Subsidence prediction of high-fill areas based on InSAR monitoring data and the PSO-SVR model
-
摘要: 基于SBAS-InSAR技术和机器学习知识对高填方区域进行地表沉降监测及预测，对工程项目的施工、检修、运营等工作都具有重要的指导意义。本文以重庆东港集装箱码头为研究对象，选取2018-2019年覆盖研究区的31景Sentinel-1A数据，利用SBAS-InSAR技术获取该区域的地表沉降数据，并进行内外精度评定；通过信息量模型分析地表沉降易发地地势特点，选择预测点位；通过灰色关联分析计算动态影响因素与沉降量之间的灰色关联度，使用主成分分析法从影响因素中提取出主成分，构建训练集和测试集，通过PSO-SVR预测模型对测试集数据进行预测。为验证该模型在高填方区域沉降预测的可靠性和优异性，将ARIMA模型作为对比模型，分别将PSO-SVR模型的预测结果和ARIMA模型的预测结果与测试集进行对比。结果表明：PSO-SVR模型的预测精度优于ARIMA模型，在高填方区域地表沉降预测中具有较好的实用性。Abstract: Based on SBAS-InSAR technology and machine learning knowledge, the monitoring and prediction of surface settlement in high-fill areas have important guiding significance for construction, maintenance, and operation of engineering projects. This study takes the Chongqing Donggang Container Terminal as the research object, and utilizes 31 scenes of Sentinel-1A data from 2018 to 2019. The surface subsidence data of the area is obtained by SBAS-InSAR technology, and the internal and external accuracy is evaluated. The topography characteristics of the prone areas of surface subsidence were analyzed through an information quantity model to select prediction points. Grey Relational Analysis （GRA） was used to calculate the grey correlation degree between dynamic influencing factors and subsidence. Principal component analysis was used to extract principal components from influencing factors, and training and testing sets were constructed. PSO-SVR prediction model was used to predict the testing set data. To verify the reliability and superiority of the model in subsidence prediction in high-fill areas, the ARIMA model was used as a comparative model, and the prediction results of the PSO-SVR model and the ARIMA model were compared with the testing set. The results show that the prediction accuracy of the PSO-SVR model is better than that of the ARIMA model, and it has better practicality in predicting surface subsidence in high-fill areas.
-
图 4 Google Earth历史影像图
（图a为2018年4月影像图；图b为2021年9月影像图）
Figure 4. Google Earth historical imagery
(Fig.4a. Apr 2018; Fig.4b. Sep 2021)
图 6 各模型预测结果
（图a为形变点1的预测结果；图b为形变点2的预测结果）
Figure 6. The prediction results of each model for each deformation points
(Fig.6a. Deformation point 1; Fig.6b. Deformation point 2)
表 1 监测点地表形变结果
Table 1. Surface deformation results of monitoring sites
点名 基于SBAS-InSAR技术获取的LOS向形变数据（mm） 基于SBAS-InSAR技术获取的垂直形变数据（mm） 水准测量获取的形变量（mm） 1 −12.75 −15.26 −15.20 2 −13.62 −16.29 −13.60 3 −13.62 −16.29 −9.20 …… …… …… …… 46 −22.75 −27.22 −39.80 47 −21.42 −25.63 −36.00 48 −20.45 −24.46 −17.40 
下载: 导出CSV
表 2 静态影响因素信息量计算结果
Table 2. Information quantity calculation results of static influencing factors
