The Application Of Robust Kalman Filtering In The Hight-Speed Railway Deformation Monitoring

Nowadays, the high-speed railway construction is moving forward at an unprecedented rate in China, It is a large complex and sophisticated projects, and demanding precise construction of all aspects. As the track basic projects-the railway underline engineering in the high-speed railway, its settlement deformation control is the first condition to provide high-speed, high smoothness and high stability track. At the same time, due to the complexity of geological conditions in China, as a long linear high-speed railway, to control the settlement deformation of underline engineerings is critically important, therefore, the security monitoring, data processing and the evaluation and analysis are guarantees to ensure the safety of the construction and operation. It must select the effective monitoring tools and data processing methods. In this thesis, Base on the high speed-railway underline engineerings construction characteristics and it exists problems, studying the application of Robust Kalman Filtering in the high-speed railway settlement deformation monitoring, and researching and analysising in conjunction with a high-speed railway monitoring data. The main contents are as follows:In this thesis, overviewing of the characteristics of the high-speed railway underline engineering, and the deformation monitoring technology and its significance are introduced.In contrast to commonly monitor data processing methods, and studying the method of establishing Kalman Filtering model and the confirmation of initial value. Through example, it is verified that it has advantages in the monitoring data processing. As the ordinary Kalman Filtering model can not effectively resist the impact of gross error. In this thesis, bases on robust estimation, deriving the recursive equation, and applying to the data processing of monitoring point. It is verified the Robust Kalman Filtering can effectively identify and exclude gross error.Combining the dynamic leveling network monitoring data of a high-speed railway and processing with the Robust Kalman Filter model, they are anatomists between the filtering value, forecasting value and difference value, so within the allowable range, the method can be used for data processing and be real-time monitored to the settlement deformation.The above studies have shown that it can meet the engineering needs using Robust Kalman Filtering model for high-speed railway deformation monitoring data processing, so it has a good practical value.