냉동시스템의 슬라이딩 모드 제어와 성능저하 옵저버 설계

Abstract

Fault is an unpermitted deviation of at least one characteristic or parameter of the system from the acceptable. Effect of fault on system usually coupled with effect of disturbance. So it is difficult to observe fault when fault and disturbance simultaneously exist on system. I designed fault diagnosis observer to estimate fault size on system which corrupted by both disturbance and fault. H infinity optimization and LMI design were used to design fault diagnosis observer. Refrigeration plant was used as target system. To concentrate on fault diagnosis, I designed robust controller which can tolerate certain size of fault and disturbance. Super-twisting algorithm based second-order sliding mode controller was used for this robust controller. The simulation results show estimated fault tracks the size of real fault.