Design and optimization of Stirling engines using soft computing methods: A review. (1st February 2021)

• Record Type: Journal Article
• Authors: Zare, Shahryar; Tavakolpour-saleh, A.R.; Aghahosseini, A.; Sangdani, M.H.; Mirshekari, Reza
• Is Part Of: Applied energy. Volume 283(2021)
• Publication Date: 2021-02-01
• Subjects: Soft computing -- Stirling engines -- Genetic algorithm -- Particle swarm optimization -- Fuzzy logic -- Artificial neural network; Power (Mechanics) -- Periodicals; Energy conservation -- Periodicals; Energy conversion -- Periodicals; 621.042
• DOI: 10.1016/j.apenergy.2020.116258
• Languages: English
• ISSNs: 0306-2619

Abstract

Highlights: The soft computing methods applied to the design of Stirling engines are reviewed. The engine frequency, pistons stroks, and poles locations can be optimized accordingly. The hybrid soft computing methods can predict the engine power, efficiency, and damping. Abstract: The need for energy converters with high thermal efficiency is a central issue in the field of renewable energies. So far, different technologies have been introduced for converting renewable energies into mechanical work. Stirling engines with an optimal design can be an appropriate choice for this aim. In this paper, the applications of soft computing methods in optimization and design of the Stirling engines are discussed. Until now, four popular soft computing approaches such as genetic algorithm, particle swarm optimization, fuzzy logic, and artificial neural network have been extensively applied to design and optimize the Stirling engines. Addressing the conducted works in this field, reveals that these soft computing methods can effectively meet the main concerns of the researchers. The performance of the Stirling engines in terms of power and efficiency can be promoted by optimizing their parameters via the soft computing methods. Moreover, the soft computing methods can be further employed to optimize the Stirling engines based on other objectives such as desired operating frequency, desired strokes of power and displacer pistons, and optimal locations of closed-loop poles of the system. On the other hand, combining these soft computing methods results in hybrid intelligent techniques that serves to predict other complex characteristics of these engines including torque, heat transfer, and damping coefficients. The hybrid techniques usually contain the artificial neural networks (or fuzzy logic) incorporating the evolutionary (or swarm intelligence) algorithms for designing, optimizing, and predicting the engine specifications.