Design of modified model-based adaptive control system for FOPDT processes

Abstract

Adaptive controllers are becoming popular for industrial closed-loop control applications, especially for various chemical processes. By incorporating adaptation mechanism, controller acquires the ability to cope with varying process dynamics. Model reference adaptive system (MRAS) is one of the most popular adaptive control techniques adopted in process industries. The most important feature of MRAS is that, it requires only one parameter i.e. adaptation gain and depending on its choice controller tuning changes according to the well-known MIT-rule. Here, we propose a modified version of MRAS based gain adaptation scheme for designing a model-based controller and a guideline is also provided for choice of an appropriate reference model. Efficacy of the proposed method is verified through simulation study for three well-known first-order-plus-dead-time (FOPDT) models with dead-time to time constant ratio (θ/τ) in the range of 0.01–1. Superiority of the proposed scheme is verified through performance based comparative study with conventional MRAS and reputed K-SIMC technique by Lee et al [11]. In addition, generalized stability of the proposed controller is also tested by Liapunov method.