New magnetic Barkhausen noise feature extraction for stress detection with slow feature analysis

Magnetic Barkhausen noise (MBN) is used in the non-destructive evaluation (NDE) of ferromagnetic materials. MBN signals have been shown to be sensitive to the stress of ferromagnetic materials. However, traditional MBN features for stress detection have poor linearity and are inaccurate for quantitative estimation of material stress. Slow feature analysis (SFA) is an unsupervised learning method that extracts the slowly-varying feature from input signals, which represents the inherent characteristics of the input signals. Since MBN signals are sensitive to the stress of ferromagnetic materials, slow feature analysis of MBN signals can be more accurate for stress detection compared with traditional MBN features.