基于数据驱动的综采装备协同控制系统架构及关键技术  

Collaborative control system architecture and key technologies of fully-mechanized mining equipment based on data drive

作　　者：路正雄 郭卫[1] 张帆[2] 张传伟[1] 赵栓峰[1] 杨满芝[1] 王渊[1] LU Zhengxiong;GUO Wei;ZHANG Fan;ZHANG Chuanwei;ZHAO Shuanfeng;YANG Manzhi;WANG Yuan(School cf Mechanical Engineering,Xi'an University of Science and Technology,Xi'an 710054,China;School of Mechanical,Electrical&Information Engineering,China University of Mining and Technology-Beijing,Beijing 100083,China)

机构地区：[1]西安科技大学机械工程学院,陕西西安710054 [2]中国矿业大学(北京)机电与信息工程学院,北京100083

出　　处：《煤炭科学技术》2020年第7期195-205,共11页Coal Science and Technology

基　　金：国家重点研发计划资助项目(2017YFC0804310);国家自然科学基金资助项目(51805428)。

摘　　要：为解决智能化综采工作面关键技术难题之一的开采设备协同控制问题,提出了基于数据驱动的综采装备仿人智能协同控制模型,重点研究了大数据背景下智能综采装备的协同控制知识自学习、开采行为自决策、分布协同自运行等关键技术理论与方法。具体包括:从数据应用的角度分析了智能综采系统的数据特点,阐明了智能综采的3大数据化特征:泛在感知(数据获取)、信息融合(数据挖掘)、智能控制(数据决策);构建了面向经验操作员决策过程表征的综采装备协同控制框架;提出了基于扩展有限状态机的综采装备运行状态演化方法和基于多标记决策信息系统的综采装备运动行为模式学习方法,来实现数据驱动下智能综采装备行为决策知识的获取;提出了面向经典采煤工艺过程的综采装备行为模态类的决策知识划分方法和基于CBR与RBR融合的决策行为混合推理方法,来实现智能综采装备动作行为的自主决策;探讨了人工控制模式下综采装备驾驶员控制策略的表征方法,发展了具有自学习、自决策、工况自适应的综采"三机"仿人智能协同控制方法;给出了基于平行系统理论的平行综采技术逻辑,为综采装备协同控制的研究提供方法。所提综采装备协同控制系统为大数据背景下的综采生产系统的协同控制提供了解决方案。In order to solve the problem of cooperative control of mining equipment,which is one of the key technical problems in intelligent fully-mechanized coal mining face,a human-like intelligent cooperative control model of fully-mechanized coal mining equipment based on data drive was proposed,focusing on the collaborative control of intelligent fully-mechanized mining equipment under the background of big data key technology theories and methods such as knowledge self-learning,mining behavior self-decision,and distributed cooperative self-operation.It includes:analyzing the data characteristics of the intelligent fully-mechanized mining system from the perspective of data application,and clarifying the three data characteristics of the intelligent fully-mechanized mining system:ubiquitous perception(data acquisition),information fusion(data mining)and intelligent control(data decision-making);the cooperative control framework of fullymechanized mining equipment oriented to the representation of decision-making process of experienced operators was constructed.A method for the evolution of fully-mechanized mining equipment operating state based on extended finite state machine and a learning method for the movement behavior mode of fully-mechanized mining equipment based on multi-marker decision information system was proposed to realize dynamic acquisition of behavior decision knowledge of intelligent fully-mechanized mining equipment driven by data.This paper also studied the decision knowledge partition method for the behavior mode class of fully-mechanized mining equipment and the decision behavior hybrid reasoning method based on CBR and RBR fusion to realize the autonomous decision-making of the intelligent fully-mechanized mining equipment’s behavior.The characterization method of the driver’s control strategy of fully-mechanized mining equipment under the manual control mode was discussed,and the"Three Machines"simulation of fully-mechanized mining with self-learning,selfdecision and self-adaptation of wor

关 键 词：数据驱动 智能综采工作面 协同控制 仿人智能控制 平行综采 

分 类 号：TD421[矿业工程—矿山机电]