Using two improved particle swarm optimization variants for optimization of daily electrical power consumption in multi-chiller systems

• MA-PSO and EPSO increase the diversity of PSO algorithm.
• EPSO produces better results than PSO and MA-PSO algorithms.
• EPSO is an efficient tool for solving DOCL problem.

One of the most important issues in multi-chiller systems (MCSs) is more energy saving by the minimization of the total electrical power consumption (TEPC) of the chillers. In this paper, daily optimal chiller loading (DOCL) problem is introduced where a 24-h cooling load profile should be satisfied by a number of chillers so that the total power consumption of the chillers during 24-h is minimized. Since in DOCL problem, the number of the decision variables which should be tuned simultaneously is 24 times greater than OCL, DOCL is a more complex optimization technique than OCL. Particle swarm optimization is an efficient stochastic metaheuristic technique which has shown a promising performance in solving the OCL optimization problem. As a result, in this paper, for efficiently solving the DOCL problem, two variants of PSO named elitism-based PSO (EPSO) and multi-agent PSO (MA-PSO) are developed. Compared with the original PSO, the proposed MA-PSO and EPSO find better results.

Figure optionsDownload as PowerPoint slide