| With the continuous development of bridge engineering construction level,the construction scale of new structures and new materials,such as long-span bridges and flexible bridges,has also been significantly improved,and they have brought great convenience to human's production and life.However,during the operation of bridge,due to the joint action of long time environment and load,its ability to resist external interference is weakening.Once its safety state crosses the safety range,it will easily lead to disastrous consequences,resulting in a large number of casualties and property losses.Therefore,the monitoring of the deformation of the bridge is of great significance to the safe operation of the bridge.GNSS has been more and more widely used in the field of bridge deformation monitoring because of its advantages of all-weather,high precision,automation and so on.Through the bridge deformation monitoring system based on GNSS,We can not only monitor the bridge spatial displacement of in real time,determine the bridge deformation information and geometric alignment,but also record the time,space and frequency information related to the displacement,so it can give better serve to the later data analysis.However,the bridge deformation monitoring based on GNSS is easily interfered by various internal and external environmental factors,resulting in lots of error information such as noise and gross error in the original bridge deformation monitoring data,so the real-time high accuracy and high reliability display of bridge deformation information can't be realized,which leads bridge managers to misidentify the bridge disaster early warning.At the same time,the direct storage of real-time bridge deformation monitoring dynamic data that contains a large amount of error information brings great challenges to the post analysis of data.Therefore,this paper takes a bridge deformation monitoring based on GNSS as the engineering background,combined with the time varying and dynamic characteristics of bridge deformation monitoring dynamic data sequence based on GNSS,the filtering method of bridge deformation monitoring dynamic data based on GNSS is studied in depth.The main contents are as follows:(1)On the basis of the principle of GNSS positioning and the principle of bridge deformation monitoring based on GNSS,the source of the error and its distribution characteristics of the dynamic data of the bridge deformation monitoring based on GNSS are analyzed.(2)Based on the dynamic and real-time requirement of dynamic data filtering for bridge GNSS deformation monitoring,by comparing and analyzing a variety of commonly used dynamic data filtering methods for bridge deformation monitoring,the feasibility of Kalman filter as the core method in bridge deformation monitoring dynamic data filtering based on GNSS is demonstrated.(3)Considering the effect of colored noise and gross errors in the dynamic data of the actual bridge deformation monitoring based on GNSS,on the basis of measurement matrix augmentation-Kalman filter(select the AR model of first class as the colored noise model),improving the measurement matrix augmentation-Kalman filter by the principle of gross error identification and correction of the first order difference method,the improved measurement matrix augmentation-Kalman filter is thus formed.The improved method can not only deal with the diversity of noise,but also deal with the diversity of gross errors(discrete state and regional state).And the effectiveness of the improved algorithm is verified by comparing with Kalman filter and measurement matrix augmentation-Kalman filter through the simulation experiment.(4)Through actual engineering case application,the actual engineering data is processed and analyzed,and the rationality and superiority of the improved measurement matrix augmentation-Kalman filter method in the bridge deformation monitoring dynamic data filtering processing based on GNSS is verified. |
PDF Full Text Request