Fast and robust fixed-point algorithms for independent component analysis
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
In view of the complexity of aero-engine system, there are massive correlated parameters, a novel method for aeroengine fault diagnosis based on Independent Component Analysis (ICA) is proposed in this paper. ICA is already widely applied in many domains, but the aeroengine fault diagnosis domain hasn't involved. ICA is used to detect fault by calculating I2 statistics, and SVM models are constructed based on separate matrix of ICA. Applications illustrate the efficiency of the proposed approach.