|제목(국문)||유전자 알고리즘을 이용한 PID 기반 자기베어링 제어기 최적화|
|제목(영문)||Optimization of PID-Based Controller for Active Magnetic Bearings Using Genetic Algorithm *권은상(충남대학교), 박영우(충남대학교|
박영우 (Y. W. Park ,충남대학교 ) ▷공저자네트워크등록하기
노명규 (- ,충남대학교 ) ▷공저자네트워크등록하기
Although optimal controllers such as H-infinity and robust control are available for controlling active magnetic bearings, variants of proportional-integral-derivative control are widely used in industry because they are intuitive, easily implementable, and tuned in a straight forward manner. However, tuning control gains require significant resources in terms of time and expertise. Further, engineers are not sure whether the tuned controller is optimal or not. In this paper, we use a genetic algorithm to search an optimal set of control gains that produce minimal output sensitivity while maximizing the stifness of the bearing and maintaining the system stability. First, non-parametric transfer function of the magnetic bearing system (amplifier, bearing, rotor, and sensor all combined) is obtained through sine sweep tests. Then, system parameters are identified through the estimation of parametic transfer functions. Actuator gains, open-loop stability, amplifier bandwidths, and flexible modes are included in the identified parameters. The identified plant transfer function is then used to check the stability, while the maximum sensitiviy is obtained by applying the control algorithm to the experimentally obtained frequency response.
|keyword||Magnetic bearing, Genetic algorithm, Control optimization
|저널명||한국정밀공학회 추계학술대회 ▷관련저널보기|