期刊文献+

直线负载模拟器及其新型CMAC改进算法控制

Linear Loading Simulator Control Based on Novel CMAC Improved Algorithm
下载PDF
导出
摘要 为检测某型直线阀是否能够达到姿/轨控发动机的推力要求,基于旋转电机加滚珠丝杠的方式,设计了一种直线负载模拟器(LLS)。为抑制多余力,提出一种基于电流环、位置环及力环的多闭环复合控制策略,并在力外环采用一种基于二维非均匀量化的新型CMAC改进算法,引入量化距离来确定高斯权重,并提出一种能够抑制"过学习"现象的新型权重学习方法。仿真及实验结果表明,该方法能够有效地抑制电动直线阀输出位移所带来多余力的干扰,明显地抑制了传统CMAC过学习现象,提高了LLS加载精度。 To test whether a linear valve can achieve the thrust requirement of the attitude/orbit control engine,the linear load simulator(LLS) was designed based on rotating motor and ball screw.To suppress the disturbance of surplus force,a hybrid control strategy with current loop,position loop and force loop is proposed.A novel CMAC improved algorithm based on two-dimensional nonuniform quantization is applied to the outer loop of force.The quantization distance is used to determine Gauss weights,and the learning method of CMAC weights is improved.The results show that this method can effectively suppress the disturbance of surplus force caused by electric valve linear's output displacement.And this method obviously restrain the over-learning phenomenon of traditional CMAC and improve the loading accuracy of LLS.
作者 徐志伟 范元勋 汪训浪 雷建杰 戴立 XU Zhi-wei;FAN Yuan-xun;WANG Xun-lang;LEI Jian-jie;DAI Li(School of Mechanical Engineering;School of Automation, Nanjing University of Science and Technolo- gy, Nanjing 210094, China)
出处 《组合机床与自动化加工技术》 北大核心 2018年第5期107-110,共4页 Modular Machine Tool & Automatic Manufacturing Technique
关键词 直线负载模拟器 电动直线阀 新型CMAC 非均匀量化 高斯权重系数 linear loading simulator electric linear servo valve novel CMAC non-unifrom quantization gaussian weighting coefficient
  • 相关文献

参考文献8

二级参考文献44

  • 1张伟,朱大奇,孔敏,李武朝.基于改进的CMAC神经网络与PID并行控制的研究[J].计算机测量与控制,2005,13(12):1359-1360. 被引量:16
  • 2胡汉金,王志良,王尚君,刘明.基于神经-模糊控制的交流伺服系统仿真研究[J].电气传动,1996,26(1):32-35. 被引量:6
  • 3刘媛,王卫红.基于CMAC的伺服系统控制研究[J].航空学报,2006,27(3):515-519. 被引量:11
  • 4Su Sh F, Tao T, Hung T H. Credit assigned CMAC and its application to online learning robust controllers [ J ]. IEEE Transactions on System, Man and Cybernetics, Part B, 2003, 33 (2) : 202 - 213.
  • 5SABOURIN C, BRUNEAU O. Robustness of the dynamic walk of a biped robot subjected to disturbing external forces by using CMAC neural networks [J]. Robotics and Autonomous Systems ,2005,51 (2-3) : 81-99.
  • 6SABOURIN C, MADAM K, BRUNEAU O. Autonomous biped gait pattern based on fuzzy- CMAC neural networks [ J ]. Integrated Computer-Aided Engineering ( ICAE ), 2007, 14(2) :173-186.
  • 7YUAN Z H, WU J D, TENG J H. Hybrid control of load simulator for unmanned aerial vehicle based on wavelet networks[C]//Proceedings of International Conference on Machine Learning and Cybernetics. New York: IEEE, 2003:715 - 719.
  • 8KU S S, LARSEN G, CETINKUNT S. Fast tool servo control for ultra-precision machining at extremely low feed rates[J]. Mechatron- ics, 1998, 8(4): 381 - 393.
  • 9LARSEN G A, CETINKUNT S, DONMEZ A. CAMC neural net- work control for high precision motion control in the presence of large friction[J]. Journal of Dynamic Systems, Measurement and Control, 1995, 117(3): 415- 420.
  • 10WANG M Y, GUO B. Design of electric dynamic load simulator based on recurrent neural networks[C]//Proceedings of IEEE Inter- national Electric Machines and Drives Conference. New York: IEEE, 2003:207 - 210.

共引文献107

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部