Adaptive Kalman Filter Based On Wavelet Analysis Applying In Metro Deformation Monitoring

As the underground traffic, subway is the product of the age. Now, city population increase rapidly and the vehicle is growing with each passing day. In order to solve the traffic congestion and offer more convenient environment for people, subway is put forward as one short cut. However, subway offers many conveniences for people, meanwhile it still exists many questions which need people to solve. It will break the original stress balance when it passes through the lines, buildings, and deep excavations. So it will lead to serious consequence if those elements can’t be controlled in time. Therefore, it is very essential to know the whole condition and survey of many-sided key area.At present, the local monitoring of metro tunnel is main thing. There are many measuring method, measuring instruments and sorting method for each monitoring region. This thesis is aimed at the monitoring of the tunnel section. It is mainly through using the method of measuring robot intelligence, and automation of data acquisition. Then the data is managed by wavelet variance method and the adaptive Kalman filter. And then the researchers can forecast the form variables. There are some main research contents.(1)This thesis expounds the deformation monitoring technology of the subway tunnel and its method of the data processing. According to the specific project requirements, the researchers put forward the data acquisition of measuring robot and methods of processing data;(2)This thesis introduces the robot’s basic structure and function. The researchers master the internal structure of TS30, program development process and monitoring system, by learning and understanding its system composition of GeoBasic, ASCⅡ, GeoCOM;(3)Instrument operation through simulation learning process, and conduct in-depth research, including the accuracy of the analysis, differential correction, the reference point of the polar coordinate stability analysis method, and the measurement results using traditional methods of calculating the adjustment as a result of the processing behind comparative data;(4)Learn regression analysis, time series, gray models commonly used monitoring data processing methods, and through broader application of wavelet de-noising, a comprehensive study of Kalman filtering theory, find the "wavelet pretreatment for monitoring subway + variance compensation adaptive Kalman filter " model. Both methods achieve filtering through matlab programming;Finally, through the knowledge what we have mentioned before, and pavilion monitor program based on building model by comparative analysis and verification in Shen Zhen, we can know that this model not only can be used in subway variable smaller projects, and also can be used for landslide, version of the rocks which is moved from variables.