Optimization of Dual-Impeller Configurations in a Gas–Liquid Stirred Tank Based on Computational Fluid Dynamics and Multiobjective Evolutionary Algorithm

Abstract

An optimization strategy combining computational fluid dynamics (CFD) with multiobjective evolutionary algorithm (MOEA) for dual-impeller design in an aerated tank was proposed to maximize the overall effective gas holdup and minimize the power consumption with six geometrical variables. The nondominated sorting genetic algorithm-II (NSGA-II) was applied to construct a Pareto front from numerous design points with greatly reduced computation. The measurement of local gas holdup and power consumption by dual electric conductivity probe and torque sensor was utilized to verify the CFD model and evaluate the optimal design. The optimal design with a pitched concave blade disk turbine as the lower impeller and a down-pumping pitched blade turbine as the upper impeller exhibited the best gas dispersion performance with efficient energy savings. This approach has the potential to greatly enhance the efficiency of industrial stirred reactors.