Extraction of feature spectra by instantaneous power spectrum and automatic generation of symptom parameters by GP for diagnosis of machinery in unsteady operating condition

This paper proposes a failure diagnosis method for plant machinery in unsteady operating conditions using the instantaneous power spectrum (IPS) and genetic programming (GP). The IPS is used to extract the feature spectra of each machine state from the measured vibration signal for distinguishing failures by the relative crossing information (RCI). Excellent symptom parameters for detecting failures are automatically generated by GP. The methods proposed in this paper are verified by applying them to the failure diagnosis of rolling bearings