Vibration fault diagnosis of steam turbine generating unit based on rough sets and support vector machine

A model of the vibration fault diagnosis for steam turbine generating unit was investigated by the method of combining rough sets(RS) theory and support vector machine(SVM).The vibration time-domain singal was transformed into frequency domain by fractional Fourier transform.RS was used to reduce redundant attributes,then a key decision table was obtained.The key table was acted as a learning sample to train SVM classifier.After training,SVM classifier can map the relationship between the attribute character and fault style,thus the objective of fault diagnosis was realized.The simulation experimental results show that the method of combining RS and SVM is efficient,which can lessen fault diagnosis time consumption.