Self-Tuning Control of a Heating, Ventilating and Air-Conditioning System

Abstract

This paper describes the minimum-variance parameter-adaptive (self-tuning) control algorithms for SISO and MIMO systems. The algorithms are designed on the basis of linear input-output system models by the combination of recursive-parameter estimation and control algorithms: a single-variable minimum-variance self-tuning controller and a multi-variable minimum-variance self-tuning controller. These controllers are applied to a three-input three-output single environmental space, which consists of an air heater, air humidifier and ventilator, and whose output variables are temperature, relative humidity and air velocity. The results of simulation indicate that it is possible to use the self-tuning controllers to stabilize the controlled system after a short adaption phase and to achieve at least a satisfactory control performance for time-varying set-points of the output variables.