Study On The Application Of Genetic Algorithms In The Induction Motor Speed Sensor-less Vector Control

During the past years, speed sensor-less vector control systems have been becoming one of the most important fields in the research of AC drives. It's an task for Chinese researchers and industries in Motor Control fields to improve the theory research level of speed sensor-less AC drive control system and industrialize it quickly. This thesis is one attempt of this field.PID Controller has been widely used in most of control systems, however, it is still a very important problem how to determine or tune its parameters, because these parameters have a great influence on the stability and the performance of the control system. On the other hand, in the nearly twenty years, GA(Genetic Algorithms) is developing fast as one method which gives us optimal answers for research, optimization and machine learning problems. Based on GA, using the related literatures and experiences of other researchers for reference, the thesis brings forward a new PI parameters algorithm to fulfill the PI controller of the speed sensor-less vector control system tuning.The thesis aims to study the induction motor speed sensor-less vector control system. Firstly it discusses the fundamental principle of the control of the induction motor, and obtains the mathematical models of the α-β and M-T shafting. The theoretical analysis of the vector control has been carried out, and the novel speed estimated mathematical model using Kalman Filter has been carefully deduced and the simulation proves that the method is effective and practical. Secondly, more research is done on GA theory, standard GA computing process and the program design of GA operator. Finally the research states, specialties, principles and common methods of PID controller parameters tuning are introduced simply. Based on the control law of the PI controller, Using the speed estimated by Kalman Filter, PI optimized tuning strategy is obtained with GA. Some searching methods of PI parametersusing GA have also been reported, while the thesis pays attention to improving searching efficiency. The control law is simulated with Matlab in two models and simulated with the speed sensor-less vector control system. The results of the simulation are analyzed and satisfactory.