Novel information fusion fault diagnosis method based on support vector machines
Jianming Lu · Systems engineering and electronics · 2005
A novel fault diagnosis method based on multi-sensor information fusion is presented. First, sensors' information is averagely fused before feature extraction. Then, signals' characteristic features are extracted by means of wavelet transform. Finally, supprt vector machines are adopted to realize character-level fusion and patter recognition. Samples of some kind of rotor platform's fault diagnosis are tested,the test results prove this method is effective and commendable.