Multi-timescale Distributed Model Predictive Control for Large-Scale Systems and a Case Study

Abstract

Abstract To solve the control problem effectively of complicated large scale system with obvious difference in the dynamic response at each channel, a strategy based on multi-timescale and distributed communication mode is presented. These systems can be regarded as combinations of fast system and slow system, the response speeds of which are in two-time scale. The algorithm takes into account the fast and slow characteristics and the coupling relationship of each subsystem, uses the Nash optimal idea and the multi-time standard information prediction method to realize the optimization control of the whole system. A simulation example is given to illustrate the effectiveness.