A Fuzzy-Based Analytic Hierarchy Process for Mechanical Noise Source Identification of a Diesel Engine

Abstract

In this paper, a novel fuzzy-based analytic hierarchy process (FAHP) is proposed for identifying noise sources of diesel engine and sorting their contributions. The hierarchy tree consists of 5 levels including the component level, the frequency band level, the test point level, the operating condition level, and the overall noise level. The variational mode decomposition (VMD) is employed to decompose the overall noise into several frequency bands. The factor weights between levels are determined by the partial coherence analysis (PCA) to consider the correlation of component vibration and radiated noise in different frequency bands. A fuzzy consistency matrix is then constructed to order the corresponding factors level by level that can avoid the consistence problem of the traditional AHP. Based on the rig test of a diesel engine, the proposed approach is implemented to weight the noise sources identified. It is shown the overall weighting order of the six major noise sources is (in descending order) the oil pan, the left block, the valve cover, the flywheel housing, and the right block.