静态影响因子 影响因子分级 信息量 高程/m 151−187 0.82 187−231 0.07 231−263 0.64 263−292 −1.28 292−333 −5.62 坡度/（°） 0-5 −0.32 5-10 0.00 10−15 0.08 15−20 0.37 >20 0.20 坡向/（°） 平坡 −1.78 北坡 0.36 东北坡 −0.02 东坡 −0.80 东南坡 −0.47 南坡 −0.16 西南坡 0.42 西坡 −0.21 西北坡 0.23 平面曲率 0−16.9 0.05 16.9−33.6 −0.22 33.6−50.5 −0.14 50.5−67.4 −0.19 67.4−81.5 0.31 剖面曲率 0 - 2.9 0.01 2.9−5.8 −0.01 5.8−9 −0.11 9−13.6 0.24 13.6−26.8 −0.23 道路缓冲区/m 0−30 −0.78 30−60 −0.19 60−90 0.13 >90 0.28 水系缓冲区/m 0−500 −0.08 500−1000 0.68 1000−1500 −0.74 1500−2000 0.48 >2000 −2.68 地形起伏度/m 0−7 −0.48 7−13 0.00 13−19 0.13 19−29 0.25 29−48 0.26 人类活动缓冲区/m 100 0.56 200 −1.07 300 −1.13 400 −0.64 >400 −1.89
下载: 导出CSV
表 3 灰色关联度
Table 3. Summary table of Grey Relational Degree
影响因素 气温 水位 地下水 NDVI 降雨量 灰色关联度 0.758 4 0.758 3 0.692 9 0.666 7 0.622 3
下载: 导出CSV
表 4 PSO-SVR模型的预测结果
Table 4. Prediction results of the PSO-SVR model
点号 日期 真实值/mm 预测值/mm 形变点1 2019/10/14 −14.71 −14.61 2019/11/7 −17.43 −17.34 2019/12/1 −20.71 −20.70 2019/12/25 −21.42 −21.32 形变点2 2019/10/14 −13.08 −13.18 2019/11/7 −16.63 −16.53 2019/12/1 −18.04 −18.03 2019/12/25 −20.29 −20.19
下载: 导出CSV
表 5 ARIMA模型的预测结果
Table 5. Prediction results of the ARIMA model
点号 日期 真实值/mm 预测值/mm 形变点1 2019/10/14 −14.71 −15.26 2019/11/7 −17.44 −17.98 2019/12/1 −20.70 −21.32 2019/12/25 −21.42 −22.14 形变点2 2019/10/14 −13.08 −13.75 2019/11/7 −16.63 −17.12 2019/12/1 −18.04 −18.63 2019/12/25 −20.29 −20.91
下载: 导出CSV
表 6 精度评定表
Table 6. Accuracy evaluation table
模型 点号 MAE MSE R2 PSO-SVR 形变点1 0.075 3 0.007 5 0.999 0 形变点2 0.075 0 0.007 5 0.998 9 ARIMA 形变点1 0.606 9 0.373 1 0.948 3 形变点2 0.593 3 0.356 8 0.947 9
下载: 导出CSV
-
[1] 刘强. 时序InSAR技术在中型城市地表形变时空特征应用及预测分析[D]. 抚州: 东华理工大学LIU Qiang. Application and prediction analysis of time series InSAR technology in temporal and spatial characteristics of surface deformation in medium-sized cities[D]. Fuzhou: East China Institute of Technology. （in Chinese with English abstract） [2] 李金超. 基于InSAR和Sentinel-1A的淮南矿区形变灾害监测研究[D]. 合肥: 合肥工业大学LI Jinchao. Research on deformation disaster monitoring in Huainan mining area based on InSAR and sentinel-1A[D]. Hefei: Hefei University of Technology. （in Chinese with English abstract） [3] FERRETTI A,PRATI C,ROCCA F. Permanent scatterers in SAR interferometry[J]. IEEE Transactions on Geoscience and Remote Sensing,2001,39(1):8 − 20. doi: 10.1109/36.898661 [4] BERARDINO P,FORNARO G,LANARI R,et al. A new algorithm for surface deformation monitoring based on small baseline differential SAR interferograms[J]. IEEE Transactions on Geoscience and Remote Sensing,2002,40(11):2375 − 2383. doi: 10.1109/TGRS.2002.803792 [5] 于书媛,杨源源,张鹏飞,等. 运用时序InSAR技术监测合肥市地面沉降及断裂活动[J]. 大地测量与地球动力学,2021,41(4):398 − 402. [YU Shuyuan,YANG Yuanyuan,ZHANG Pengfei,et al. Monitoring land subsidence and fault activity in Hefei city based on MT-InSAR[J]. Journal of Geodesy and Geodynamics,2021,41(4):398 − 402. (in Chinese) doi: 10.14075/j.jgg.2021.04.014 [6] 代聪,李为乐,陆会燕,等. 