Expert robust decentralised controller for uncertain large-scale systems

An expert decentralised robust control algorithm is developed for continuous large-scale systems with dynamic interconnections and external disturbances, taking into consideration the possible structured variation in the system and input matrices. The proposed scheme has three parallel processes: the decentralised robust controller and observer, the overlapped observers for fault detection and, finally, the expert system. The expert system uses the differences between the same states as estimated by different overlapped observers to specify the failure position. Furthermore, it selects the appropriate control and observer gains that guarantee the overall system stability while responding in a satisfactory manner