Fault Diagnosis of a Gearbox Under Varying Speed Based on STFT and SVM

Gearboxes, one of the most important part of rotating machine transmission systems, be diagnosed and classified quickly and efficiently make great signification in condition monitoring. By detecting and analyzing the vibration signals of gearboxes, we can judge the situation under varying speed conditions. In this paper, a novel method of the feature extraction based on Short Time Fourier Transform (STFT) is proposed for the diagnosis of faults in gearboxes with the vibration signals. Then, Support Vector Machine is used to categorize the malfunctions. This method mainly includes three stages: Firstly, we measure and analyze the vibration signals collected under normal and fault conditions respectively based on STFT. Secondly, the average velocity of each window used in the last step is calculated to compose a novel fault characteristic matrix, combined with the result of first stage. After that, we use the matrix as inputs to train the SVM. Two experiments are carried out to verify the effectiveness of the proposed method.