Modelling of Material Properties Using Frequency Domain Information from Barkhausen Noise Signal

Riku-Pekka Nikula, Aki Sorsa, Suvi Santa-aho, Minnamari Vippola, Kauko Leiviskä

    Research output: Contribution to journalArticleScientificpeer-review

    Abstract

    Frequency spectrum, bispectrum and bicoherence which are computed from Barkhausen noise (BN) signal are used to model material properties. The use of frequency domain information can be a significant addition to the more common time domain data analysis of the BN signals. The frequency spectrum shows the magnitude of the spectral components present in the signal. These components can also have interaction which is revealed only by the higher-order spectra. Third order spectrum can be used to detect the quadratic phase coupling phenomenon, which is a result of nonlinearity in the signal. In this study, a special attention is paid on the segment biphase to distinguish the quadratic phase coupling from constant non-zero biphase. Partial least squares regression models are made to model the surface hardness and residual stress properties from a set of carburizing case-hardened steel samples.
    Original languageEnglish
    Pages (from-to)201-206
    Number of pages6
    JournalIFAC-PapersOnLine
    Volume48
    Issue number17
    DOIs
    Publication statusPublished - 2015
    Publication typeA1 Journal article-refereed

    Keywords

    • Barkhausen noise
    • bicoherence
    • biphase
    • bispectrum
    • feature extraction
    • ferromagnetic materials
    • frequency spectrum
    • partial least squares regression

    Publication forum classification

    • No publication forum level

    Fingerprint

    Dive into the research topics of 'Modelling of Material Properties Using Frequency Domain Information from Barkhausen Noise Signal'. Together they form a unique fingerprint.

    Cite this