Précédent|Prochain

Ground subsidence prediction model in Xinmi City based on TSC—LSTM

  • role： First author第一作者
  • Affiliation：

    School of Geophysics & Geomatics, China University of Geosciences, Wuhan430074, China

  • Email：hewenzhao@cug.edu.cn
  • Introduction：赵贺文，研究方向为基于InSAR技术的城市地面沉降监测和预测。E-mail： hewenzhao@cug.edu.cn
ZHAO Hewen,  
  • role： Corresponding author通信作者
  • Affiliation：

    School of Geophysics & Geomatics, China University of Geosciences, Wuhan430074, China

  • Email：taochen@cug.edu.cn
  • Introduction：陈涛，研究方向为遥感地学应用。E-mail： taochen@cug.edu.cn
CHEN Tao*

résumé

Predicting land subsidence is crucial for conducting in-depth analyses of and providing early warnings for urban land subsidence patterns. However, traditional numerical prediction models frequently experience difficulty in accurately capturing the intricate characteristics of land subsidence data, leading to less precise predictions. This study focuses on Xinmi City and endeavors to improve the accuracy of land subsidence prediction by combining time series feature extraction methods with time series prediction techniques.In this study, 232 interference images provided by HyP3 were utilized to acquire land subsidence information in Xinmi City from January 2018 to December 2022, employing Small Baseline Subset (SBAS)-Interferometric Synthetic Aperture radar (InSAR) technology. Recognizing the challenge of achieving high accuracy in directly predicting land subsidence data, this study developed a land subsidence prediction model that integrates trend and seasonal characteristics by using a Long Short-term Memory (LSTM) network, namely, the Trend Seasonal Characteristics-LSTM (TSC-LSTM). The TSC-LSTM model capitalizes on the strengths of weighted regression seasonal trend decomposition (STL) in extracting time series features from settlement data and the LSTM model for addressing the vanishing gradient problem in time series prediction.This fusion of techniques allows for a precise analysis of land subsidence data and enables highly accurate predictions. Distinguishing itself from the conventional LSTM model, the TSC-LSTM model refrains from directly inputting ground subsidence data. Instead, it employs STL to meticulously extract trend and seasonal characteristics from land subsidence data. This approach maximizes the utilization of characteristic information inherent in land subsidence data. Subsequently, these features are fed into the LSTM model for prediction. This unique methodology reduces noise interference and significantly enhances the accuracy of model predictions.This research leverages time-series InSAR data for Xinmi City from 2018 to 2022. It employs the TSC-LSTM model, deep learning architectures (recurrent neural network and LSTM), and conventional machine learning algorithms (multilayer perceptron and support vector regression) to forecast the cumulative subsidence data for five subsidence centers by using SBAS-InSAR. This study identifies the two most optimal models and validates their efficacy in single-point prediction scenarios, utilizing domain-specific terminologies. Research findings indicate the following. (1) Between 2018 and 2022, Xinmi City experienced a land subsidence rate that ranged from -60.3 mm to 51.96 mm per annum, resulting in the identification of five distinct land subsidence center areas. Among these, the highest cumulative settlement and uplift reached 304.9 mm and 197.68 mm, respectively. The universality of the TSC-LSTM model across diverse datasets has been corroborated, demonstrating its high precision, exceptional generalization capability, and stable high performance in the prediction of land subsidence, employing specialized terminologies. (2) The TSC-LSTM model exhibited exceptional performance in predicting the five subsidence center areas. The R² values for the TSC-LSTM model range from 0.9985 to 0.9992, significantly surpassing the second-best model, i.e., LSTM, which has an R² range of 0.9662 to 0.9872. Moreover, the root mean square error values for the prediction accuracy of the TSC-LSTM model are less than 2 mm, achieving a range of 1.2426 mm to 1.7403 mm . (3) Single-point prediction results demonstrate the superior ability of the TSC-LSTM model to accurately capture local changes in the cumulative settlement data. The TSC-LSTM model proposed in this study outperforms the traditional LSTM model in terms of prediction accuracy and model stability, providing robust support for in-depth research on urban land subsidence.

