李东印, 王祖洸, 王伸,等.特厚煤层采放协调智能群组放煤工艺模型及关键技术研究[J].河南理工大学学报（自然科学版）,2025,44(3):22-33.

LI D Y, WANG Z G, WANG S, et al. Research on the intelligent coordinated top-coal caving model and key technologies in extra-thick coal seams[J]. Journal of Henan Polytechnic University(Natural Science) , 2025, 44(3): 22-33.

特厚煤层采放协调智能群组放煤工艺模型及关键技术研究

李东印1,2, 王祖洸1, 王伸1,2,3, 李化敏1, 彭维平4, 李红斌1,2, 张国澎5, 张旭和6

1.河南理工大学 能源科学与工程学院，河南 焦作  454000；2.煤炭安全生产与清洁高效利用省部共建协同创新中心，河南 焦作  454000；3.河南理工大学 地质资源与地质工程博士后流动站，河南 焦作  454000；4.河南理工大学 信息化建设与管理中心，河南 焦作  454000；5.河南理工大学 电气工程与自动化学院，河南 焦作  454000；6.郑州煤矿机械集团股份有限公司，河南 郑州  450000

摘要: 目的 综放开采是特厚煤层安全高效的开采方法，放煤环境复杂多变、采放协调难度大、放煤工艺及参数受控因素多，导致智能化综放开采尤其是智能放煤控制技术发展较为缓慢。针对特厚煤层综放面群组放煤下的智能化放煤控制难题，  方法 采用理论分析、实验室仿真试验、计算机控制软件研发、现场试验与示范等方法，以“智能群组放煤机理-采放协调控制理论及方法-放煤工艺决策模型-放煤软件现场应用”为主线展开深入研究。  结果 结果表明：提出多放煤口群组协同放煤方法，以连续群组放煤为例，揭示了多放煤口群组协同放煤下的顶煤放出体和煤岩分界面的空间关系；深入剖析综放工作面的采放协调关系及群组放煤逻辑，提出特厚煤层综放面采放协调智能放煤控制理论，构建了综放面采放时间、空间和运能协调控制模型和基于强化学习及隐马尔科夫随机场模型的群组放煤协同控制决策方法，为智能化放煤决策与精准控制提供可靠理论基础；深度融合地质信息与采放工艺参数，建立了综放面多源信息数据库，为智能决策及数字孪生提供数据基础。  结论 基于群组放煤条件下的采放协调特征，构建了智能放煤决策模型，自主研发智能放煤决策系统（软件），实现了智能放煤工作面年产1 500万t。

关键词:特厚煤层;综放开采;智能放煤;采放协调;群组放煤

doi:10.16186/j.cnki.1673-9787.2024060059

基金项目:国家重点研发计划项目（2018YFC0604502）；河南理工大学安全学科“双一流”创建国家级重点项目（AQ20240306）；河南省科技攻关计划项目（242102320210）

收稿日期:2024/06/28

修回日期:2025/03/17

出版日期:2025-04-18

Research on the intelligent coordinated top-coal caving model and key technologies in extra-thick coal seams

LI Dongyin1,2, WANG Zuguang1, WANG Shen1,2,3, LI Huamin1, PENG Weiping4, LI Hongbin1,2, ZHANG Guopeng5, ZHANG Xuhe6

1.School of Energy Science and Engineering， Henan Polytechnic University， Jiaozuo  454000， Henan， China；2.Collaborative Innovation Center of Coal Work Safety and Clean High Efficiency Utilization， Jiaozuo  454000， Henan， China；3.Postdoctoral Station of Geological Resources and Geological Engineering， Henan Polytechnic University， Jiaozuo  454000， Henan， China；4.Information Construction and Management Center， Henan Polytechnic University， Jiaozuo  454000， Henan， China；5.School of Electrical Engineering and Automation， Henan Polytechnic University， Jiaozuo  454000， Henan， China；6.Zhengzhou Coal Mining Machinery Group Co.， Ltd.， Zhengzhou  450000， Henan， China

Abstract: Objectives Fully mechanized top-coal caving is a safe and efficient mining method for extra-thick coal seams. However， the complexity of the caving environment， the challenges in coordinating mining and caving， and the numerous factors influencing caving parameters have hindered the development of intelligent fully mechanized top-coal caving， particularly in terms of automated coal caving control technology. Methods To address the challenges of intelligent coal caving control under group caving conditions in fully mechanized top-coal caving faces of extra-thick coal seams， this study employs theoretical analysis， laboratory simulation experiments， computer control software development， and field trials. The research follows the framework of “mechanism of intelligent group caving-theory and methods of mining-caving coordination control-decision-making model for coal caving processes-field application of coal caving software.”  Results The study proposes a multi-outlet group cooperative caving method. Taking the continuous group caving method as an example， the spatial relationship between the top-coal caving body and the coal-rock interface under group cooperative caving is revealed. By analyzing the coordination relationship between mining and caving and the logic of group caving in fully mechanized top-coal caving faces， this study establishes a theoretical framework for intelligent mining-caving coordination control in extra-thick coal seams. A coordinated control model for mining and caving in terms of time， space， and material transport capacity is developed， along with a group caving decision-making method based on reinforcement learning and the Hidden Markov Random Field （HMRF） model， providing a reliable theoretical foundation for intelligent caving decision-making and precise control. Furthermore， a multi-source information database for fully mechanized top-coal caving faces is established by integrating geological information with mining and caving process parameters， laying the groundwork for intelligent decision-making and digital twin applications.  Conclusions Based on the mining-caving coordination characteristics under group caving conditions， an intelligent caving decision-making model is developed， and an intelligent caving decision support system （software） is independently designed and implemented. Field application of this software has enabled an annual coal production capacity of 15 million tons in intelligent fully mechanized top-coal caving faces.

Key words: extra-thick coal seam; fully mechanized top-coal caving; intelligent coal caving;mining-caving coordination; group caving