基于单神经元控制器的异步电动机矢量控制  

The Vector Control of Induction Motor Drive Based on Single Neuron

作　　者：付华[1] 冯爱伟[1] 徐耀松[1] 王传英[1] 孟宪敬[1] 

机构地区：[1]辽宁工程技术大学,辽宁省阜新市123000

出　　处：《中国电机工程学报》2006年第1期127-131,共5页Proceedings of the CSEE

基　　金：辽宁省教育厅教育基金项目(C-011)

摘　　要：文中提出了采用单神经元智能控制器代替传统PID控制器以改善异步电动机矢量控制的性能。在分析单神经元控制器结构与控制原理的基础上,为了提高单神经元控制器的学习能力与自适应性,将无监督的Hebb学习规则与有监督的Delta学习规则相结合,运用改进的学习与控制算法,实现单神经元控制器的参数优化与在线自动调整。采用Matlab软件建立单神经元控制器与异步电动机矢量控制模型,进行仿真研究;并将单神经元控制器的控制软件应用于异步电动机矢量系统,进行实验研究。仿真与实验结果表明,单神经元控制器可以改善异步电动机矢量控制的性能,具有较强的自适应性与鲁棒性。Presented substituting an intelligent single neuron controller for the traditional PiED controller to improve the Field Oriented Vector Control performance of induction motor. The supervisory Delta study rule was combined with non-supervisory Hebb study rule and the improved study and control algorithms were adopted to achieve the parameter optimum and automatic adjustment on line in order to inhance the self-study capability and self-adaptivity of the single neuron controller on the basis of analyzing the construction and control principle of the single neuron controller. MATLAB software was used to build the neuron controller model and induction motor vector control model to carry on emulation research ,and the control software of the single neuron controller was used in the induction motor vector control system to carry on experiment research. The emulation and experiment results Show that the single neuron controller can improve the Field Oriented Vector Control property of induction motor and has stronger self-adaptability and robustness.

关 键 词：PID控制器 单神经元 异步电动机 矢量控制 自适应性 鲁棒性 

分 类 号：TM921[电气工程—电力电子与电力传动]