Multivariable PI control for a boiler plant benchmark using the Virtual Reference Feedback Tuning

Abstract

In this work, the Virtual Reference Feedback Tuning is applied to the control of a boiler plant benchmark. Only data (without any modelling step) was used to compute a series of PI controllers for both a decentralized and a complete multi-loop control strategies. When compared with other PI controllers, it was found that a better tuning of the controller was achieved yielding better closed-loop results. The original VRFT method was enhanced with a series of constraints during the optimization process that allows the user to select the position of the zero of the controller.