A Comparison Model for Mine Subsidence Estimation Based on Bilinear Variations and Multi-learning Machine Algorithms

The surface subsidence caused by coal mining is an extremely complicated process, and the influencing factors such as geological conditions and physical and mechanical properties of rock and soil in the mining area can not be clearly described or inferred, so it is difficult to accurately predict or estimate the surface subsidence caused by coal mining. In order to promote the intelligence of coal mine safety management information system (MIS) and innovate the application of geographic information system (GIS) in coal mine safety management, based on bilinear change and multi-learning machine algorithm, this article establishes a new system for predicting surface subsidence and deformation in horizontal coal seam mining. Because of the complexity and uncertainty of surface subsidence, based on the decision of rock movement data and the precision comparison of various machine algorithms, this study puts forward a comparative analysis method of maximum subsidence estimation. The simulation results show that the urban mine subsidence estimation model in this article has obvious advantages over the comparison method, and the accuracy rate is over 96%. The algorithm in this article provides a new method for the prediction of surface subsidence and deformation in mining, promotes the automation and visualization of mine management, and improves the scientific and accurate comprehensive evaluation of mine environment.