大范围地表沉降时序深度学习预测法

doi:10.11947/j.AGCS.2021.20200038

测绘学报 ›› 2021, Vol. 50 ›› Issue (3): 396-404.doi: 10.11947/j.AGCS.2021.20200038

大范围地表沉降时序深度学习预测法

刘青豪^1,2, 张永红², 邓敏¹, 吴宏安², 康永辉², 魏钜杰²

1. 中南大学地球科学与信息物理学院, 湖南长沙 410083;
2. 中国测绘科学研究院, 北京 100830

收稿日期:2020-02-05 修回日期:2020-08-01 发布日期:2021-03-31
通讯作者: 张永红 E-mail:yhzhang@casm.ac.cn
作者简介:刘青豪(1996-),男,硕士,研究方向为时空数据挖掘,InSAR形变预测。E-mail:185012100@csu.edu.cn
基金资助:
国家自然科学基金（41874014；41730105）；政府间国际科技创新合作重点专项（2017YFE0107100）；中国测绘科学研究院基本科研业务费（AR1938）

Time series prediction method of large-scale surface subsidence based on deep learning

LIU Qinghao^1,2, ZHANG Yonghong², DENG Min¹, WU Hongan², KANG Yonghui², WEI Jujie²

1. College of Earth Science and Information Physics, Central South University, Changsha 410083, China;
2. China Academy of Surveying and Mapping, Beijing 100830, China

Received:2020-02-05 Revised:2020-08-01 Published:2021-03-31
Supported by:
The National Natural Science Foundation of China (Nos. 41874014;41730105);The Key Project of International Science and Technology Innovation Cooperation between Governments (No. 2017YFE0107100);The Basic Scientific Research Business Cost Project of China Academy of Surveying and Mapping Sciences (No. AR1938)

摘要/Abstract

摘要： 地表沉降不仅影响社会经济的可持续发展，还威胁人类的生命安全。高精度的地表沉降预测对人类预防地质灾害具有重要意义。现有的预测方法因模型参数难以获取或相关数据的缺乏而难以得到可靠的预测结果，针对此问题，本文提出一种基于深度学习的地表沉降预测方法。首先采用多主影像相干目标小基线干涉技术MCTSB-InSAR获取大区域高精度地表形变时序反演结果；其次利用循环神经网络作为网络架构，用长短期记忆（LSTM）模型进行地表沉降特征学习；最后采用网格搜索的方法调整模型参数，进而获取最优的模型参数组合方案。实际观测结果显示，相较于现有地表沉降预测方法，本文提出的预测模型平均绝对误差（0.3 mm）至少降低了27.3%，差分沉降量平均预测精度至少提高了8.9%。空间格局分析的结果表明，LSTM模型对于大区域时序形变的短期预测是有效的。

关键词: 地表沉降, 时间序列预测, 深度学习, 长短期记忆, InSAR

Abstract: Surface subsidence not only affects the sustainable development of social economy, but also threatens the safety of human life. High precision prediction of surface subsidence is of great significance for the prevention of geological disasters. However, the existing prediction methods are difficult to obtain reliable prediction results because of the model parameters or the lack of relevant data. For this problem, a method of surface subsidence prediction based on deep learning is proposed. Firstly, the multiple master-image coherent target small-baseline InSAR (MCTSB-InSAR) is used to obtain the inversion results of large area and high precision ground deformation time series. Secondly, the cyclic neural network is used as the network framework, and the long short-term memory (LSTM) model is used to learn the characteristics of ground settlement. Finally, the grid search method is used to adjust the model parameters, then get the optimal combination scheme of model parameters. The actual observation results show that the average absolute error (0.3 mm) of the prediction model proposed in this paper is reduced by 27.3% at least, and the average prediction accuracy of differential settlement is improved by 8.9% at least. The results of spatial pattern analysis show that the LSTM model is effective for the short-term prediction of large-scale time series deformation.

Key words: ground settlement, time series prediction, deep learning, LSTM, InSAR

中图分类号:

P237

刘青豪, 张永红, 邓敏, 吴宏安, 康永辉, 魏钜杰. 大范围地表沉降时序深度学习预测法[J]. 测绘学报, 2021, 50(3): 396-404.

LIU Qinghao, ZHANG Yonghong, DENG Min, WU Hongan, KANG Yonghui, WEI Jujie. Time series prediction method of large-scale surface subsidence based on deep learning[J]. Acta Geodaetica et Cartographica Sinica, 2021, 50(3): 396-404.

