Neural Network-based Adaptive Control In Ac Servo System Applied Research

This paper is to design a new type intelligent controller aiming at a servo system of some model rocket launcher -- nerve net self-adapting controller. Alternating current servo motor is the driver device in the servo system ,which works under hard circumstances like large inertia, variety load , and the disturb of the moment. The controller designed in this paper can satisfy the servo system need of speeding , accuracy and stability preferably.The paper adopts PC/104 modules when designing the hardware structure of the servo system. PC/104 modules have some advantages such as small size , combinated conveniently , working stably. A nerve net controller have been designed on the basis of the hardware design of the AC servo system and system model. The controller include two forward nerve net with three layers . One is designed to identify the AC servo system model and another is to control. The controller use special learning algorithm which to control system at the same time and improve its constringency pace through introducing new MFBP algorithm. The intelligent control algorithmic is programmed using C++ language after the controller have been designed . According to the results of the simulation and experiments, the nerve net self-adapting controller has the high quality of static-dynamic performance and restrain the infection to system performance under hard circumstances like large inertia variety load and the huge change of the torsion...