Three-dimensional computational fluid dynamics simulation of stirling engine

Abstract

A three-dimensional Computational Fluid Dynamics (CFD) simulation for the GPU-3 Stirling engine is performed. Firstly, the performance of six different eddy-viscosity models are assessed to identify the most appropriate model for the engine simulation. Secondly, a comprehensive characterization of the thermal and fluid flow fields during the cycle is presented. The effects of the unsteady behaviors of both thermodynamics and fluid dynamics on the heat transfer phenomena are also investigated. Results show that the realizable k-ε-enhanced wall treatment model produced the most accurate predictions for engine power with a robust convergence and a reasonable computational time. The average deviation in the CFD results with this model compared with the experimental ones is about 4%. Within the compression and expansion spaces, the dominant heat transfer rates occur during the expansion strokes due to the significant impinging effect of the tumble vortices generated from the flow jetting. Furthermore, the jetting and ejecting processes into the regenerator is characterized by a significant temperature gradient and large matrix temperature oscillation. Also, the temperature profile through the regenerator is almost logarithmic.