A set of accurate and time-saving methods were proposed to predict the rotation noise of vehicle alternator cooling fan. First

to predict the overall noise and main order noise magnitudes of the original fan

an acoustic analogy method combined with large eddy simulation (LES) and Ffows-Williams and Hawkings (FW-H) equations was applied. Second

to predict the change of rotation noise caused by blade angle modification

a vector composition method was proposed in case only the blade distribution angle was changed. The main order noise magnitudes of the modified fan were eventually calculated. Calculations and experimental results indicate that the maximum prediction error of the overall noise of the original fan is 4.3dB

and the prediction errors of main rotation noise magnitudes in orders 12 and 18 are 1.24dB and 4.26dB

respectively. The rotation noises of the modified fan in orders 12 and 18 are decreased by 9.3dB and 10.5dB

while the prediction errors of their sound pressure level (SPL) variations are 0.36dB and 0.43dB

respectively. The results show that this set of methods can effectively predict the rotation noise of fan

providing a significant guide to the design of blade distribution angles.