摘要
针对盘营铁路专线、哈大铁路专线沿线沉降监测研究较少,采用InSAR技术获取了研究区地表形变信息,还对其进行了相关分析。用SBAS-InSAR对35景Sentinel-1A SAR数据进行处理,获取VV、VH极化下的年均沉降速率及沉降序列;以年均沉降速率为研究对象,进行沿线沉降特征分析及交叉验证;利用小波变换对沉降序列降噪处理,用改进BP神经网络对降噪后沉降序列预测分析。研究结果表明,研究区内高速铁路沿线共监测出6个明显沉降区域,最大沉降速率达50 mm/a;两种极化年均沉降速率具有较高的一致性,降噪处理后的沉降序列更加平滑;改进BP神经网络具有较高的收敛速度,其预测精度有较大提高。
In view of the lack of research on settlement monitoring along Panying railway line and Hada railway line,this paper used InSAR technology to obtain the surface deformation information of the study area,and also carried on the correlation analysis.Firstly,35 scenes of Sentinel-1 A SAR data were processed by SBAS-InSAR to obtain the average annual settlement rate and settlement sequence under the polarization of VV and VH.Secondly,taking the average annual settlement rate as the research object,the settlement characteristics along the line were analyzed and cross verified.Finally,the wavelet transform was used to denoise the settlement sequence,and the improved BP neural network was used to predict and analyze the denoising settlement sequence.The research results showed that there were six obvious settlement areas along the high-speed railway in the study area,the maximum settlement rate was 50 mm/a.The average annual settlement rates of the two polarizations had high consistency,and the settlement sequence after noise reduction was smoother.The improved BP neural network had higher convergence speed,and its prediction accuracy was greatly improved.
作者
游洪
米鸿燕
李勇发
王志红
刘岚
熊鹏
YOU Hong;MI Hongyan;LI Yongfa;WANG Zhihong;LIU Lan;XIONG Peng(School of Land and Resources Engineering,Kunming University of Science and Technology,Kunmming 650093,China;Guizhou University of Engineering Science,Bijie,Guizhou 551700,China;Tonghai Real Estate Registration Center,Tonghai,Yunnan 652704,China)
出处
《测绘科学》
CSCD
北大核心
2021年第7期67-75,共9页
Science of Surveying and Mapping
基金
国家自然科学基金项目(51574242)
贵州省教育厅自然科学研究项目([2018]071,[2018]405)。
