论文部分内容阅读
电主轴内部的定子电阻在运行时会因受定子电流、温度和运行频率的影响而发生变化,这一变化会在采用直接转矩控制策略来控制电主轴的运行时影响直接转矩的控制效果。为减小这一影响,改善直接转矩的控制性能,借助Matlab神经网络工具箱设计了基于RBF的定子电阻辨识器对定子电阻进行辨识,并用Matlab进行了仿真测试。测试结果表明,采用RBF神经网络来对电主轴的定子电阻进行辨识是可行的,其能够在(-0.01～0.011)Ω误差范围内较为准确的观测定子电阻的阻值,辨识能力较强。
The stator resistance inside the spindle changes during operation due to stator current, temperature and operating frequency. This change will affect the control of direct torque when using direct torque control strategy to control the spindle operation . In order to reduce this influence and improve the control performance of direct torque, RBF-based stator resistance recognizer was designed based on Matlab neural network toolbox to identify the stator resistance and simulated by Matlab. The test results show that using RBF neural network to identify the stator resistance of the spindle is feasible. It can accurately measure the resistance of the stator resistance within the range of (-0.01 ~ 0.011) Ω error, and has a strong ability of identification.