采煤机扭矩轴扭断参数的神经网络算法辨识  

Neural network algorithm identification of twisting parameters of shearer torque axis

作　　者：万丰[1] 蔡桂英[2] Wan Feng Cai Guiying(School of Mechanical Engineering, Heilongjiang University of Science ＆ Technology, Harbin 150022, China School of Computer ＆ Information Engineering, Heilongjiang University of Science ＆ Technology, Harbin 150022, China)

机构地区：[1]黑龙江科技大学机械工程学院,哈尔滨150022 [2]黑龙江科技大学计算机与信息工程学院,哈尔滨150022

出　　处：《黑龙江科技大学学报》2017年第2期118-122,共5页Journal of Heilongjiang University of Science And Technology

摘　　要：针对实际工况下采煤机扭矩轴工作性能不稳定的问题,提出基于Hopfield神经网络的辨识算法来优化扭矩轴参数。利用该算法拟合采煤机扭矩轴的实验数据,并以该数据为样本,辨识采煤机扭矩轴参数,应用MATLAB软件比较分析实验与仿真的误差。结果表明,基于Hopfield神经网络的辨识算法对采煤机扭矩轴传动系统建模工作量小、通用性好,且仿真误差小于1 N·m,可应用于实际工程。This paper seeks to overcome the unstable performance of shaft torque of the coal winning machines operating in the actual working conditions and presents an identification algorithm based on Hopfield neural network to optimize the torque shaft parameters. The study using this algorithm involves fit- ting the experimental data of coal winning machine torque shaft; identifying coal winning machine shaft torque parameters using the data as sample; and comparing and analyzing errors occurring between the experiment and simulation by applying MATLAB software. The results show that the algorithm promises a practical engineering application thanks to its advantages, such as a smaller modeling workload in coal winning machine torque shaft driving system, a better versatility, and simulation error of less than 1 N · m

关 键 词：采煤机 扭矩轴 HOPFIELD神经网络 刚度 系统辨识 

分 类 号：TD421.6[矿业工程—矿山机电]