Non-Contact Fault Detection Based on Smart Phone: Analysis and Design

The detection and diagnosis of rotating machinery faults are of great significance. Although many online detection technologies have been developed, obtaining an accurate estimation of the faults of a complex system is still a practical challenge. Due to economic reasons concerning the cost related to data acquisition and analysis, or due to practical reasons such as the inaccessibility of locations for measurement, non-contacted diagnosis has drawn worldwide attention. In this study, a non-contact portable fault detector based on smart phone is designed using the improved maximum correlated kurtosis deconvolution (IMCKD) technique. The designed fault detector integrates acoustic signal acquisition, IMCKD signal processing technique, fault diagnosis and monitoring into a smart phone, which avoids the traditional sensor placement problem and achieves a real-time heath monitoring and efficient fault detection for rotating machinery. The applications in terms of functionality, flexibility and precision in faulty diagnosis is verified by two case studies with different damage scenarios for a real locomotive rolling bearing.