The Application Of Wavelet Analysis In The Multi-source Data Fusion Of Bridge Health Monitoring

With the development of the bridge, more and more bridges accident also arise. Thisgreat threat to the safety of the people. So it particularly important that maintaining thenormal work of the bridge and working well to bridge diseases warning work. To this kind ofsituation, bridge health testing getting people’s attention, more and more. The final result ofthe traditional test methods is always not ideal for a variety of reasons. And it is not soobvious in identifying the damage effect. In order to be able to find a kind of effective methodto recognize and test the damage of the bridge, this paper doing a series of experiments andstudied. The main content of the concrete includes three aspects. First, doing time historyanalysis to the bridge under the dynamic loading action; the key point is the dynamic responseof the displacement and acceleration under the dynamic loading action. Extracting therelevant time history analysis of data from it for the data sources of the following analysis.Second, importing the time history analysis of data of displacement and acceleration to‘matlab’ which from ‘midas’, then converting the data types to what we need. Undering this,we make wavelet decomposition and reconstruction to the two sets of data with the waveletfunction which is in wavelet toolbox. Because of this, we can get the preliminary conclusions.The third we make wavelet fusion using the time history data of displacement andacceleration. Then we can do decomposition and reconstruction to the data which is combinedby the wavelet function. Except this, we use the different wavelet function doingdecomposition and reconstruction to the same data. The end, we make analysis respectively indifferent speed and load situations. The analysis results show that the wavelet fusion canjudge the damage location. Comparing the different analysis results and then we can get thefinal conclusion. Through the analysis and comparison of the data，it can be a very good proofthat wavelet multi-source data fusion can recognize the bridge damage exactly. And thisprovides a reference method for the bridge health detection in future.