Multi-objective optimization of a novel butterfly-wing vortex generator fabricated in a rectangular microchannel based on CFD and NSGA-II genetic algorithm

• Published in: Applied thermal engineering Vol. 234; p. 121187
• Main Authors: Das, Ananta Kumar, Hiremath, Somashekhar S
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 05.11.2023
• Subjects: Butterfly-wing; Entropy-generation; Friction factor; NSGA-II; Nusselt number; Taguchi design; TOPSIS; NSGA-II; Nusselt number; Friction factor; Entropy-generation; Butterfly-wing; TOPSIS; Taguchi design
• ISSN: 1359-4311
• DOI: 10.1016/j.applthermaleng.2023.121187

•Fabrication of butterfly-wing in the microchannel.•Validation of numerical model with experiment results.•Sensitivity analysis carried out by Taguchi design using S/N ratio.•Multi-objective optimization performed using NSGA-II with TOPSIS. A novel butterfly-wing vortex generator is fabricated inside the microchannel of 1 mm width (W), 0.63 mm height (H) and 50 mm length (L) by micro milling process using KERN EVO micro machining center. The Numerical model is validated by the experiments performed in the microchannel with the butterfly-wing of 0.36 mm bigger width (w), 0.18 mm smaller width (b), 0.2 mm wing height (h), 1 mm wing length (l), and a fixed number of wing (n) i.e., 5. Further, the numerical simulation is carried out to study the thermo-hydraulic performance and entropy-generation in the microchannel by employing Taguchi’s design of experiment. Thus, the L27 experimental design is formed by varying the input variables in 3 levels, i.e., w (0.24–0.48) mm, b (0.12–0.24) mm, h (0.2–0.5) mm, l (1–3) mm, and Reynolds number (Re) (142–544). A sensitivity analysis is performed to know the influence of each input parameter by calculating the signal-to-noise ratio for the friction factor (f), average Nusselt number (Nuavg), and total entropy-generation (SG), and the same is verified by the CFD simulation results. A multi-objective optimization is carried out to balance the trade-off between the increment of Nuavg and the decrement of f and SG. In this optimization study, the performance factor (Pf) and augmentation entropy-generation number (Ns,a) are considered as output responses because both the parameters explain the phenomenon of fluid flow, heat transfer and entropy-generation in the microchannel. Further, utilizing the simulation results, an empirical model is developed by response surface method for both the output responses. With the developed empirical model, the optimization is performed using NSGA-II to maximize the Pf and minimize the Ns,a. A decision-making technique, i.e., TOPSIS is used to get the best optimal solution from the acquired pareto-front. The optimal input parameters obtained from the TOPSIS are 0.48 mm (w), 0.12 mm (b), 1 mm (l), 0.38 mm (h) and 544 (Re) with the output responses of Pf = 1.35 andNs,a = 0.67.