多信息融合的采煤机智能控制策略及其关键技术研究

项目名称： 多信息融合的采煤机智能控制策略及其关键技术研究

项目编号： No.U1504503

项目类型： 联合基金项目

立项/批准年度： 2016

项目学科： 机械、仪表工业

项目作者： 刘俊利

作者单位： 河南理工大学

项目金额： 27万元

中文摘要： 项目针对采煤机自动控制中的多源信号识别、煤岩分界、截割路径优化等难题，在通过相似实验完成采煤机截割不同材料振动特性分析的基础上，以匀质煤层工作面为研究对象，提出一种多元信息融合的采煤机姿态自动控制方法。研究中，利用独立分量分析法识别采煤机工作时振动、噪音信息，结合当前采煤机工作参数，提出一种基于自适应模糊神经推理系统(ANFIS)的采煤机截割振动预测模型的间接煤岩识别反演推理方法；采用图像识别技术对其截割痕迹图像进行边缘识别，确定当前截割路径上的煤岩分界线,为下一步序截割路径优化提供依据；利用当前步序下采煤机运行姿态、煤岩分界信息及设备高效、连续运行条件，融合出采煤机下一步序行走姿态与截割轨迹优化控制策略，达到采煤机姿态自动控制目的，为建立自动化综采工作面提供新的思路，为实现数字化矿山奠定基础。

中文关键词： 动态工艺规划；采煤机；姿态控制；煤岩识别

英文摘要： To solve the technological problems in the automatic shearer control that include multi-source signal identification, coal-rock interface recognition, cutting path optimization and so on.The automation control method of the shearer's running posture with multi-information fusion is studied on the basis of the similar model experiments about shear's vibration characteristics with different materials being cut. Firstly, it presents a method to recognize the coal-rock interface with the vibration prediction model of the shearer cutting based on the adaptive neuro fuzzy inference system (ANFIS) in the working face with the homogeneous coal seam. In the model, the Independent Component Analysis is used to identify the multi-source signals when the shearer is working. The image of cut-mark of coal-rock interface is analyzed by means of the Image Edge Detection.So the next control parameters are obtain on the basis of the information of the shearer's running posture, the coal-rock interface and the condition of machine sequences working, and finally the automatic control method of the shearer's running posture with multi-information fusion is obtained,which provides a theoretical basis for the building of digitization mines. 

英文关键词： dynamic process plan;coal cutter;attitude control;identification of coal and rock