The design of NMSS fractional-order predictive functional controller for unstable systems with time delay

Abstract

Open-loop unstable systems are more difficult to control than stable processes. In presence of time delay and uncertainty, the complexity of problem increases. In this paper, non-minimal state space predictive functional control (NMSS-PFC) infrastructure has been generalized for control of unstable systems with time delay. At first, NMSS system representation has been extended for a so-called coprime-factorized equivalent model of the unstable processes. Then, the proposed NMSS-fractional-order PFC (NMSS-FOPFC) has been formulated via a fractional order cost function in terms of the output tracking error vector. In the developed formulation and via the NMSS structure, the constraints on inputs of the system could be easily formulated and employment of fractional order cost function led to improved performance even in case of disturbances, perturbations or uncertainties. Closed loop robust stability analysis was also performed based on small gain theorem and verified via simulations. Simulation examples show that the proposed NMSS-FOPFC results in improved nominal and perturbed responses compared to conventional methods. Comparison has been carried out by considering integral of absolute error (IAE) and integral of squared error (ISE) as well as step response transient and steady state properties and the control effort.

Keywords: Coprime-factorization; Fractional-order predictive functional control; Non-minimal state space; Robust stability; Unstable systems.