The use of fuzzy control in industrial processes has increased greatly in the last few decades, especially in processes with difficult mathematical models, due to its ability to function in a system based exclusively on specialist knowledge and also to inter-relate all the process variables. In a liquid chiller, quality is directly related to the capacity to maintain the temperature of the secondary fluid exactly at the value demanded by the process. The development and experimental evaluation of the fuzzy-PI and fuzzy-PID controllers were studied under a desired secondary refrigerant temperature in a refrigeration system for liquid cooling. The system was submitted to load and set point disturbances, an analysis being carried out based on the time-integral error criteria and transient response. The results showed that fuzzy controllers were satisfactorily adaptable to liquid chiller processes. (c) 2005 Elsevier Ltd. All rights reserved.