Kriging with Meta-Heuristic Methods for Optimal Design to Reduce the Noise of the Engine Cooling Fan

Abstract

This paper proposes an optimal design scheme to reduce the noise of the engine cooling fan by adapting Kriging with two meta-heuristic techniques. An engineering model has been developed for the prediction of the noise spectrum of the engine cooling fan. The noise of the fan is expressed as the discrete frequency noise peaks at the BPF and its harmonics and line spectrum at the broad band by noise generation mechanisms. The object of this paper is to find the optimal design for noise reduction of the engine cooling fan. We firstly show a comparison of the measured and calculated noise spectra of the fan for the validation of the noise prediction program. Then, L18 orthogonal array is applied as design of experiments because it is suitable for Kriging. With these simulated data, we can estimate a correlation parameter of Kriging by solving the nonlinear problem with genetic algorithm and find an optimal level for the noise reduction of the cooling fan by optimizing Kriging estimates with simulated annealing. We notice that this optimal design scheme gives noticeable results. Therefore, an optimal design for the cooling fan is proposed by reducing the noise of its system.