Prediction of mine water inflow under water-inrush conditions based on a LSTM-Transformer model

Li Zhenhua; Jiang Yufei; Du Feng; Wang Wenqiang

doi:10.16186/j.cnki.1673-9787.2025080014

Abstract：Objectives Accurate prediction of mine water inflow is crucial for preventing water hazard acci-dents and ensuring safe production.This study aims to construct a water inflow prediction model suitable for mines in North China-type coalfields affected by water hazards from the underlying L1-4 limestone aqui-fer and Ordovician limestone aquifer.Methods Based on hydrogeological monitoring data from a typical coal mine in Henan Province,a coupled LSTM-Transformer model was proposed.The LSTM component cap-tures the dynamic temporal characteristics of mine water inflow,while the multi-head attention mechanism of the Transformer analyzes the complex temporal correlation between aquifer water level variations and mine water inflow.This framework enables accurate prediction of mine water inflow driven by dynamic wa-ter level changes.Results The coupled LSTM-Transformer model significantly outperformed LSTM,CNN,Transformer,and CNN-LSTM models in prediction accuracy,with a root mean square error(RMSE)of 20.91 m3/h,mean absolute error(MAE)of 16.08 m3/h,and mean absolute percentage error(MAPE)of 1.12%.Furthermore,compared to the single-factor water inflow prediction model,the two-factor(water level and water inflow)prediction model showed greater stability.Conclusions The LSTM-Transformer coupled model successfully overcomes the limitations of traditional methods in capturing the dynamic water level-discharge relationship within complex hydrogeological systems.It provides a solution for dynamic mine water inflow prediction with strong interpretability and robustness,offering a novel methodology for predict-ing water inflow under similar geological conditions.