Use of genetic algorithms to develop an adaptive fuzzy logic controller for a cooling coil

Abstract

An adaptive approach to control a water valve for a cooling coil, called an adaptive fuzzy logic controller (AFLC), is developed and validated in this study. The AFLC calculates the error between the supply air temperature and the supply air temperature set point for air in an air handling unit (AHU) of a heating, ventilating, and air conditioning (HVAC) system and continues to improve the fuzzy controller parameters to minimize the error. The AFLC uses genetic algorithms (GAs) to improve the fuzzy rule matrix and fuzzy membership functions for the AHU in HVAC systems. In this paper, the application of genetic algorithms for developing the AFLC is presented. After a brief background on fuzzy logic controllers and GA theory, the use of GAs is explained. Three methods of modifying the fuzzy rule matrix using the GAs are presented along with simulation and real-time experimental results. Experimental results indicate that GAs can be successfully applied to modify an AFLC rule matrix to achieve a better controller.