论文部分内容阅读
针对传统的异步电机直接转矩控制在电机低速运行时系统性能受定子电阻变化影响较大的问题,详细分析了定子电阻变化对系统控制性能的影响,提出了一种基于RBF神经网络的定子电阻辨识方法。该方法应用梯度算法训练RBF神经网络各参数。用该方法对定子电阻进行辨识,具有辨识精度高,响应迅速等优点。对该方法在基于Simulink仿真软件上进行仿真,并与BP神经网络对定子电阻辨识时进行比较。仿真结果表明,该方法优于BP神经网络,可以有效地提高直接转矩控制系统的低速运行性能。
Aiming at the problem that the traditional direct torque control of induction motor has a great influence on the stator resistance when the motor runs at low speed, the influence of stator resistance change on the control performance of the system is analyzed in detail. A stator resistance based on RBF neural network Identification method. The method uses gradient algorithm to train the parameters of RBF neural network. Using this method to identify the stator resistance, it has the advantages of high identification accuracy and rapid response. The method is simulated on the basis of Simulink simulation software and compared with BP neural network to identify stator resistance. The simulation results show that this method is superior to BP neural network, which can effectively improve the low speed performance of DTC system.