甘肃省舟曲县城周边活动滑坡InSAR探测[J]. 武汉大学学报(信息科学版),2021,46(7):994 − 1002. [DAI Cong,LI Weile,LU Huiyan,et al. Active landslides detection in Zhouqu County,Gansu Province using InSAR technology[J]. Geomatics and Information Science of Wuhan University,2021,46(7):994 − 1002. (in Chinese) [7] 赵现仁. 基于时序干涉SAR的北极地区冰川运动监测分析[D]. 北京: 北京建筑大学ZHAO Xianren. Monitoring and analysis of glacier movement in Arctic based on time series interferometric SAR[D]. Beijing: Beijing University of Civil Engineering and Architecture. （in Chinese with English abstract） [8] 周定义,左小清. 基于SBAS-InSAR和PSO-BP神经网络算法的矿区地表沉降监测及预测[J]. 云南大学学报(自然科学版),2021,43(5):895 − 905. [ZHOU Dingyi,ZUO Xiaoqing. Surface subsidence monitoring and prediction in mining area based on SBAS-InSAR and PSO-BP neural network algorithm[J]. Journal of Yunnan University (Natural Sciences Edition),2021,43(5):895 − 905. (in Chinese) [9] 师芸,李杰,吕杰,等. 结合SBAS-InSAR与支持向量回归的开采沉陷监测与预测[J]. 遥感信息,2021,36(2):6 − 12. [SHI Yun,LI Jie,LV Jie,et al. Monitoring and prediction of mining subsidence based on SBAS-InSAR and improved support vector regression[J]. Remote Sensing Information,2021,36(2):6 − 12. (in Chinese) doi: 10.3969/j.issn.1000-3177.2021.02.002 [10] CORTES C,VAPNIK V. Support-vector networks[J]. Machine Learning,1995,20(3):273 − 297. [11] 喜文飞,杨正荣,赵子龙,等. 基于SBAS-InSAR技术的小区域沉降监测研究[J]. 云南师范大学学报(自然科学版),2022,42(4):41 − 44. [XI Wenfei,YANG Zhengrong,ZHAO Zilong,et al. Study on settlement monitoring in small area based on the SBAS-InSAR technology[J]. Journal of Yunan Normal University (Natural Sciences Edition),2022,42(4):41 − 44. (in Chinese) [12] 杨康,薛喜成,李识博. 信息量融入GA优化SVM模型下的地质灾害易发性评价[J]. 安全与环境工程,2022,29(3):109 − 118. [YANG Kang,XUE Xicheng,LI Shibo. Geological hazard susceptibility assessment by incorporating information value into GA optimized SVM model[J]. Safety and Environmental Engineering,2022,29(3):109 − 118. (in Chinese) doi: 10.13578/j.cnki.issn.1671-1556.20210976 [13] 谭学瑞,邓聚龙. 灰色关联分析:多因素统计分析新方法[J]. 统计研究,1995,12(3):46 − 48. [TAN Xuerui,DENG Julong. 灰色关联分析:多因素统计分析新方法[J]. Statistical Research,1995,12(3):46 − 48. (in Chinese with English abstract) [14] 何旭,何毅,张立峰,等. 联合InSAR与PCA的北京平原地面沉降时空分析[J]. 光谱学与光谱分析,2022,42(7):2315 − 2324. [HE Xu,HE Yi,ZHANG Lifeng,et al. Spatio-temporal analysis of land subsidence in beij ing plain based on InSAR and PCA[J]. Spectroscopy and Spectral Analysis,2022,42(7):2315 − 2324. (in Chinese) [15] KENNEDY J, EBERHART R. Particle swarm optimization[C]//Proceedings of ICNN'95 - International Conference on Neural Networks. November 27 - December 1, 1995, Perth, WA, Australia. IEEE, 2002: 1942-1948. [16] 韩冰,袁颖. 基于主成分分析的GA-SVM地表沉降预测模型[J]. 