摘要
针对传统PID控制器参数固定,处理非线性时变系统的局限性,提出神经元自适应PID控制算法。以传统PMSM转速控制为基础,采用神经元自适应PID控制方式,选用Hebb学习算法,以误差平方为性能指标实现参数在线调整,进而实现电机转速的快速跟踪。逆变器环节采用非正交坐标系下SVPWM算法,简化矢量算法步骤。仿真结果表明,对比传统PID控制方式,神经元自适应PID控制系统作用永磁同步电机,可以达到响应快、精度高的控制效果。
Since the traditional PID controller has fixed parameters,and is limited to deal with the nonlinear time-varying system,the neuron self-adaptive PID control algorithm is proposed. On the basis of the speed controller of traditional permanent magnet synchronous motor(PMSM),the neuron adaptive PID control method and Hebb learning algorithm are adopted,and the parameters online adjustment is realized by taking the square of error as performance index to fast trace the motor speed. The SVPWM algorithm under non-orthogonal coordinate system is adopted in the inverter to simplify the steps of vector algorithm.The simulation results show that,in combination with the traditional PID control method,the neuron adaptive PID as the control method of PMSM has the control effects of rapid response speed and high precision.
引文
[1]傅平,郭吉丰,丁敬,等.基于神经元自适应PID的超声波电机速度位置控制[J].电工技术学报,2007,22(2):28-33.
[2]邓赞高,吴必君,张峰.基于改进单神经元自适应PID励磁控制器[J].电力自动化设备,2006,26(11):61-63.
[3]丁文,高琳,梁得亮,等.永磁同步电机矢量控制系统的建模与仿真[J].微电机,2010,43(12):66-71.
[4]张宏宇,闫镔,陆利忠.永磁同步电机控制策略及应用研究综述[J].微电机,2008,41(4):69-73.
[5]李毓洲,罗玉涛,赵克刚.基于单神经元控制器的永磁同步电机矢量控制仿真[J].微电机,2007,40(3):5-8.
[6]张建峡,杨勇,徐德志,等.单神经元自适应PID控制交流调速系统[J].电机与控制学报,2007,11(2):130-133.
[7]丁军,徐用懋.单神经元自适应PID控制器及其应用[J].控制工程,2004,11(1):27-31.
[8]何罡,杨耕,窦曰轩.基于非标准正交基的空间电压矢量快速算法[J].电力电子技术,2003,37(6):1-3.