The Application Of GPS In Deformation Monitoring Of Bridges

During the design phase of the bridge, certain loads are pre-assumed, such as max traffic load and max wind load. However, in reality, these loads are not always in according to the pre-assumed one, even exceed the pre-assumed one. Therefore it is very important to monitor the health of such a structure, in order to keep a safe running. At present, the traditional tools used to monitor structures are displacement sensors, accelerometers, tiltmeters, laser interferometers, total stations and precise leveling. To a certain extent, these tools have their merits, however, they also exit some shortcomings. In recent years, with the development of GPS, especially the RTK, which now allows subcentimeter accuracy to be achieved at rate of up to 20Hz, it makes such a system be capable of detecting deflection of the long bridge. Furthermore, GPS has the advantage of long-term stability, full automation in data collection, all weather observation and time synchronization in application of bridge monitoring. Based on the theme project of the Ministry of Construction - "Health monitoring of large bridges and system study" and according to the GPS based dynamic trial of nanpu bridge, some crucial problems in GPS based dynamic monitoring are investigated in this paper.Firstly, the GPS based dynamic trial of nanpu bridge is introduced, including trial method, sampling rate, scheme of monitored stations, data process solution, establishment of the bridge coordinate system, coordinate transformation and data quality control. Secondly, the single epoch positioning algorithms of the dual frequency GPS are discussed, then three of single epoch algorithms are given, and the feasibility and practicability of each algorithm is demonstrated by using three practical examples. Meanwhile the defects of two algorithms are analyzed and ameliorated in this part. Thirdly, the multipath effect which is the main error source in GPS applications is investigated, and the result demonstrates that the multipath error has stronger repeatability on successive days. Based on this characteristic of multipath, the adaptive filtering method is used to mitigate the effect of multipath. On the other hand, by using the wavelet decomposition and reconstruction, the structure vibration signal is separated from measure noise in the observed data series. Finally, the data quality and the static and dynamic component of the bridge deformation are analyzed.From the study and analysis of this subject mentioned above, some inspiring achievements are obtained. The feasibility of GPS used in bridge monitoring applications is verified further, and also this research provides certain assistance to the topic of GPS on-line monitoring system.