摘要
针对综采面生产过程机理复杂、数学模型难以建立等问题,通过数据挖掘和深度学习技术,挖掘隐藏在数据中的规律,通过智能建模技术和多目标优化技术,根据矿井综合生产指标对工艺控制参数进行模拟,建立工艺参数优化模型、通过海量历史数据对模型训练,给出合理工艺参数优化控制策略情况预测,以提升工作面生产效能为目的,选择出优化的、合理的工艺控制参数,为降低生产成本和能耗、提高生产效率提供智能决策方案,为矿山工作人员提供辅助决策方法。
As it is difficult to establish mathematical model for the production process of fully mechanized coal mining face because of its complex mechanism, data mining and deep learning technology are applied to find the hidden laws in the data. Through intelligent modeling and multi-objective optimization, the process control parameters are simulated according to the comprehensive production index of the mine. The optimization model of process parameters is established, and trained with massive historical data. In order to improve the production efficiency of working face, the optimized and reasonable process control parameters are selected, thus to reduce the production cost and energy consumption, improve the production efficiency to provide intelligent decision-making scheme, offer auxiliary decision-making methods for mine staff.
作者
王挨荣
陈汉章
郭微
潘涛
赵洪泽
贾灵强
徐洪洋
WANG Ai-rong;CHEN Han-zhang;GUO Wei;PAN Tao;ZHAO Hong-ze;JIA Ling-qiang;XU Hong-yang(Shangwan Coal Mine,CHN Energy Shen dong Coal Group,Ordos 017010,China;Guoneng Wangxin Technology(Beijing)Co.,Ltd.,Beijing 100011,China;China University of Mining and Technology(Beijing),Beijing 100083,China)
出处
《煤炭工程》
北大核心
2022年第4期62-67,共6页
Coal Engineering
基金
国家重点研发计划重点专项资助项目(2017YFC0804300)。
关键词
综采工作面
多目标优化
工艺参数优化
遗传算法
深度学习
神经网络
fully mechanized mining face
multi-objective optimization
process parameter optimization
genetic algorithm
deep learning
neural network
作者简介
王挨荣(1979-),男,内蒙古鄂尔多斯人,高级工程师,国能神东煤炭集团上湾煤矿机电矿长,现主要从事煤矿机电自动化方面的研究管理工作,E-mail:airong.Wang@chnenergy.com.cn。
