【摘要】 为解决感应电机无速度传感器矢量控制系统的转速辨识问题,在给定的无速度传感器感应电机间接矢量控制系统中,根据感应电机的数学模型,经过一定的变换,利用电机易于检测到的定子电压和电流,以及基于BP算法的两层神经网络,用期望状态与实际状态之间的偏差来调整神经网络模型的权值,达到实时辨识电机转速的目的。该方法简单、直观,不仅利用了神经网络的优点,又能适应感应电机调速系统实时控制的要求。仿真结果验证了该方法的有效性。更多还原

【Abstract】 In order to get the identified speed of speed sensorless induction motor, an indirect vector control strategy of speed sensorless induction motor is given in this paper. According to transformed form of a mathematical model of induction motor, the easily - measured motor stator voltage and current and the two - layered back - propagation neural network technique are used to provide a real time speed estimation of the induction motor. The error between the desired state variable and the actual state variable is back propagated to adjust the weight of the neural model, and then the speed can be identified exactly. The algorithm is simple and intuitionistic. It can not only take advantage of neural networks, but also fit the demands of the real time control of the induction motor. The effectiveness of the system has been verified by simulation.更多还原