Data Processing And Forecasting Of Mining Subsidence Movement Deformation Based On Adaptive Kalman Filter

With the development of China's national economy and the progress of society,China's energy structure is changing towards energy conservation.However,the status of coal in China's basic energy is still unshakable,and it still occupies an important and irreplaceable role in the main energy structure.Status.The surface subsidence caused by the high-intensity mining of coal resources has destroyed the cultivated land and brought great influence to agricultural production.The coal gangue pollutes the ecological environment of the mining area and pollutes the water source,causing irreversible impact on the ecological environment.In order to protect the ecological environment of the mining area and carry out restoration and reconstruction,it is especially important to obtain timely and accurate and reliable information on the surface deformation and deformation of the mining area and accurately predict the deformation information.The Kalman filter model describes the dynamic process of the system by establishing the state equation and the measurement equation for the mining surface deformation.It requires the mathematical model and noise prior knowledge of the subsidence surface movement deformation system,but the mining subsidence process has some stages.It is very unstable,which makes it cause a large deviation in the data processing and forecast analysis of this stage,which will significantly reduce the accuracy and reliability.Aiming at the problem of low precision of data processing and forecasting analysis of mining subsidence surface,this paper proposes several adaptive Kalman filter models for mining subsidence surface movement deformation monitoring system,which is combined with the 1222(1)working face mining of Zhuji East Mine.Several adaptive Kalman filter models are proposed to investigate the processing and analysis of the monitoring data of mining subsidence surface deformation.The work and results of this paper are as follows:The filtering model proposed in this paper is used to filter the surface deformation and deformation data of the mining area.It is concluded that the residual stability of each period of the adaptive Kalman filter is high,but the residual values are positive or negative,and there may be systematic deviation;variance and residual error in adaptive Kalman filter residuals The stability is inferior to the variance component adaptive Kalman filter;the error of the variance component adaptive residual is the smallest and the residual of each period is relatively stable,which can significantly reduce the large residual filter residual of Kalman filter,and the effect is obvious.The three-dimensional spatial position coordinate sequence of the GNSS CORS surface mobile automation real-time monitoring station is forecasted and analyzed by the filtering proposed in this paper.The variance component Kalman filter is used to predict that the residual stability is high and the intensity is high.The overall residual error is reduced by 60%in the X coordinate direction,52%in the Y coordinate direction,and 69%in the H direction.The difference dispersion degree distribution is more uniform;the plane position prediction accuracy and elevation prediction accuracy are significantly better than other types of filtering.The square position coordinate and Kalman filter in this paper are used to filter the plane position coordinates and elevation of the single base station CORS RTK.After filtering,the plane accuracy is increased by about 50%,the elevation accuracy is increased by about 70%,the plane and elevation accuracy are improved,and the elevation accuracy is more obvious.The difference between the measured value and the leveling of the filtered elevation is significantly reduced.And the reliability is improved,which can basically meet the accuracy requirement of the elevation of the mining subsidence surface displacement deformation parameter.Figure [53] table [20] reference [85]...