Research on Multi-Objective Optimization of Bioreactor Cupped Impeller Cfd Based on Improved Proxy Model

Bioreactors are widely used to liquid-liquid mixing in the chemical and biological fields. The study aims to analyze the impact of bioreactor cupped impeller structures on performance and to optimize the structure using an improved surrogate model. Firstly, a complete Computational Fluid Dynamics (CFD) and Fluid-Structure Interaction (FSI) simulation model is established, and its applicability is verified using the component transfer method. Consideration of impeller structural variables(X, k, i,M) , the influence of impeller structure on mixing and mechanical performance was explored with turbulent flow energy and equivalent stress as the analysis objectives. Secondly, sensitivity analysis is innovatively introduced to improve proxy model, solving the problem of excessive training sample size. Then, the mapping relationship between impeller structure parameters and performance was established, and the key parameters were optimized by genetic algorithm. The results showed that the best mixing effect was attained with four cups. But this configuration also led to the highest equivalent stress. Furthermore, adjusting the cup angle changed the nearby vortices, affecting mixing performance, and the magnitude of the equivalent stress was associated with the size of the pressure values and their distribution. The optimized impeller configuration ranged: X as one, M approximately between 160 mm and 165 mm, and angle β around 6° or 35°. Concurrently, compared with the optimal solution of the range analysis, the optimization effect of the improved surrogate model on the turbulent flow energy of the impeller is increased by 45.63%, and the maximum equivalent stress is reduced by 48.67%.