Abstract:
In view of the problems that the traditional monitoring technology cannot monitor the mining surface subsidence for a long time and the existing prediction model relies too much on the subsidence data and has a single model, a mining surface subsidence monitoring and prediction model based on Small Baseline Subsets Interferometric Synthetic Aperture Radar (SBAS-InSAR) and Particle Swarm Optimization -Back Propagation (PSO-BP) neural network algorithm is proposed. Firstly, SBAS-InSAR technology was used to obtain the monitoring values of the surface subsidence in the mining area. Then, the influencing factors of the surface subsidence in the mining area and the obtained monitoring values were selected to construct the PSO-BP prediction model from the perspective of multiple factors. Finally, the effectiveness and rationality of the method were analyzed. The experimental results show that using SBAS-InSAR can effectively monitor the long time surface subsidence in the mining area. With the increase of training samples, the residual difference between PSO-BP predicted value and SBAS-InSAR subsidence value decreases gradually, the algorithm convergence iteration speeds up, and the mean square error decreases. Compared with the existing monitoring methods and prediction models, the advantages of SBAS-InSAR in the monitoring of long-term surface subsidence in mining area and the validity and rationality of PSO-BP model in the prediction of surface subsidence in mining area are proved. This method can be used as an effective means for monitoring and forecasting long-term surface subsidence in mining area.