Optimization of the SMX static mixer types thermal and hydraulic performance by coupling CFD-Genetic Algorithm

Abstract The thermo-fluid performance of static mixers was evaluated by numerical simulation of the flow field and heat transfer in turbulent flows. Four types of mixers_ namely: twisted tape turbulator, LPD, SMX, and KSM_ were studied at different Reynolds numbers. The validation of numerical results confirmed their agreement with the experimental results in the literature. According to results, twisted tape and SMX had the lowest and highest effects on enhancing heat transfer rate, respectively. However, SMX showed a higher pressure drop than other mixers. Therefore, Genetic Algorithm (GA) was used to optimize its configuration. Design parameters were considered to be number of mixers, Reynolds number and installation angle of plates. Design of Experiments (DoE) method was used to select the design points. In addition, Response Surface Methodology (RSM) and GA were used to conduct optimization process. Maximum heat transfer rate and minimum pressure drop were defined as objective functions. Sensitivity analysis showed that the objective functions represent maximum sensitivity to Reynolds number. GA derived optimum point at the following conditions: Reynolds number = 96,500, number of mixers = 5, central plate's angle = 45o and lateral plates' angle = 49o. The prediction accuracy of optimization algorithm was assessed by simulation and it was proved that it could predict the thermo-fluid performance of the studied mixers with an acceptable accuracy (prediction error