神经网络在永磁同步电机变频调速系统中的应用

Application of artificial neural network(ANN) in variable frequency speed control system of PMSM

永磁同步电机调速系统被广泛应用于工业、农业等各个领域。为了能够减少能源的消耗,节约成本,须在工业设计和控制技术方面实现高效率、高节能的要求。采用基于空间矢量脉宽调制(SVPWM)策略的矢量控制方案能够使系统的调速范围较宽,精度较高,抗扰动能力强,得到较好的控制效果,但其超调量较大,而利用神经网络的精准建模特性可有效解决该问题。采用MATLAB/SIMULINK仿真软件设计了转速、电流双闭环SVPWM型交流变频调速系统,对永磁同步电机运用了id=0的矢量控制方法。提出在转速控制器的设计中,用单神经元控制器取代传统的PI控制器。通过调节控制器参数并进行仿真,验证了系统模型的合理性,有效地提高了系统的动态性能,改善了系统的稳定性。

The speed control system of the permanent magnet synchronous motor （PMSM） has been widely used in industry, agriculture and many other fields. In order to reduce energy consumption and save cost effectively, it is necessary to accomplish the requirements of high efficiency and high energy saving on the aspects of industrial de- sign and control technology. The vector control method based on the space vector pulse width modulation （SVP- VIM） strategy provides wide speed range, high precision and good ability to resist disturbance for the system. The control effect is excellent, but its overshoot is large. The precise modeling property of the neural network is used to effectively solve this problem. This paper designed a kind of variable frequency speed control system based on SVP- WM that contains double closed-loop of speed and current by MATLAB/SIMULINK, and the vector control method of ia = 0 was used to control the PMSM.In this study,we adopt single neuron controller to replace the traditional PI controller in the process of designing a speed controller. Moreover, the rationality of the system model was verified by adjusting the parameters of the controller, and the dynamic performance and stability of the system are improved.