Fault diagnosis of rotating machinery based on FA-DSAE

As one of the most widely used equipment, rotating machinery fault diagnosis is significant. The data-driven methods aim at mining the relationship between monitoring data and fault and realizing automatic and accurate equipment fault diagnosis results. However, current fault diagnosis methods generally depend on prior knowledge and the experienced diagnosticians, unsuitable for a variety of conditions. The deficiencies limit the development of intelligent fault diagnosis technology of rotating machinery. In this research, an adaptive fault diagnosis method of rotating machinery based on firefly algorithm and deep sparse autoencoder (FA-DSAE) is proposed to ensure accurate and stable diagnosis result while greatly reduce the dependence on expert experience and prior knowledge. The method is applied for roll bearing and hob fault diagnosis and have a good performance.