Characteristics Analysis Of Health Monitoring Data Of Cable Stayed Bridge And Recognition Of Heavy Vehicle Load

With the growing trend towards complex and large-scale of bridge structures, manual detection using traditional methods cannot meet the security needs of large-span bridges, to build a bridge health monitoring system to ensure the safe operation of bridge has become an important means. The value of scientific guidance Bridge Health Monitoring System is decided by comprehensive background data collection capabilities and good reception performance data, yet the data performance capabilities of most health monitoring system is much weaker than its data collection capabilities currently. The current health monitoring system is able to collect huge amounts of data, but the study of the data is not so deep enough that massive monitoring data cannot be properly treated and utilized. These data have actually become the bottleneck of restricting the significance of health monitoring system. Therefore, the study aiming at the characteristics of massive health monitoring data is of great significance to data processing and mining. The main research contents of this paper are as follows:(1) Strain and temperature data of the stable operational phase collected from North Branch of the Huangpu Bridge cable-stayed bridge were pretreated, then the characteristics of the processed monitoring data are analyzed. The measured temperature field of steel box girder are analyzed in detail, and the relationship between strain and temperature is discussed.(2) Through the analysis of the whole model of the cable-stayed bridge, the internal stress of the key beam section is obtained. Apply these internal stress to the finite element model of the partial beam and impose temperature field to the bottom plate of the box girder. Combined with the details layer information from the wavelet analysis, the positive correlation between the strain and temperature of the steel box girder bridge can be verified.(3) Based on a large number of strain data collected by the monitoring system, an attempt is made to identify the abnormal events of the overweight vehicle load. Based on wavelet analysis, this paper identifies the location of the mutation. And then calculate the value of the strain caused by the catastrophe event. Through the comparison with the strain variation of the overload vehicle through finite element simulation, whether the mutation position is a heavy truck load anomaly event can then be determined. What is more, through the recognition of the results of statistical analysis, to explore the safety of each section of the cable stayed bridge.