Optimization of micromixer with different baffle shapes using CFD, DOE, meta-heuristic algorithms and multi-criteria decision making

Micromixers are devices that have the task of mixing on a small scale. In this study, to improve the mixing index in micromixers with minimum pressure drop, the optimal values of the affecting factors on the micromixer performance were obtained by using computational fluid dynamics (CFD), design of experiment (DOE), metaheuristic algorithms and multi-criteria decision making (MCDM). The most influential factors on the performance of micromixers are Reynolds number, blockage ratio and the ratio of the distance between baffles to baffle diameter, which is examined in the different baffle shapes. The dimensionless ratios obtained from these parameters, which were achieved by using the mentioned methods, significantly improved the efficiency of micromixers and they are facilitators and necessary for future studies. The obtained mathematical models from DOE were optimized by meta-heuristic algorithms and then the results of algorithms were compared with each other by four evaluation criteria. It was shown multi-objective Particle Swarm Optimization (MOPSO) had the best performance in presenting the preferred solutions. The designs were ranked among the members of the Pareto front by MCDM. The preferred solution with the ratio of mixing index to pressure drop equal to 0.78 was the most optimal case.

Automated summary

By utilizing CFD, DOE, metaheuristic algorithms, and MCDM, this study identified the optimal values of factors affecting micromixer performance, resulting in significant improvement in efficiency. Multi-objective Particle Swarm Optimization was found to be the most effective in providing preferred solutions, with the most optimal case having a mixing index to pressure drop ratio of 0.78.