mots-clés

Ground settlement prediction;TSC—LSTM;SBAS—InSAR;Cumulative settlement data decomposition;Xinmi City;LSTM

References

  1. 1.
    Abbasimehr H and Paki R. 2022. Improving time series forecasting using LSTM and attention models. Journal of Ambient Intelligence and Humanized Computing, 13(1): 673-691
  2. 2.
    Apaydin H, Sattari M T, Falsafian K and Prasad R. 2021. Artificial intelligence modelling integrated with Singular Spectral analysis and Seasonal-Trend decomposition using Loess approaches for streamflow predictions. Journal of Hydrology, 600: 126506
  3. 3.
    Bai L, Jiang L M, Wang H S and Sun Q S. 2016. Spatiotemporal characterization of land subsidence and uplift (2009-2010) over Wuhan in Central China revealed by TerraSAR-X InSAR analysis. Remote Sensing, 8(4): 350
  4. 4.
    Berardino P, Fornaro G, Lanari R and Sansosti E. 2002. A new algorithm for surface deformation monitoring based on small baseline differential SAR interferograms. IEEE Transactions on Geoscience and Remote Sensing, 40(11): 2375-2383
  5. 5.
    Cao J, Li Z and Li J. 2019. Financial time series forecasting model based on CEEMDAN and LSTM. Physica A: Statistical Mechanics and Its Applications, 519: 127-139
  6. 6.
    Cao X Y, Zhu L, Gong H L, Guo L, Wei Y J, Guo T, Chen B B, Wang H G and Li H J. 2022. Land subsidence simulation in the east of Beijing plain based on the AM-LSTM network. National Remote Sensing Bulletin, 26(7): 1302-1314
  7. 7.
    Chen Y, He Y, Zhang L F, Chen B S, He X, Pu H Y, Cao S P, Gao L Y and Yang W. 2022. Surface deformation prediction based on TS-InSAR technology and long short-term memory networks. National Remote Sensing Bulletin, 26(7): 1326-1341
  8. 8.
    Chen Y Y, Zhao B K, Wang H, Zheng J Z and Gao Y H M. 2024. Time-series InSAR ground deformation prediction based on LSTM model. Yangtze River, 55(3): 146-152
  9. 9.
    Cheng S, Sui B B, Shen Y and Wang T. 2015. Prediction of mining ground settlement based on BP neural network. Geomatics and Spatial Information Technology, 38(3): 18-20
  10. 10.
    Cigna F and Tapete D. 2022. Urban growth and land subsidence: multi-decadal investigation using human settlement data and satellite InSAR in Morelia, Mexico. Science of the Total Environment, 811: 152211
  11. 11.
    Ding Q, Shao Z F, Huang X, Altan O, Zhuang Q W and Hu B. 2021. Monitoring, analyzing and predicting urban surface subsidence: a case study of Wuhan City, China. International Journal of Applied Earth Observation and Geoinformation, 102: 102422
  12. 12.
    Ding S, Zhang H H, Tao Z and Li R J. 2022. Integrating data decomposition and machine learning methods: an empirical proposition and analysis for renewable energy generation forecasting. Expert Systems with Applications, 204: 117635
  13. 13.
    Ferretti A, Prati C and Rocca F. 2000. Nonlinear subsidence rate estimation using permanent scatterers in differential SAR interferometry. IEEE Transactions on Geoscience and Remote Sensing, 38(5): 2202-2212
  14. 14.
    Goorabi A, Karimi M, Yamani M and Perissin D. 2020. Land subsidence in Isfahan metropolitan and its relationship with geological and geomorphological settings revealed by sentinel-1A InSAR observations. Journal of Arid Environments, 181: 104238
  15. 15.
    Guo C N, Wang X R, Wang X S and Lian J X. 2023. Land subsidence prediction method based on ARIMA-GRU model. Urban Geology, 18(2): 138-143
