Monitoring and predicting railway subsidence using InSAR and time series prediction techniques
Monitoring and predicting railway subsidence using InSAR and time series prediction techniques
复制标题
DOI：
--
复制
发表时间：
2015-12
期刊：
影响因子：
--
通讯作者：
Ziyi Yang
中科院分区：
文献类型：
--
作者：
Ziyi Yang
铁路能力的提高导致了更重的轴载荷和更高的运行速度，这增加了轨道上的动态载荷。因此，铁路沉降已成为影响铁路良好性能和铁路安全运营的一大威胁。本文通过对铁路沉降的监测与预测，为铁路性能评价提供了一种途径。铁路沉降监测选择了InSAR技术，该技术能够监测大范围、长时间的铁路沉降。利用InSAR获取的时间序列变形记录，建立沉降预测模型，预测铁路沉降的未来趋势。本文采用了三种时间序列预测模型：ARMA模型、神经网络模型和灰色模型。通过对高铁1号线沉降监测与预测的两个实例研究，对高铁1号线的沉降性能进行了评价。实例分析表明，高铁1号线经过10年的运行，除部分有潜在沉降的区域外，未发生大规模沉降，运行状态稳定。此外，神经网络模型对HS1的沉降预测效果最好。
Improvements in railway capabilities have resulted in heavier axle loads and higher speed operations, which increase the dynamic loads on the track. As a result, railway subsidence has become a threat to good railway performance and safe railway operation. The author of this thesis provides an approach for railway performance assessment through the monitoring and prediction of railway subsidence. The InSAR technique, which is able to monitor railway subsidence over a large area and long time period, was selected for railway subsidence monitoring. Future trends of railway subsidence should also be predicted using subsidence prediction models based on the time series deformation records obtained by InSAR. Three time series prediction models, which are the ARMA model, a neural network model and the grey model, are adopted in this thesis. Two case studies which monitor and predict the subsidence of the HS1 route were carried out to assess the performance of HS1. The case studies demonstrate that except for some areas with potential subsidence, no large scale subsidence has occurred on HS1 and the line is still stable after its 10 years' operation. In addition, the neural network model has the best performance in predicting the subsidence of HS1.