声响信号分析的柴油机故障诊断方法

Fault Diagnosis Method for Diesel Engines Based on Sound Signal Analysis

柴油机声响信号中蕴含了丰富的柴油机异常或故障状态信息。部件磨损、零件松动、配合间隙增大、装配不当和断裂损坏时,均会伴随着各种异常声响。有经验的维修人员凭借耳听结合个人经验就可以判断出故障部位及原因。通过实时检测和采集柴油机工作时的声响信号,并运用数字滤波、小波变换和功率谱分析法从中提取和分离异常的声响信号,对其进行量化分析,提取故障声音的特征参数,从而识别故障的类型。

Sound signal of diesel engines contains a lot of abnormal information and fault information. For example, worn-out and failure of components, parts loosing, clearance increasing and poor assemblage will cause various abnormal sounds. The experienced experts can determine the positions and reasons of the faults by listening to the abnormal sounds with their own experiences. This paper used a new method to detect and collect the real-time sound signals of the diesel engines. Digital filtering, wavelet transforms and power spectrum analysis were applied to extract and separate the different sound signals. Meanwhile, abnormal sound signals were analyzed quantitatively in order to extract the characteristic parameters and identify the faults.