Prediction of Mine Subsidence Based on InSAR Technology and the LSTM Algorithm: A Case Study of the Shigouyi Coalfield, Ningxia (China)

Abstract

The accurate prediction of surface subsidence induced by coal mining is critical to safeguarding the environment and resources. However, the precision of current prediction models is often restricted by the lack of pertinent data or imprecise model parameters. To overcome these limitations, this study proposes an approach to predicting mine subsidence that leverages Interferometric Synthetic Aperture Radar (InSAR) technology and the long short-term memory network (LSTM). The proposed approach utilizes small baseline multiple-master high-coherent target (SBMHCT) interferometric synthetic aperture radar technology to monitor the mine surface and applies the long short-term memory (LSTM) algorithm to construct the prediction model. The Shigouyi coalfield in Ningxia Province, China was chosen as a study area, and time series ground subsidence data were obtained based on Sentinel-1A data from 9 March 2015 to 7 June 2016. To evaluate the proposed approach, the prediction accuracies of LSTM and Support Vector Regression (SVR) were compared. The results show that the proposed approach could accurately predict mine subsidence, with maximum absolute errors of less than 2 cm and maximum relative errors of less than 6%. The findings demonstrate that combining InSAR technology with the LSTM algorithm is an effective and robust approach for predicting mine subsidence.

Analysis and Prediction of Urban Surface Transformation Based on Small Baseline Subset Interferometric Synthetic Aperture Radar and Sparrow Search Algorithm–Convolutional Neural Network–Long Short-Term Memory Model

“…LSTM is a kind of neural network with a unique time series processing ability in order to prevent the problem of gradient disappearance and the explosion of a recurrent neural network, the core of which lies in the forgetting gate, input gate, and output gate, which is able to make full use of the historical time series data and capture the temporal features in the data, as well as selectively retaining and forgetting the information to predict the future data more accurately [ 37 , 38 , 39 , 40 ]. The structure of its network model unit is shown in Figure 10 .…”

Urban Ground Subsidence Monitoring and Prediction Using Time-Series InSAR and Machine Learning Approaches: A Case Study of Tianjin, China

“…The Extremely Randomized Trees (ERT) algorithm extended from RF by introducing greater randomness in feature selection and node splitting thresholds (Shen et al 2023a). Lastly, the Long Short-Term Memory (LSTM) model, a deep learning approach based on recurrent neural networks, effectively captured long-term dependencies in sequential data through memory units and gating mechanisms(Chen et al 2021; Ma et al 2023). These models were rigorously validated through cross-validation techniques.…”

“…In recent years, machine learning and deep learning methods have emerged as promising approaches to overcome the limitations of traditional prediction methods, exhibiting vast potential in ground deformation forecasting (Li et al 2023; Wang et al 2023a; Xu et al 2023). Conventional machine learning techniques, such as Support Vector Machines (SVMs), Gradient Boosting Decision Trees (GBDTs), Random Forests (RFs), and Extremely Randomized Trees (ERTs), have demonstrated their effectiveness in ground deformation prediction (Dagès et Ma et al 2023). However, existing studies have primarily focused on evaluating single-scenario areas, lacking systematic comparative analyses across different urban scenarios (e.g., along metro lines, construction sites) and in-depth investigations into the potential in uencing factors (e.g., rainfall, groundwater extraction, construction activities) and underlying mechanisms.…”

Prompt determination of predictive parameters for mining-technogenic landscape objects

“…[20][21][22]. To solve the problems of constructing spatial digital relief models based on GIS-technologies, software packages such as Surpac, Surfer, AutoCAD Civil 3D, VRMesh and others in synthesis with aerospace and satellite data are widely used [23][24][25][26]. However, very often the processes of constructing digital models are labor-intensive, and the purchase of GIStechnology software packages requires additional material costs for scientists and engineers.…”