Using genetic algorithms to optimize controller parameters for HVAC systems

Abstract This paper presents an adaptive learning algorithm based on genetic algorithms (GA) for automatic tuning of proportional, integral and derivative (PID) controllers in Heating Ventilating and Air Conditioning (HVAC) systems to achieve optimal performance. Genetic algorithms, which are search procedures based on the mechanics of Darwin's natural selection, are employed since they have been proved to be robust and efficient in finding near-optimal solutions in complex problem spaces. The modular dynamic simulation software package HVACSIM + has been modified and incorporated in the genetic algorithm-based optimization program to provide a complete simulation environment for detailed study of controller performance. Three performance indicators—overshoot, settling time, and mean squared error—are considered in the objective function of the optimization procedure for evaluation of controller performance. The simulation results show that the genetic algorithm-based optimization procedures as implemented in this research study are useful for automatic tuning of PID controllers for HVAC systems, yielding minimum overshoot and minimum settling time.

[1]  Arthur L. Dexter.  Intelligent Buildings: Fact or Fiction? , 1996 .

[2]  A. J. Katz,et al.  Generating Image Filters for Target Recognition by Genetic Learning , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  John J. Grefenstette,et al.  Optimization of Control Parameters for Genetic Algorithms , 1986, IEEE Transactions on Systems, Man, and Cybernetics.

[4]  Wai-Hung Ng,et al.  A self-tuning software package for direct digital controllers , 1990 .

[5]  Jean-Michel Renders,et al.  Genetic Algorithms and Their Potential for Use in Process Control: A Case Study , 1991, ICGA.

[6]  Daniel R Clark,et al.  HVACSIM+ building systems and equipment simulation program - user's guide , 1985 .

[7]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[8]  P. A. Stoecker,et al.  Microcomputer Control of Thermal and Mechanical Systems , 1988 .

[9]  H. N. Lam,et al.  Optimising neural network weights using genetic algorithms: a case study , 1995, Proceedings of ICNN'95 - International Conference on Neural Networks.

[10]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[11]  Katsuhiko Ogata,et al.  Modern Control Engineering , 1970 .

[12]  Lawrence. Davis,et al.  Handbook Of Genetic Algorithms , 1990 .

[13]  David Clark,et al.  HVACSIM+ building systems and equip-ment simulation program reference manual , 1985 .