采煤机摇臂振动信号分析及其截割模式识别方法研究

发布时间：2019-01-19 13:45

【摘要】：采煤机是实现煤矿安全高效生产的关键设备之一,作为综采成套装备的主要组成部分,其智能化水平是实现综采工作面“无人化”或“少人化”关键因素,截割模式的准确识别是实现采煤机智能开采的前提,而采煤机摇臂的振动信号能够直接反映采煤机的截割状态。因此,有必要对采煤机摇臂振动信号及截割模式进行深入研究,为实现采煤机的自动截割和自适应控制奠定基础。实际工况中采煤环境极其恶劣,采煤机在运行过程中其摇臂同时受到截割煤壁、机身姿态突变、牵引速度波动等外界作用的干扰,表现为一种非线性复杂带噪信号。本文以采煤机摇臂复杂振动信号为研究对象,研究在不同时间尺度下的特征向量提取方法,建立采煤机截割模式分类模型,基于改进支持向量机实现不同截割模式的识别。论文的主要工作及研究成果如下:(1)在分析采煤机基本结构及工作过程的基础上,研究了采煤机摇臂振动信号变化机理,论述了利用加速度信号进行截割模式识别的可行性,给出了不同顶、底板和煤层特性下的截割模式类别。(2)针对摇臂复杂振动信号信噪比、虚假分量和特征维数存在的问题,利用多阈值小波包对不同频段信号分别进行去噪处理;基于K.L散度剔除EMD分解过程中出现的虚假分量,并结合拉普拉斯分值,实现了振动信号的多尺度模糊熵特征提取。(3)为提高采煤机截割模式识别精度,提出了基于改进支持向量机的截割模式分类方法,研究了一种人工鱼群与粒子群融合的优化算法,实现了支持向量机核参数、惩罚因子的优化,并在此基础上设计了采煤机截割模式识别系统框架及实现流程。(4)搭建了摇臂振动信号采集系统,并在张家口煤矿机械有限公司国家能源采掘装备研发实验中心进行了地面实验。实验结果表明,基于融合算法改进的SVM截割模式识别精度为98.86%,高于人工鱼群改进SVM的97.15%和粒子群改进SVM的97.71%,验证了所提方法的正确性和有效性。
[Abstract]:The shearer is one of the key equipments to realize the safe and efficient production of coal mine. As the main part of the complete set of equipment for fully mechanized coal mining, its intelligent level is the key factor to realize "no man" or "less person" in the fully mechanized mining face. The accurate recognition of cutting pattern is the premise of realizing intelligent mining of shearer, and the vibration signal of the rocker arm of shearer can directly reflect the cutting state of shearer. Therefore, it is necessary to deeply study the vibration signal and cutting mode of the rocker arm of the shearer, so as to lay a foundation for the automatic cutting and adaptive control of the shearer. In the actual working conditions, the coal mining environment is extremely bad, and the rocker arm of the shearer is disturbed by the external actions such as cutting the coal wall, the fuselage attitude abrupt change, the traction speed fluctuation and so on, which is a kind of nonlinear complex noise signal. In this paper, the complex vibration signal of shearer rocker arm is taken as the research object, the feature vector extraction method under different time scales is studied, the classification model of cutting pattern of shearer is established, and the recognition of different cutting pattern is realized based on improved support vector machine. The main work and research results are as follows: (1) based on the analysis of the basic structure and working process of the shearer, the mechanism of vibration signal variation of the rocker arm of the shearer is studied. The feasibility of cutting pattern recognition by acceleration signal is discussed, and the cutting mode categories under different roof, floor and coal seam characteristics are given. (2) aiming at the signal-to-noise ratio of rocker arm complex vibration signal, For the problems of false component and feature dimension, the multi-threshold wavelet packet is used to Denoise the signals in different frequency bands. Based on K.L divergence, the false components in EMD decomposition process are eliminated, and the multi-scale fuzzy entropy feature extraction of vibration signal is realized by combining Laplace score. (3) in order to improve the accuracy of cutting pattern recognition of shearer, A cutting pattern classification method based on improved support vector machine (SVM) is proposed, and an optimization algorithm based on the fusion of artificial fish swarm and particle swarm is studied. The kernel parameters and penalty factors of SVM are optimized. On the basis of this, the frame and realization flow of shearer cutting pattern recognition system are designed. (4) the vibration signal acquisition system of rocker arm is built. The ground experiment was carried out at the National Energy Extractive equipment Research and Development Center of Zhangjiakou Coal Mine Machinery Co., Ltd. The experimental results show that the accuracy of SVM cutting pattern recognition based on the improved fusion algorithm is 98.86%, which is higher than that of artificial fish swarm improved SVM and particle swarm improved SVM 97.71%. The correctness and effectiveness of the proposed method are verified.
【学位授予单位】：中国矿业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TD421.6

【参考文献】