Damage detection and multi-faults classification of gear transmission system

A method of intelligent fault detection and diagnosis based on the support vector machine (SVM) was proposed. By measuring the vibration signals of the gear system at different rotating speeds with different conditions and faults, the testing signals were collected. The feature signals of system were extracted and analyzed. SVM was used for gear fault diagnosis, the classifiers of two and multi-classifications were set up, and the algorithms for two and multi-classifications of SVM were discussed. By analyzing, training and testing the samples of simulation data and gear vibration signals, the various damages in different running conditions of gear system were detected, classified and diagnosed. Based on these, the various representative gear damages in different conditions can be well distinguished, the detection rate is as higher as 95% in low rotating speed, and especially the identification rate of multi-faults diagnosis is over 81%. The results show that the support vector machine in gear fault diagnosis is of excellent diagnostic and identifying abilities and has development prospect in engineering applications.