A new feature of magnetic memory signal for quantitative identification of stress concentration

Abstract

Magnetic memory detection technology is a new nondestructive testing technology developed in recent 20 years, which can detect the stress concentration of ferromagnetic materials. At present, the quantification of stress concentration in magnetic memory testing is always a difficult problem. In the quantitative research of magnetic memory detection, the key is to extract the effective characteristics of magnetic memory signal. In this paper, a feature of magnetic memory signal based on wavelet packet energy spectrum is proposed. The experimental results show that the stress concentration of ferromagnetic materials is related to the wavelet packet energy spectrum distribution of magnetic memory signals. The lower the stress concentration, the more scattered the energy spectrum distribution of wavelet packet. The higher the stress concentration is, the more concentrated the energy spectrum of wavelet packet is, and the energy of wavelet packet is mainly concentrated in the low frequency band.