摘要：

煤矿开采安全问题尤其是采空区地表沉降现象会对人员安全及工程安全造成威胁, 研究合适的矿区地表沉降预测方法具有很大意义。矿区地表沉降影响因素复杂, 单一的深度学习模型对矿区地表沉降数据拟合效果差且现有的地表沉降预测研究多是单独进行概率预测或考虑时序特性进行点预测, 难以在考虑数据的时序特征的同时对其随机性进行定量描述。针对此问题, 在对数据本身性质进行观察分析后选择差分整合移动平均自回归(ARIMA)模型进行时序特征的概率预测, 结合长短时记忆(LSTM)网络模型来学习复杂的且具有长期依赖性的非线性时序特征。提出基于ARIMA-LSTM的地表沉降预测模型, 利用ARIMA模型对数据的时序线性部分进行预测, 并将ARIMA模型预测的残差数据辅助LSTM模型训练, 在考虑时序特征的同时对数据的随机性进行描述。研究结果表明, 相较于单独采用ARIMA或LSTM模型, 该方法具有更高的预测精度(MSE为0.262 87, MAE为0.408 15, RMSE为0.512 71)。进一步的对比结果显示, 预测结果与雷达卫星影像数据(经SBAS-INSAR处理后)趋势一致, 证实了该方法的有效性。

关键词: 煤矿采空区, 地表沉降预测, 时序概率预测, 差分整合移动平均自回归, 长短时记忆网络

Abstract:

Safety concerns in coal mining, particularly surface subsidence in goaf areas, threaten personnel and engineering safety. Investigating appropriate prediction methods for surface subsidence in mining areas is important. The factors influencing surface subsidence in mining areas are complex. Single deep learning models exhibit poor fitting performance for subsidence data, and existing subsidence prediction studies often focus on either probabilistic or point predictions that consider temporal characteristics. Describing randomness quantitatively while simultaneously considering the temporal features of the data is challenging. To address this issue, after observing and analyzing the nature of the data, we select the Autoregressive Integrated Moving Average (ARIMA) model for probabilistic prediction of temporal features. This is combined with a Long Short-Term Memory (LSTM) model to learn complex and long-term dependent nonlinear temporal sequences. We propose an ARIMA-LSTM based surface subsidence prediction model. The ARIMA model predicts the temporal linear part of the data, and the residual data predicted by the ARIMA model assists in training the LSTM model. This approach allows for the consideration of both temporal features and the randomness of the data. Experimental results demonstrate that the proposed method achieves superior prediction accuracy compared to ARIMA or LSTM models alone. Specifically, the Mean Square Error (MSE) is 0.262 87, Mean Absolute Error (MAE) is 0.408 15, and Root Mean Square Error (RMSE) is 0.512 71. Further comparative validation indicates that the predicted results align with trends observed in radar satellite imagery data processed using a Small Baseline Subset Interferometric Synthetic Aperture Radar (SBAS-INSAR), confirming the effectiveness of the proposed approach.

Key words: coal mining subsidence area, surface subsidence prediction, temporal probabilistic forecasting, Autoregressive Integrated Moving Average (ARIMA), Long Short-Term Memory (LSTM) network