Multi-objective heat transfer optimization of 2D helical micro-fins using NSGA-II

Abstract

A numerical simulation of helical micro-fins is implemented in ANSYS Fluent 15. The model is scripted to automatically set up and execute given three input parameters: fin height, helix angle, and number of starts. The simulation results reasonably predict experimental pressure drop and heat transfer for multiple micro-fin geometries. A multi-objective parameter optimization is implemented based on the NSGA-II algorithm to estimate the optimal trade-off (Pareto front) between Nusselt number and friction factor of a micro-fin tube for 0.0006 < e/D < 0.045, 10 < Ns < 70, at Reynolds number of 49,013. The resulting Pareto front is analyzed and compared with several experimental data points. From the optimal results, a distinct difference in flow characteristics was identified between geometries above and below a helix angle of approximately 45°. How the Pareto front can be used to choose micro-fin geometries for different performance evaluation criterion scenarios is demonstrated. Optimal results from various existing correlations are also compared to the optimization results.