Direct controller designs from plant data

Abstract

Model-based PID and IMC designs are popular control strategies. However, model-based methods require the availability of a process model, the identification of which is usually demanding. Therefore an alternative strategy, namely data-based approach, is proposed. Two optimizations derived from approximating the model reference problem formulated in the frequency domain are developed for the feedback system and IMC system, respectively. In the former design, the optimization is solved by using an adjustable parameter in the reference model. PID controller parameters are thus obtained by the least square solution of the optimization problem after the determination of the optimal parameter in reference model. In the latter design, IMC controller and model are computed simultaneously from the optimization problem. Extensive simulation results show that the proposed PID and IMC controllers give comparable or better control performance compared to their respective model-based counterparts that have been tuned on-line to achieve best possible control performances.