  16. 16.
    Hochreiter S and Schmidhuber J. 1997. Long short-term memory. Neural Computation, 9(8): 1735-1780
  17. 17.
    Hogenson K, Kristenson H, Kennedy J, Johnston A, Rine J, Logan T, Zhu J, Williams F, Herrmann J, Smale J and Meyer F. 2024. Hybrid Pluggable Processing Pipeline (HyP3): a cloud-native infrastructure for generic processing of SAR data. Computer software
  18. 18.
    Jia S M, Tian F and Qi G. 2019. Urban “chronic disease”: the origin and comprehensive control countermeasures of land subsidence. City and Disaster Reduction, (3): 22-27
  19. 19.
    Lanari R, Casu F, Manzo M, Zeni G, Berardino P, Manunta M and Pepe A. 2007. An overview of the small BAseline subset algorithm: a DInSAR technique for surface deformation analysis. Pure and Applied Geophysics, 164(4): 637-661
  20. 20.
    Li G Y, Zhang R, Liu G X, Yu B, Zhang B, Dai K R, Bao J W and Wei B W. 2018. Land subsidence detection and analysis over Beijing-Tianjin-Hebei area based on Sentinel-1A TS-DInSAR. Journal of Remote Sensing (in Chinese), 22(4): 633-646
  21. 21.
    Li M W, Zhang X D, Bai Z C, Xie H Y and Chen B. 2022. Land subsidence in Qingdao, China, from 2017 to 2020 based on PS-InSAR. International Journal of Environmental Research and Public Health, 19(8): 4913
  22. 22.
    Li Y T, Bao T F, Gong J, Shu X S and Zhang K. 2020. The prediction of dam displacement time series using STL, extra-trees, and stacked LSTM neural network. IEEE Access, 8: 94440-94452
  23. 23.
    Liu C Z and Huang S. 2022. Research on “7·20” mountain torrents and geological disasters in Zhengzhou city, Henan province of China. Journal of Engineering Geology, 30(3): 931-943
  24. 24.
    Liu Q H, Liu H M, Zhang Y H, Wu H A and Deng M. 2022. A spatio-temporal prediction method of large-scale ground subsidence considering spatial heterogeneity. National Remote Sensing Bulletin, 26(7): 1315-1325
  25. 25.
    Liu Q H, Zhang Y H, Deng M, Wu H A, Kang Y H and Wei J J. 2021. Time series prediction method of large-scale surface subsidence based on deep learning. Acta Geodaetica et Cartographica Sinica, 50(3): 396-404
  26. 26.
    Lü C Z and An D D. 2021. Application of grey prediction model on land subsidence. Geomatics and Spatial Information Technology, 44(3): 73-75
  27. 27.
    Peng H J, Zhou Y, Hu X F, Zhang L, Peng Y Z and Cai X Y. 2023. A PM2.5 prediction model based on deep learning and random forest. National Remote Sensing Bulletin, 27(2): 430-440
  28. 28.
    Sabrian P G, Saepuloh A, Kashiwaya K and Koike K. 2021. Combined SBAS-InSAR and geostatistics to detect topographic change and fluid paths in geothermal areas. Journal of Volcanology and Geothermal Research, 416: 107272
  29. 29.
    Sun X P, Lu X Y, Wen X H, Zhan Y and Wang L. 2016. Monitoring of ground subsidence in Chengdu plain using SBAS-InSAR. Remote Sensing for Land and Resources, 28(3): 123-129
  30. 30.
    Vagale A, Šteina L and Vēciņš V. 2021. Time series forecasting of mobile robot motion sensors using LSTM networks. Applied Computer Systems, 26(2): 150-157
  31. 31.
    Xiong J C, Nie Y J, Luo Y and Li Y F. 2019. Monitoring urban land subsidence by dual-polarization Sentinel-1 data: a case study of Shanghai. Bulletin of Surveying and Mapping, 11: 98-102, 129
  32. 32.