参考文献

[1] 苏河源, 胡兆璋. 国外地面沉降研究状况述评[J]. 上海地质, 1980(2):65-77. SU Heyuan, HU Zhaozhang. A review of land subsidence research abroad[J]. Shanghai Geology, 1980(2):65-77.
[2] 任连伟, 周桂林, 顿志林, 等. 采空区建筑地基适宜性及沉降变形计算工程实例分析[J]. 岩土力学, 2018, 39(8):2922-2932, 2940. REN Lianwei, ZHOU Guilin, DUN Zhilin, et al. Case study on suitability and settlement of foundation in goaf site[J]. Rock and Soil Mechanics, 2018, 39(8):2922-2932, 2940.
[3] CHEN Mi, TOMÁ S R, LI Zhenhong, et al. Imaging land subsidence induced by groundwater extraction in Beijing (China) using satellite radar interferometry[J]. Remote Sensing,2016, 8(6):468.
[4] THINH H P, STROKOVA L A. Prediction maps of land subsidence caused by groundwater exploitation in Hanoi, Vietnam[J]. Resource-Efficient Technologies, 2015, 1(2):80-89.
[5] ZHANG Yuanliang, ZHANG Yihu. Land subsidence prediction method of power cables pipe jacking based on the peck theory[J]. Advanced Materials Research, 2013, 634-638:3721-3724.
[6] SHEARER T R. A numerical model to calculate land subsidence, applied at Hangu in China[J].Engineering Geology,1998, 49(2):85-93.
[7] WANG Yue, YANG Guang. Prediction of composite foundation settlement based on multi-variable gray model[J]. Applied Mechanics and Materials, 2014, 580-583:669-673.
[8] LI Peixian, TAN Zhixiang, YAN Lili, et al. Building settlement forecast based on BP neural network[C]//Proceedings of 2011 International Conference on Electric Technology and Civil Engineering.Lushan, China:IEEE, 2011:2024-2027.
[9] GAO Hui, SONG Qichao, HUANG Jun. Subgrade settlement prediction based on least square support vector regession and real-coded quantum evolutionary algorithm[J]. International Journal of Grid and Distributed Computing, 2016, 9(7):83-90.
[10] SHAHIN M A, MAIER H R, JAKSA M B. Settlement prediction of shallow foundations on granular soils using B-spline neurofuzzy models[J]. Computers and Geotechnics, 2003, 30(8):637-647.
[11] 岳荣花. 小波神经网络在沉降预测中的应用研究[D]. 南京:河海大学, 2007. YUE Ronghua. The application of wavelet neural network in settlement prediction[D]. Nanjing:Hohai University, 2007.
[12] WANG Yinghua, XU Chang. Using genetic artificial neural network to model dam monitoring data[C]//Proceedings of 2010 Second International Conference on Computer Modeling and Simulation. Sanya, Hainan, China:IEEE, 2010.
[13] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[C]//Proceedings of the 25th International Conference on Neural Information Processing Systems.Red Hook, NY:Curran Associates Inc., 2012.
[14] ZHOU Pan, LIU Cong, LIU Qingfeng, et al. A cluster-based multiple deep neural networks method for large vocabulary continuous speech recognition[C]//Proceedings of 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.Vancouver, BC, Canada:IEEE, 2013.
[15] SUN Yi, CHEN Yuheng, WANG Xiaogang, et al. Deep Learning Face Representation by Joint Identification-Verification[J]. Proceedings of the 27th International Conference on Neural Information Processing Systems.Cambridge, MA:MIT Press, 2014.
[16] GERS F A, SCHMIDHUBER J, CUMMINS F. Learning to forget:continual prediction with LSTM[J]. Neural Computation, 2000, 12(10):2451-2471.
[17] HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural Computation, 1997, 9(8):1735-1780.
[18] GRAVES A, SCHMIDHUBER J. Framewise phoneme classification with bidirectional LSTM and other neural network architectures[J]. Neural Networks, 2005, 18(5-6):602-610.
[19] 贾澎涛, 何华灿, 刘丽, 等. 时间序列数据挖掘综述[J]. 计算机应用研究, 2007, 24(11):15-18, 29. JIA Pengtao, HE Huacan, LIU Li, et al. Overview of time series data mining[J]. Application Research of Computers, 2007,24(11):15-18, 29.
[20] KINGMA D, BA J. ADAM:a method for stochastic optimization[J]. Computer Science, 2014.
[21] DUCHI J C, HAZAN E, SINGER Y. Adaptive subgradient methods for online learning and stochastic optimization[J]. Journal of Machine Learning Research, 2011, 12(7):257-269.
[22] TIELEMAN T, HINTON G. Lecture 6.5-RMSProp:divide the gradient by a running average of its recent magnitude[J].COURSERA:Neural Networks for Machine Learning,2012, 4(2):26-31.
[23] WEN Quan. Spare parts demand prediction research of navigation marks based on the method of quadratic exponential smoothing[C]//Proceedings of the 7th International Conference on Applied Science,Engineering and Technology (ICASET 2017).[S.l.]:Atlantis Press,2017.
[24] PERCIVAL D B. Spectral analysis for physical applications[M]. Cambridge:Cambridge University Press, 1993.
[25] BOX G E P, JENKINS G M. Time series analysis:forecasting and control[J]. Journal of Time, 2010, 31(3):303-303.
[26] NIKOLOPOULOS K, GOODWIN P, PATELIS A, et al. Forecasting with cue information:a comparison of multiple regression with alternative forecasting approaches[J]. European Journal of Operational Research, 2007, 180(1):354-368.
[27] ZHOU Hongyue, WANG Yunjia, YAN Shiyong, et al. Monitoring of recent ground surface subsidence in the Guangzhou region by the use of InSAR time-series technique with multi-orbit Sentinel-1 TOPS imagery[J]. International Journal of Remote Sensing, 2018, 39(21-22):8113-8128.
[28] 张永红, 刘冰, 吴宏安, 等. 雄安新区2012~2016年地面沉降InSAR监测[J]. 地球科学与环境学报, 2018, 40(5):152-162. ZHANG Yonghong, LIU Bing, WU Hongan, et al. Ground subsidence in Xiong'an New Area from 2012 to 2016 monitored by InSAR technique[J]. Journal of Earth Sciences and Environment, 2018, 40(5):152-162.
[29] 佚名.CH/T 6006-2018《时间序列InSAR地表形变监测数据处理规范》概述[J]. 测绘标准化, 2019, 35(2):24. Anonymous.CH/T 6006-2018"time series InSAR surface deformation monitoring data processing specification" overview[J]. Standardization of Surveying and Mapping, 2019, 35(2):24.
[30] 朱建军, 李志伟, 胡俊. InSAR变形监测方法与研究进展[J]. 测绘学报, 2017, 46(10):1717-1733. DOI:10.11947/j.AGCS.2017.20170350. ZHU Jianjun, LI Zhiwei, HU Jun. Research progress and methods of InSAR for deformation monitoring[J]. Acta Geodaetica et Cartographica Sinica, 2017,46(10):1717-1733. DOI:10.11947/j.AGCS.2017.20170350.

