基于单神经元PI的船用异步电机无速度传感器仿真

针对船用异步电机转速辨识不精确而导致的控制系统不稳定问题,使用基于模型参考自适应系统的无速度传感器技术取代传统的转速检测装置对转速信号进行估测。根据神经网络的自学习和自适应能力,设计了一种单神经元自适应PI调节器以实现控制参数的在线自整定,克服了传统PI控制器在非线性和时变系统中控制能力差的缺陷。结合异步电机矢量控制技术,使用基于转子磁链模型的转速辨识算法对电机转速进行估测,将单神经元PI控制器应用于转速估测系统,实现对电机转速的精确估计。仿真结果表明,相对于传统控制器,设计的单神经元自适应PI调节器响应速度快,参数自调节性能强,转速估测更为精确。

Sensorless Speed Estimation Study of Marine Asynchronous Motor Based on Single Neuron PI

In order to solve the problem of control system instability caused by inaccurate speed identification of Marine asynchronous motor, the sensorless speed estimation technology based on model reference adaptive system is used to replace the traditional speed detection device to estimate the speed signal. According to the self-learning and self-adaptive ability of neural network, a single-neuron adaptive PI regulator is designed to realize on-line self-tuning of control parameters, which avoids the defects of traditional PI controllers with poor control ability in nonlinear and time-varying systems. Combined with the vector control technology of asynchronous motor, the speed identification algorithm based on rotor flux model is used to estimate the motor speed, and the single neuron PI controller is applied to the speed estimation system to realize the accurate estimation of motor speed. The simulation results show that compared with the traditional controller, the designed single neuron adaptive PI regulator has fast response speed, strong self-adjusting performance and more accurate speed estimation.