The Research On Intellectual Control Of AC Drive Based On Neural Networks

The thesis is chiefly contributed to the deep study of AC drive intellectual control based on neural networks. The AC motor drive has been one of the important areas in automation research. The performance of the AC motor drive has been highly improved recently. The thesis dissertates on the principles of vector control and field orientation detailedly at first. The vector control can improve the torque control performance of the AC motor drive, but due to some uncertain reasons during the control process, the controller's dynamic performance of the AC drive will become bad. The traditional PID controller can not satisfy the requests of high-performance control, such as fast dynamic response and accurate speed or torque control and robustness. The intelligent control based on neural networks offer an effective control method for high-performance AC drive systems. Therefore the combination of intelligent control and vector control has become a hotspot. Neural network has been applied in many fields, but it is under the research how to use neural network controller effectively in the drive systems. So it is very meaningful in the research of these aspects. On the basis of analyzing in depth the principle of the neural network, the paper has researched on the applications of neural network in the AC motor drive. Two kinds of neural network speed controllers have been studied. One is the single-neuron adaptive PID controller. The other is the hybrid neural network controller. In order to test the performance of the drive systems based on neural networks, an induction motor speed drive system of field-oriented control with the neural network controller has been designed for simulation. At the same time, the neural network strategy of vector control systems is digitally realized in TMS320LF2407A DSP controller. The results show that the performance is very good in the drive systems based on neural network.