基于单神经元的永磁同步电机解耦控制

Decoupling control for permanent magnet synchronous motor based on single neuron

针对永磁同步电机(PMSM)磁场定向控制为代表的传统解耦策略难以实现高性能控制的问题,本文利用神经网络不依赖对象模型的特点以及出色的学习能力,提出了一种基于单神经元的永磁同步电机解耦控制策略.在传统磁场定向控制模型的基础上,构建了基于单神经元的永磁同步电机解耦控制系统,进行了仿真,并搭建以数字信号处理器为核心的电机控制实验平台上进行实验论证.结果表明,基于单神经元解耦的永磁同步电机控制系统具有快速响应能力,并且几乎达到无静差、无超调,实现了PMSM的高性能控制.

The conventional decoupling strategy for the magnetic field-oriented control of permanent magnet syn- chronous motors （PMSM） is difficult to achieve high performances. To deal with this problem, we put forward a decoupling control strategy based on the single neuron, by exploiting the independency of neural network on the object model as well as its excellent learning ability. On the basis of the model of traditional magnetic field-oriented control, we build a single neuron decoupling control system for the PMSM, and perform the simulation as well as the experimental tests on the plat- form with digital signal processor （DSP） as the core element. The experimental results show that the PMSM control system based on the single neuron decoupling has rapid response ability and is almost with no static error and overshoot. It realizes high performance in PMSM control.