中国科技论文,2018,13(9):1045 − 1049. [HAN Bing,YUAN Ying. Application of GA-SVM model based on principal component nalysis to prediction of surface settlement of shield construction[J]. China Sciencepaper,2018,13(9):1045 − 1049. (in Chinese) [17] 冯小蔓. 基于InSAR技术的地表形变与雪深监测研究[D]. 太原: 太原理工大学FENG X （M/W）. Research on monitoring of surface deformation and snow depth based on InSAR technology[D]. Taiyuan: Taiyuan University of Technology. （in Chinese with English abstract） [18] 潘建平,邓福江,徐正宣,等. 基于轨道精炼控制点精选的极艰险区域时序InSAR地表形变监测[J]. 中国地质灾害与防治学报,2021,32(5):98 − 104. [PAN Jianping,DENG Fujiang,XU Zhengxuan,et al. Time series InSAR surface deformation monitoring in extremely difficult area based on track refining control points selection[J]. The Chinese Journal of Geological Hazard and Control,2021,32(5):98 − 104. (in Chinese with English abstract) doi: 10.16031/j.cnki.issn.1003-8035.2021.05-12 [19] FAN H,WANG L,WEN B,et al. A new model for three-dimensional deformation extraction with single-track InSAR based on mining subsidence characteristics[J]. International Journal of Applied Earth Observation and Geoinformation,2021,94:102223. doi: 10.1016/j.jag.2020.102223 [20] 中国国土资源航空物探遥感中心. DD2014-11地面沉降干涉雷达数据处理技术规程[S]. 北京: 中国地质调查局, 2014China Aero Geophysical Survey and Remote Sensing Center for Land and Resources. DD2014-11 Technical Specification for Data Processing of Land Subsidence Interference Radar[S]. Beijing: China Geological Survey, 2014. （in Chinese with English abstract） [21] 杜国梁,杨志华,袁颖,等. 基于逻辑回归-信息量的川藏交通廊道滑坡易发性评价[J]. 水文地质工程地质,2021,48(5):102 − 111. [DU Guoliang,YANG Zhihua,YUAN Ying,et al. Landslide susceptibility mapping in the Sichuan-Tibet traffic corridor using logistic regression-information value method[J]. Hydrogeology & Engineering Geology,2021,48(5):102 − 111. (in Chinese with English abstract) [22] 刘刚,徐成华,施威,等. 南京河西地区地面沉降成因分析[J]. 地质论评,2023,69(2):639 − 647. [LIU Gang,XU Chenghua,SHI Wei,et al. Cause analysis of land subsidence in Hexi area,Nanjing[J]. Geological Review,2023,69(2):639 − 647. (in Chinese) doi: 10.16509/j.georeview.2022.08.081 [23] 葛伟丽,李元杰,张春明,等. 基于InSAR技术的内蒙古巴彦淖尔市地面沉降演化特征及成因分析[J]. 水文地质工程地质,2022,49(4):198 − 206. [GE Weili,LI Yuanjie,ZHANG Chunming,et al. An attribution analysis of land subsidence features in the city of Bayannur in Inner Mongolia based on InSAR[J]. Hydrogeology and Engineering Geology,2022,49(4):198 − 206. (in Chinese) doi: 10.16030/j.cnki.issn.1000-3665.202106022 [24] 秦胜伍,张延庆,张领帅,等. 基于Stacking模型融合的深基坑地面沉降预测[J]. 吉林大学学报(地球科学版),2021,51(5):1316 − 1323. [QIN Shengwu,ZHANG Yanqing,ZHANG Lingshuai,et al. Prediction of ground settlement around deep foundation pit based on stacking model fusion[J]. Journal of Jilin University (Earth Science Edition),2021,51(5):1316 − 1323. (in Chinese) -
下载:
点击查看大图
图(6) / 表(6)
计量
- 文章访问数: 909
- HTML全文浏览量: 2140
- PDF下载量: 121
- 被引次数: 0
邮件订阅
RSS