Data Processing And Analysis For Monitoring Landslide Deformation Based On Kalman Filtering

Landslide is one of the geological disaster that influence the security of human's life and property. It has become a important developing direction that to monitor and forecast the landslide by using different mathematical model currently. On the basis of the predecessors' experience, the model of kalman filtering with nine parameters is applicated to monitor and predict the deformation of the landslide. Based on the theory,the software was developed which can process and analyse the GPS monitoring data of the landslide. the main research are the following aspects:The status quo is summarized, include the theoretical studies and the technical development of landslide deformation monitoring. the research and the application development status of the kalman filter model is also given.Introduced The accuracy design,the net form design,the benchmarking design and the observation design of GPS monitoring network. The GPS monitoring network data processing methods are classified into static data processing and dynamic data processing. Then briefly introduced the mathematical model of static data processing and it's quality evaluation.and also the deformation analysis method of GPS deformation monitoring network is given.Deduced the kalman filtering model of the landslide dynamic monitoring and forecast. The GPS data processing model includes nine state parameters:the 3d position parameter, the 3d velocity parameters and the 3d acceleration parameters. the baseline vector was selected as the observed quantity and the acceleration change values was selected as the dynamic noise.Based on the above theories, the software, GPS landslide monitoring data processing software, was designed started with baseline vector.The landslide monitoring data of three gorges reservoir area was processed and analysed by using this software, and draw some related conclusions.