基于神经网络逆系统永磁同步电机解耦控制.pdfVIP

·1226 · 2008 全国博士生学术论坛电气工程论文集 基于神经网络逆系统永磁同步电机解耦控制 孙晓东 朱熀秋 江苏大学电气信息工程学院 江苏 镇江 212013 【摘 要】 针对永磁同步电机这一多变量、非线性、强耦合的控制对象，提出了一种基于神经网络在 线辨识的永磁同步电机逆系统解耦控制新方法。通过静态神经网络加积分器来构造永磁同 步电机的逆系统，并在实际运行中不断地修正神经网络权值，使其更精确地逼近逆系统。 将逆系统与永磁同步电机原系统复合成两个伪线性子系统，使永磁同步电机解耦成二阶线 性转速子系统和一阶线性磁链子系统，在此基础上，运用线性系统理论进行综合。仿真试 验表明这种控制策略能够实现永磁同步电机转速和定子磁链之间的动态解耦控制，并且系 统具有良好的动静态性能。 【关键词】 永磁同步电机 神经网络 逆系统 在线辨识 解耦控制 Decoupling Control of Permanent Magnet-type Synchronous Motor Based on Neural Network Inverse System Sun Xiaodong Zhu HuangQiu School of Electrical and Information Engineering ，Jiangsu University ，Zhenjiang 212013 ，Jiangsu ，China Abstract ：A novel approach of inverse system decoupling control based on RBF neural network on-line identification is proposed for permanent magnet-type synchronous motor （PMSM ），which is multi- variable ，nonlinear and coupled system. A static neural network and three integrators are used to construct the neural network inverse of PMSM ，and the connection value of neural network can be amended on-line in order to approach the inversion exactly. Consequently ，two pseudo-linear subsystems are completed by combining the neural network inverse system with PMSM. PMSM is decoupled into second-order linear speed subsystem and first-order linear flux subsystem. Then the linear system theory is used to design the closed-loop linear regulators to control each of the subsystems. Simulation results have showed that this kind of control strategy can realize dynamic decoupling control between speed and flux