An integrated state space partition and optimal control method of multi-model for nonlinear systems with state estimation

Abstract

Linear analysis and control system are more popular and simple, which leads to the popularity of dealing with complex systems by multiple model approaches. To answer the question that how to obtain the optimal boundaries between the ranges where the submodels are active, a systematic method of state space partition based on closed-loop systems by hybrid systems optimal control theory is utilized. Additionally, a MLD-based MPC controller with corresponding optimal control problem is designed. In order to address the puzzle brought by unavailable states, this paper also develops an adaptive EKF estimator for the PWA model with fictitious noises. By applying this whole strategy to a CSTR process, the simulation demonstrates a satisfactory result.