基于SBAS-InSAR和GM-SVR的居民区形变监测与预测

采用合成孔径雷达时序分析方法，利用2016-12～2017-05(12 d为一个周期)连续13景哨兵卫星雷达影像对淮南矿区内的居民区杨聚庄进行形变监测。根据矿区形变特征，提出一种基于灰色支持向量机（GM-SVR）的组合预测模型对矿区形变进行预测，并与传统的单一灰色模型和支持向量机预测模型进行对比。结果表明，时序InSAR技术和GM-SVR模型的结合，可以实现对矿区形变的快速监测和灾害预防，为矿区灾害监测与预警提供了一种可靠手段。

Deformation Monitoring and Prediction of Residential Areas Based on SBAS-InSAR and GM-SVR

In this paper, we use the synthetic aperture radar time series analysis method to analyze the deformation of Yangjuzhuang, a residential area in the Huainan mining area. We use 13 consecutive Sentinel satellite radar images from December 2016 to May 2017 (one cycle of 12 d). According to the deformation characteristics of the mining area, we propose a combined prediction model of grey support vector machine (GM-SVR) to predict the deformation of the mining area. The results of this model are compared with the results of the traditional single gray model and the support vector machine prediction model. The results show that the combination of InSAR time series analysis technology and the GM-SVR model can realize the rapid deformation monitoring and disaster prevention of the mining area, and provide a new method for the monitoring and early warning of mining disasters.