Owing to the nonlinear and non-stationary characteristics of noise of diesel engines,it is difficult to diagnose its abnormal noise.Currently,research on abnormal noise diagnosis is mostly based on order analysis and signal processing methods.However,these methods are not suitable for non-order and high-frequency abnormal noise sources.To address this issue,a method for abnormal noise diagnosis based on feature extraction and acoustic array technology was proposed.First-ly,the possible frequency bands of the abnormal noise were determined based on the results of feature loudness MAP,tone to noise ratio MAP,and prominence ratio MAP.Then,noise sample filtering design was carried out according to the feature extraction results,and the frequency spectrum distribution of the abnormal noise was determined through subjective evalua-tion experiments.Finally,the abnormal noise source identification was realized through acoustic array technology.The pro-posed method was applied to the abnormal noise diagnosis of a diesel engine.It is found that the abnormal noise is mainly distributed in the range of 11700-12200 Hz,and the abnormal noise source is the turbocharger compressor.The method proposed in this study is universal for abnormal noise source diagnosis of power machinery.
万方数据
维普
