Mechanism of Electromechanical Fault Diagnosis of Large Machinery Based on Acoustic Image Pattern Recognition
Based on the advantages of non-contact acoustic fault diagnosis technology, this paper develops the acoustic imaging pattern recognition technology based on support vector machine. This can be introduced into the field of fault diagnosis. According to the characteristics of a mechanical fault, the acoustic image is obtained by using beamforming algorithm to identify and locate the noise source, and then the acoustic image is processed. In this paper, the texture features and singular value features of the sound image are extracted, and the support vector machine (SVM) is used for training classification, which is then used for the diagnosis of mechanical working state. Through simulation and experiment, a better diagnosis effect is obtained. The experimental results show that the image feature extraction technology based on acoustic image can be applied to mechanical fault diagnosis after combining with support vector machine, which provides a reference for the application of acoustic imaging method in the field of fault diagnosis.
Keywords-sound image, pattern recognition, fault diagnosis, data fusion.