Spatiotemporal modeling of land subsidence using a geographically weighted deep learning method based on PS-InSAR

Li, Huijun; Zhu, Lin; Dai, Zhenxue; Gong, Huili; Guo, Tao; Guo, Guijie; Wang, Jingbo; Teatini, Pietro

doi:10.1016/j.scitotenv.2021.149244

Cited by 41 publications

(24 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Step 3:  ℎ,𝑡 is defined as the partial derivative of the loss function with respect to ℎ 𝑡 , denoted as shown in Equation (10).…”

Section: 2 Lstm-amsgdmentioning

confidence: 99%

See 1 more Smart Citation

Land Subsidence Time Series Prediction Method Based on LSTM-AMSGD

Qian¹,

Shi²,

Lv³

et al. 2022

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Accurate prediction of geological subsidence is of great importance for geological hazard risk assessment. Various existing prediction models do not take into account the time correlation between geological subsidence, and the prediction effect lacks practical significance. In this paper, an LSTM-AMSGD-based land subsidence prediction method is proposed. Firstly, the high-precision time series inversion results of large-area land surface deformation are obtained by the small baseline interference technique with multiple principal image coherent targets. Secondly, a recurrent neural network (LSTM-AMSGD) is used as the network architecture. The final cumulative subsidence prediction error is within 0.3 mm, and the single-step prediction of more than 400,000 observation points can be completed in 126s. Therefore, the LSTM-AMSGD model in this paper is effective for the prediction of geological subsidence.

show abstract

“…Step 3:  ℎ,𝑡 is defined as the partial derivative of the loss function with respect to ℎ 𝑡 , denoted as shown in Equation (10).…”

Section: 2 Lstm-amsgdmentioning

confidence: 99%

“…In recent years, deep learning has received a lot of attention as a new machine learning method in many industries [10]. Long Short Term Memory (LSTM) models enable the current operation of the network to be related not only to the current input data, but also to previous data.…”

Section: Introductionmentioning

confidence: 99%

Land Subsidence Time Series Prediction Method Based on LSTM-AMSGD

Qian¹,

Shi²,

Lv³

et al. 2022

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

show abstract

“…The mathematical prediction model have a wider range of applications than physical prediction model. However, mathematical prediction model has a poor prediction performance when the dispersion of land subsidence value is large (Li et al 2021). This phenomenon should be attributed to the fact that the mathematical prediction model cannot explain the mechanism of land subsidence.…”

Section: Introductionmentioning

confidence: 97%

“…The land subsidence prediction model based on machine learning method can integrate the advantages of the physical prediction model and mathematical prediction model and overcome the shortcomings of the physical prediction model and mathematical prediction model. Previous studies have shown that the land subsidence prediction model based on machine learning method can obtain reliable prediction results (Shi et al 2020;Li et al 2021). This paper was organized as follows.…”

Section: Introductionmentioning

confidence: 99%

Land subsidence prediction model based on its influencing factors and machine learning methods

Liu

Tao

et al. 2022

Nat Hazards

View full text Add to dashboard Cite

Land subsidence has caused huge economic losses in the Beijing plains (BP) since 1980s. Building land subsidence prediction models that can predict the development of land subsidence is of great signi cance for improving the safety of cities and reducing economic losses in Eastern Beijing plains.The pattern of evolution of land subsidence is affected by many factors including groundwater level in different aquifers, thicknesses of compressible layers, and static and dynamic loads caused by urban construction. First, we used the small baseline subset Interferometric Synthetic Aperture Radar (SBAS-InSAR) technology on 47 ENVISAT ASAR images and 48 RADARSAT-2 images and used Persistent Scatterers Interferometric Aperture Radar (PS-InSAR) technology on 27 Sentinel-1 images to obtain the land subsidence monitoring results from June 2003 to September 2018. Second, the accuracy of the InSAR monitoring results were validated by using leveling benchmark land subsidence monitoring results.Finally, we built land subsidence rate prediction models and land subsidence gradient prediction models by combining land subsidence in uencing factors and four machine learning methods including support vector machine (SVM), Gradient Boosting Decision Tree (GBDT), Random forest (RF) and Extremely Randomized Trees (ERT). The ndings show: (1) The InSAR monitoring results revealed that the maximum land subsidence rate reached − 110.7 mm/year, -144.4 mm/year and − 136.8 mm/year during the 2003-2010, 2011-2015 and 2016-2018 periods, respectively. (2): The InSAR monitoring results agreed well with the leveling benchmark monitoring results with the Pearson correlation coe cients of two monitoring results were 0.97, 0.96 and 0.95 during the 2003-2010, 2011-2015 and 2016-2018 periods, respectively. (3): We found that the land subsidence prediction based on ERT method is the optimal model among four land subsidence prediction models and that the prediction performance of land subsidence prediction model based on ERT method will be greatly improved when apply this prediction model in sub study areas where the land subsidence mechanism is similar owning to the similar hydrogeological parameters.

show abstract

“…ese catastrophes include uneven settlement of foundations, cracking of walls, and tilting of buildings [1][2][3]. It is possible that the aforesaid scenario will occur and the structure will be demolished without any prior planning that will result in several difficulties including resource waste and environmental degradation [4][5][6][7].…”

Section: Introductionmentioning

confidence: 99%

A Discrete-Time Model-Based Method for Predicting Settlement of Geotechnical Foundations in Buildings

Yang

Jiang

2022

Mobile Information Systems

View full text Add to dashboard Cite

The selection of a single prediction method is difficult to adapt to the actual engineering situation in the construction geotechnical foundation’s settlement prediction. This article proposes a modified optimized comprehensive prediction model to independently mine the construction geotechnical foundation settlement monitoring data from different angles. The proposed model analyzes the construction geotechnical foundation settlement’s change law and realizes the comprehensive prediction of construction geotechnical foundation settlement. Initially, the preliminary integrated discrete-time prediction model is established based on the combination of the hyperbolic method and the GM (1, 1) model’s modeling mechanism. The GM (1, 1) modeling mechanism is based on the idea of arithmetic weighted average combination. Then the real-time correction weight coefficients are constructed and the real-time correction amount is calculated to modify the initial integrated prediction model. The modified optimized integrated prediction model is established too. Finally, the modified optimized integrated prediction model is utilized to predict the settlement of building geotechnical foundations. The experimental results of A-Ma city project, China show that the modified optimized integrated prediction model has better prediction accuracy and has universal applicability than the hyperbolic method and GM (1, 1) in the prediction of foundation settlement of building geotechnical foundations. It can effectively reduce the prediction error after the combination of single methods.

show abstract

Spatiotemporal modeling of land subsidence using a geographically weighted deep learning method based on PS-InSAR

Cited by 41 publications

References 50 publications

Land Subsidence Time Series Prediction Method Based on LSTM-AMSGD

Land Subsidence Time Series Prediction Method Based on LSTM-AMSGD

Land subsidence prediction model based on its influencing factors and machine learning methods

A Discrete-Time Model-Based Method for Predicting Settlement of Geotechnical Foundations in Buildings

Contact Info

Product

Resources

About