采矿与岩层控制工程学报2026，Vol.8Issue(2)：284-298,15.DOI:10.13532/j.jmsce.cn10-1638/td.2025-1347
榆神府矿区多煤层重复采动覆岩裂隙带高度预测研究
Study on height prediction of overburden fracture zone under repeated mining of multi-coal seams in Yushenfu mining area
摘要
Abstract
The Yushenfu mining area is characterized by shallow coal seams,thin overlying bedrock,and thick loose layers,and most mines in this mining area involve repeated mining of multiple coal seams.Affected by multiple factors such as coal seam mining height and spacing,the spatial interaction of surrounding rock in the upper and lower stopes makes it challenging to accurately predict fracture zone height.In this paper,the fracture zone height under multi-coal seam repeated mining in typical coal mines in the Yushenfu mining area was taken as the research object,and the research methods of physical similarity simulation,theoretical analysis,and deep learning were used.First,the fracture development law under multi-coal seam repeated mining was analyzed.Subsequently,a multi-factor coupling nonlinear regression model was established to describe the relationship between the fracture zone height and key parameters,including coal seam mining height,spacing,burial depth,dip angle,working face length,and interval rock strength.On this basis,the prediction method of fracture zone height under multi-coal seam repeated mining based on the SSA-BP neural network was established,and its accuracy was verified.The results indicate that the fracture development under repeated mining in Ciyaota Coal Mine exhibits a three-stage characteristic,i.e.,localized slow growth,nonlinear rapid increase through interconnection,and dynamic stabilization.The ultimate height of the fracture zone reaches 139.0 m.The nonlinear regression model incorporating the coupled effects of coal seam mining height,interlayer spacing,strength of intervening rock strata,and working face length achieves an R2 value of 0.880,confirming these parameters as key influencing factors for the fracture zone height.Compared to predictions from traditional empirical formulas and the BP model,the SSA-BP model demonstrates reductions in MAPE values by 22.96％and 6.70％,respectively,and attains a low RMSE of 1.79,indicating superior stability.Validation at the 14205 working face of Zhonghui Funeng Coal Mine in the Yushenfu mining area shows a relative error of 1.3％between the predicted and measured heights,well below 5％.The study demonstrates strong generalizability for predicting the height of water-conducting fracture zones under multi-coal seam repeated mining in the Yushenfu mining area and provides valuable insights for water hazard prevention and control under such mining conditions.关键词
榆神府矿区/多煤层开采/裂隙带高度/非线性回归/SSA-BP神经网络Key words
Yushenfu mining area/multi-coal seam mining/fracture zone height/non-linear regression/SSA-BP neural network分类
矿业与冶金引用本文复制引用
王红伟,左开永,陈玉涛,董国良,李延军,焦建强,白金源,王力涛..榆神府矿区多煤层重复采动覆岩裂隙带高度预测研究[J].采矿与岩层控制工程学报,2026,8(2):284-298,15.基金项目
国家自然科学基金面上资助项目(52474149,51974230) （52474149,51974230）
陕西省杰出青年科学基金资助项目(2023-JC-JQ-42) （2023-JC-JQ-42）
