Mining Subsidence Monitoring and Prediction Based on UAV Tilt Photogrammetry Technology and SVR Algorithm
In order to realize high-precision monitoring and prediction of mining subsidence,UAV tilt photography tech-nology was combined with Support Vector Regression Algorithm(SVR)to monitor and predict the subsidence of Liujia Coal mining Area in Chifeng City.The predicted results are compared with the measured results of GPS.The results show that:the subsidence measured by GPS in the mining area in 10 days is 0.2~15.8 mm,the average subsidence in the east,west and north regions is 51.2,31.5,46.6 and 130.8 mm,respectively,and the subsidence in the north side is 2.55,4.15 and 2.81 times that in the east,west and south,respectively.The subsidence in the north side of the mining area is a serious area.It is necessary to focus on monitoring and develop reasonable and effective prevention and control measures.The cumulative average settlement in the east,west,north and south regions of the mining area predicted based on human-machine oblique photography technology and SVR algorithm is 49.4,31.5,45.8 and 134.6 mm,respectively.The relative error values measured by the two measurement methods at all measuring points are within 5%,reflecting the high prediction accuracy of the proposed method in this study.It provides a reference for realizing high-precision monitoring and prediction of mining subsidence.
NETL
NSTL
万方数据
