基于神经网络逆系统的感应电机变频系统解耦控制

感应电机反馈线性化方法的优点之一 ,是在电机参数准确已知的前提下 ,可将感应电机数学模型解耦为相互独立的转速子系统和磁通子系统 但是电机参数是时变的 ,不易精确获得 ,为进一步完善反馈线性化方法 ,提高调速性能 ,本文应用神经网络逆系统控制理论 ,使用神经网络直接替代现有解耦控制方法中的对应逆系统模型 ,理论分析与实验的结果表明 :该方法成功地实现了转速与磁链的解耦 ,同时对感应电机变频调速系统负载的变化具有较强的抗扰性和鲁棒性

The Decoupling Control of AC Variable Frequency Motor System Based on Artificial Neural Network Inverse System Method

Induction motors can be decoupled into rotor speed and flux subsystems by the feedback linearing method if the parameters can be precisely obtained. But the parameters are so changeable that they are difficult to obtain. In order to improve the method and the speed regulation property as well, based on ANN (artificial neural network), a new method to obtain inverse system model is proposed to solve the problem. The results of analyzsis and experiment show that the proposed method can decouple the AC variable frequency control system, which is valid, robust and stiff.