Negative-selection algorithm based approach for fault diagnosis of rotary machinery

The main task of the immune system is to detect and eliminate the non-self materials called antigens such as virus and cancer cells which come from inside and outside of the living system. The negative-selection algorithm proposed which is inspired by the negative-selection mechanism of the immune system is able to detect any change. In this paper, the algorithm is improved to make it suitable for the fault diagnosis of mechanical equipment. Moreover, based on the improved algorithm a new approach of fault diagnosis of rotary machinery is proposed and its feasibility is demonstrated by the on-line fault diagnosis of rotary machinery successfully. In addition, the approach of the novel fault diagnosis can also be applied to other fields.