大范围地表沉降时序深度学习预测法

Time series prediction method of large-scale surface subsidence based on deep learning

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	张良培, 何江, 杨倩倩, 肖屹, 袁强强. 数据驱动的多源遥感信息融合研究进展[J]. 测绘学报, 2022, 51(7): 1317-1337.
[2]	许强, 朱星, 李为乐, 董秀军, 戴可人, 蒋亚楠, 陆会燕, 郭晨. “天-空-地”协同滑坡监测技术进展[J]. 测绘学报, 2022, 51(7): 1416-1436.
[3]	李志伟, 许文斌, 胡俊, 冯光财, 杨泽发, 李佳, 张恒, 陈琦, 朱建军, 王琪洁, 赵蓉, 段梦. InSAR部分地学参数反演[J]. 测绘学报, 2022, 51(7): 1458-1475.
[4]	程涛, 张洋, James Haworth. 基于网络和图的时空智能——概念、方法和应用[J]. 测绘学报, 2022, 51(7): 1629-1639.
[5]	龚健雅, 宦麟茜, 郑先伟. 影像解译中的深度学习可解释性分析方法[J]. 测绘学报, 2022, 51(6): 873-884.
[6]	张勤, 赵超英, 陈雪蓉. 多源遥感地质灾害早期识别技术进展与发展趋势[J]. 测绘学报, 2022, 51(6): 885-896.
[7]	龚健雅, 张觅, 胡翔云, 张展, 李彦胜, 姜良存. 智能遥感深度学习框架与模型设计[J]. 测绘学报, 2022, 51(4): 475-487.
[8]	毛文婧, 王卫林, 焦利民, 刘安宝. 基于深度学习的中国连续空间覆盖PM_2.5浓度预报[J]. 测绘学报, 2022, 51(3): 361-372.
[9]	金飞, 官恺, 刘智, 韩佳容, 芮杰, 李庆高. 无监督密集匹配特征提取网络性能分析[J]. 测绘学报, 2022, 51(3): 426-436.
[10]	何直蒙, 丁海勇, 安炳琪. 高分辨率遥感影像建筑物提取的空洞卷积E-Unet算法[J]. 测绘学报, 2022, 51(3): 457-467.
[11]	马张烽, 蒋弥, 李桂华, 黄腾. 空间网络对时序InSAR相位解缠的影响——以Delaunay与Dijkstra网络为例[J]. 测绘学报, 2022, 51(2): 248-257.
[12]	张永生, 张振超, 童晓冲, 纪松, 于英, 赖广陵. 地理空间智能研究进展和面临的若干挑战[J]. 测绘学报, 2021, 50(9): 1137-1146.
[13]	艾廷华. 深度学习赋能地图制图的若干思考[J]. 测绘学报, 2021, 50(9): 1170-1182.
[14]	龚健雅, 许越, 胡翔云, 姜良存, 张觅. 遥感影像智能解译样本库现状与研究[J]. 测绘学报, 2021, 50(8): 1013-1022.
[15]	杨必胜, 韩旭, 董震. 适用于城市场景大规模点云语义标识的深度学习网络[J]. 测绘学报, 2021, 50(8): 1059-1067.