RESUMEN
SBAS InSAR has long been used to monitor the mining surface deformation, and its research has been of great interest to researchers worldwide. For the unsatisfactory accuracy of SBAS InSAR-monitored mining surface deformation results, a new corrected model is proposed by integrating SBAS InSAR and Logistic Function. Firstly, the time series deformation results of the mining area were obtained by SBAS InSAR, and the variation law of the differences between SBAS InSAR- and leveling-monitored deformation values was statistically analyzed. Subsequently, the corrected model was constructed using the logistic linear regression analysis function and solved using the Levenberg-Marquardt algorithm. Finally, the corrected high-precision time series deformation results in the mining area were obtained. A mining area in Shandong Province of China was taken as the research object, and the practical application effect of the proposed corrected model was verified. Results showed that the Logistic Function could describe the variation law of the differences relatively accurately, and the corrected results were significantly better than the SBAS InSAR-monitored results, and the RMSEs of the corrected results were improved by 33-58%. The accuracy of SBAS InSAR-monitored mining surface deformation was effectively improved.
Asunto(s)
Algoritmos , Monitoreo del Ambiente , China , Factores de TiempoRESUMEN
In coal mining areas, surface subsidence poses significant risks to human life and property. Fortunately, surface subsidence caused by coal mining can be monitored and predicted by using various methods, e.g., probability integral method and deep learning (DL) methods. Although DL methods show promise in predicting subsidence, they often lack accuracy due to insufficient consideration of spatial correlation and temporal nonlinearity. Considering this issue, we propose a novel DL-based approach for predicting mining surface subsidence. Our method employs K-means clustering to partition spatial data, allowing the application of a gate recurrent unit (GRU) model to capture nonlinear relationships in subsidence time series within each partition. Optimization using snake optimization (SO) further enhances model accuracy globally. Validation shows our method outperforms traditional Long Short-Term Memory (LSTM) and GRU models, achieving 99.1% of sample pixels with less than 8 mm absolute error.