Study On The Optimization For Impeller Of G4-73 Centrifugal Fan Based On Parametric Modeling

This paper takes the optimization of the G4-73 centrifugal fan impeller parameter to enhance its total pressure and efficiency. Due to outstanding advantages of numerical simulation technology in solving engineering problems, the paper take numerical simulation as the main method for the research of the fan of different impeller parameters.Based on parametric thought, a parametric numerical platform was developed using VB programming language, which can greatly simplify operations of the numerical simulation pre-treatment. Take different values of the blade number, export installation blade angle and impeller width as modeling parameters to make the simulation based on the platform, 48 groups simulation results were got which can be used for optimization by neural network and genetic algorithms. Genetic algorithms were used for multi-objective optimization. And finally a set of optimized parameters of the impeller were obtained, which would be used for simulation to verify the optimization results.Practice showed that the secondary development to the numerical simulation software with the parametric method was feasible, which can improved the working efficiency obviously. On the analysis of the total pressure and efficiency of the fan following single impeller parameter's alteration, we found that the total pressure increases with the increase of the blades number which changes around the original parameters of 12, but the efficiency can not keep increasing when blades number larger than 11. The total pressure and efficiency achieved the highest value when the impeller export installation angle got 45 degree. They got better when impeller width in the range of 18.3 to 19.8 centimeters. The optimization result by the genetic algorithm was 13 blades, 44.6 degrees and 20.25 centimeters, the result of the numerical simulation in the design flow rate based on the platform showed that the total pressure and efficiency have increased by 2.12% and 0.67% over the corresponding model respectively. Total pressure and efficiency were obviously improved around the vicinity of the design flow rate and the velocity distribution of the impeller and volute was also improved, thus the purpose of the optimization was achieved ultimately.