    Xu L, Ou Y X, Cai J J, Wang J, Fu Y and Bian X Y. 2023. Offshore wind speed assessment with statistical and attention-based neural network methods based on STL decomposition. Renewable Energy, 216: 119097
  33. 33.
    Yan X X, Yang T L, Xu Y, Tosi L, Stouthamer E, Andreas H, Minderhoud P, Ladawadee A, Hanssen R, Erkens G, Teatini P, Lin J X, Bonì R, Chimpalee J, Huang X L, Da Lio C, Meisina C and Zucca F. 2020. Advances and practices on the research, prevention and control of land subsidence in coastal cities. Acta Geologica Sinica-English Edition, 94(1): 162-175
  34. 34.
    Ye Y C, Yan C D, Luo X X, Zhang R F and Yuan G J. 2022. Analysis of ground subsidence along Zhengzhou metro based on time series InSAR. National Remote Sensing Bulletin, 26(7): 1342-1353
  35. 35.
    Yin H L, Jin D, Gu Y H, Park C J, Han S K and Yoo S J. 2020. STL-ATTLSTM: vegetable price forecasting using STL and attention mechanism-based LSTM. Agriculture, 10(12): 612
  36. 36.
    Yu W, Gong H L, Chen B B and Zhou C F. 2022. Spatial-temporal evolution characteristics and prediction of land subsidence in the eastern plain of Beijing. Remote Sensing for Natural Resources, 34(4): 183-193
  37. 37.
    Yue Z H, Shen T, Mao X and Ma W J. 2020. Study on prediction method of land subsidence based on recurrent neural network. Science of Surveying and Mapping, 45(12): 145-152
  38. 38.
    Zhang T F, Xing X M, Peng W, Zhu J, Liu X B, Ge J W and Lei M C. 2024. Incorporation of Coordinate-Time Function (CT-PIM) time-series InSAR deformation prediction for salt mining areas: case study of the Huaian Salt Mine. National Remote Sensing Bulletin, 28(6): 1615-1631
  39. 39.
    Zhang Y J, Fattahi H and Amelung F. 2019. Small baseline InSAR time series analysis: unwrapping error correction and noise reduction. Computers and Geosciences, 133: 104331
  40. 40.
    Zhang Y M, Wang P, Luo X, Zhang Q and Chen H. 2017. Monitoring Xi'an land subsidence using Sentinel-1 images and SBAS-InSAR technology. Bulletin of Surveying and Mapping, (4): 93-97
  41. 41.
    Zhao F C, Miao F S, Wu Y P, Xiong Y, Gong S Q and Sun D K. 2024. Land subsidence susceptibility mapping in urban settlements using time-series PS-InSAR and random forest model. Gondwana Research, 125: 406-424
  42. 42.
    Zheng X X, Wu Q, Hou Y S and Ying Y F. 2002. Some frontier problems on land subsidence research. Acta Geoscientia Sinica, 23(3): 279-282
  43. 43.
    Zhou D Y, Zuo X Q, Zhao Z F, Xi W F and Ge C. 2023. Prediction of urban land subsidence by SBAS-InSAR and improved BP neural network. Geological Bulletin of China, 42(10): 1774-1783
  44. 44.
    Zhou W, Chen F L and Guo H D. 2015. Differential radar interferometry for structural and ground deformation monitoring: a new tool for the conservation and sustainability of cultural heritage sites. Sustainability, 7(2): 1712-1729
  45. 45.
    Zhu J J, Li Z W and Hu J. 2017. Research progress and methods of InSAR for deformation monitoring. Acta Geodaetica et Cartographica Sinica, 46(10): 1717-1733
  46. 46.
    Zhu X G and Li G E. 2021. Urban land subsidence monitoring using multi-sensor SAR datasets. Bulletin of Surveying and Mapping, (6): 71-75

Lire l'article complet

The above content is generated by Large Model Translation. The translated content is for reference only. We do not assume any commercial or legal responsibilty for any consequences arising from the use of our website