Optimization Research of Centrifugal Fan with Different Blade Number and Outlet Blade Angle

The characteristics of the three-dimension flow field of the G4-73 centrifugal fan were numerically simulated based on the k e − turbulence model with the Fluent, and then verified the simulating result by experiment. Taking the efficiency η maximizing as a function goal, while the blade number and the angle of blade outlet as the variable quantities, the fan impeller parameters are optimized based on least squares method. The optimizing results showed that the performance of centrifugal fan was improved by lowering the energy loss which caused by the secondary flow vortex, the volute tongue, the wake-jet and the angle of attack. After the optimization, the total pressure and efficiency increased 3.7% and 0.5% respectively. Total pressure and efficiency are important parameters of fan performance. In deduction of the energy equation for centrifugal fan, one of the assumption is that the impeller has unlimited blade. In fact, the blade number is always limited, which results in lower total pressure. Slip factor reflect the influence of limited blade on theoretical total pressure. The results show that the slip factor relate to the number of blade, the outlet blade angle, the ratio of inside diameter to outside diameter, the dynamic viscosity of the fluid and the runner's surface roughness(2). Therefore, the number of fan blades and the size of the outlet blade angle directly impact on the fan performance, but it is still difficult to calculate it by theoretical methods(3-4). Cheng xinde studied the selection of the number of the forward-curved blade by experiments, and few papers did research on the impact of blade number and the size of the outlet blade angle on the performance of fan by numerical method. This paper will use the software of Fluent to calculate the total pressure and efficiency of centrifugal fans with different blade numbers and blade outlet angles, and optimize the fan for a higher efficiency based on the least square method.