摘要
为提高PMSM(Permanent Magnet Synchronous Motor,永磁同步电机)伺服系统的控制水平,本文提出一种基于RBF(Radial Basis Function,径向基函数)神经网络和单神经元PID的PMSM速度控制器,可根据PMSM伺服系统的实时状态进行速度控制器内部参数自整定,由于具备这种能力,使PMSM伺服系统的速度环实现全局最优控制。实际中,工业过程控制对平稳度要求较高,针对此要求对单神经元PID算法进行改进,将固定比例系数改进为跟随系统误差变化的变比例系数,这种改进有效的减少了系统超调,使控制更加平稳。
In order to improve the control performance of PMSM(Permanent Magnet Synchronous Motor)servo system, the paper proposes a PMSM speed controller based on RBF(Radial Basis Function)neural network and single neuron PID, which realizes self-tuning of the speed controller' s internal parameters in accordance with the real-time status of PMSM servo system.Due to this ability, it makes the system' s speed loop realize the optimal control on the whole.In reality, industrial process control has a higher requirement for smoothness. Therefore, improvement is made to the single neuron PID algorithm by modifying the fixed scale factor into a variable proportional coefficient changing with system error, which effectively reduces the system overshoot and thereby makes the control smoother.
出处
《电子世界》
2017年第2期122-126,共5页
Electronics World