification of bridge dynamic characteristics

Abstract

The indirect measurement method of extracting the dynamic characteristics of the bridge structure from the dynamic response of passing vehicles is a new method for bridge structure health monitoring, which has the advantage of cost-effectiveness. However, how to improve the accuracy of the indirect measurement method to identify the dynamic characteristics of the bridge is the current research focus. In this paper, combining the characteristics and advantages of bandpass filter (BPF) and wavelet transform (WT), a simple and effective joint filtering method BPF-WT is proposed for indirect identification of bridge dynamic characteristics. This method can eliminate the peak caused by the known vehicle frequency, and realize the layer-by-layer decomposition and reconstruction of the signal through the advantage of WT, to achieve the purpose of automatically executing the packet filtering task. This paper takes a simple trolley as an example to conduct a scaled test study to identify the dynamic characteristics of a bridge and proves that the BPF-WT joint filtering method can perform bridge dynamics under three conditions: different vehicle weights, different driving speeds, and driving on bridge decks with different roughness. Characteristic identification results and their influence rules. According to the research findings, the method of combined filtering suggested in this paper can efficiently filter out the noise produced by the vehicle’s frequency and the surroundings. This method can also distinctly detect the initial three frequencies of the bridge. Therefore, it demonstrates the practicality and efficiency of the BPF-WT combined filtering technique, which has a broad range of potential applications.