Multi-Objective Optimization for the Impeller Parameters of Centrifugal Fan Based on Kriging Model and GA-PSO Algorithm

Background: Large centrifugal fans with high efficiency are being developed fast recently, reflecting the trends that saving energy is becoming more and more important. Objective: The purpose of this study is to explore the design method of efficient impeller, and to form the design ideas and concepts, this can provide reference for the enterprises to design efficient impeller in the future, and improve enterprises' production efficiency. Methods: In order to improve the comprehensive quality of centrifugal fan, an integrated optimization strategy based on Kriging surrogate model and Genetic algorithm combined with Particle Swarm Algorithm (GA-PSO) was proposed. The optimization strategy was used to optimize the impeller parameters of centrifugal fan. Coupled with Computational Fluid Dynamics (CFD) analysis, the Orthogonal test method was used to arrange the experimental points. In the paper, inlet blade angle, outlet blade angle, inlet width, outlet width, inlet radius, outlet radius and blade number were considered as impeller significant structure parameters, Then Kriging surrogate model was introduced to establish predictive model between centrifugal fan comprehensive aerodynamic quality and the impeller structure parameters. Besides, GA-PSO was used to seek the best result of the predictive model in the feasible solution space. Results: After optimization, the efficiency is increased by about 2%, and the efficiency increased significantly. The optimal impeller parameters were inlet blade angle of 36.35 degree, outlet blade angle of 58.87 degree, inlet width of 189.2mm, outlet width of 91.9mm, inlet radius of 239.45mm, outlet radius of 515.65, and blade number of 16. Conclusion: CFD analysis and actual experiment achieved good agreement, showing that the integrated optimization strategy was feasible and reasonable in centrifugal fan impeller quality optimization.