Four-variable simultaneous optimization of the cooling and acoustic power with particle swarm optimization

Abstract

The low performance of the thermoacoustic refrigerator has made it uncompetitive to currently available refrigeration systems and hence its path towards commercialization is being restricted. Recently, evolutionary algorithm such as genetic algorithm has become popular among researchers in optimizing the performance of the thermoacoustic refrigerator due to its capability to provide a solution with a global maximum or minimum through simultaneous optimization of several objectives. The purpose of this study was to maximize the performance of the thermoacoustic refrigerator using the Multi-Objective Particle Swarm Optimization (MOPSO), an evolutionary optimization tool that has not been tried in this field before. By optimizing the two conflicting objectives which are maximizing the cooling power and minimizing the acoustic power required, simultaneous optimization of inter-dependent controlling parameters has been performed for two, three and four parameters. Comparing with the results of past studies, MOPSO has improved the stack COP by 6.92% compared to the parametric optimization approach and 2.96% higher than the maximum COP achieved by multi-objective genetic algorithm (MOGA) with an optimum COP of 1.39. Also, a maximum cooling power of 10.8 W was obtained. This study has highlighted the potential of MOPSO in providing optimized conditions for conflicting objectives desired for a thermoacoustic system.