Research On Key Technology Of Stress Detection System Based On Magnetic Barkhausen Noise

Ferromagnetic components are exposed to high temperature,high pressure,high load and other harsh environment for a long time,which easily leads to material performance degradation and structural damage.Nondestructive testing is an important technical means to evaluate material performance,which can find material defects in time and ensure safety.Magnetic Barkhausen noise(MBN signal for short)is a phenomenon produced when ferromagnetic materials are magnetized.When the ferromagnetic material is stressed,the energy state of the magnet and the instantaneous motion characteristics of the domain wall change,and the MBN signal changes accordingly.Therefore,MBN signal method has gradually become a new nondestructive testing method of ferromagnetic material stress,which is widely used in various fields.In this paper,the key factors of MBN signal are analyzed from the angle of micro domain and domain wall,combined with magnetization technology,and the stress detection system based on MBN signal is proposed and established.The main contents include: excitation design,MBN signal detection components and MBN signal stress characterization feature extraction(1)Excitation design: the finite element simulation analysis of excitation module is carried out to determine the distribution trend of magnetic line and magnetic field intensity,and silicon steel sheet is selected for yoke design;the influence law of excitation amplitude,frequency and waveform on MBN signal is studied;the influence law is obtained by finite element simulation analysis of yoke lift off effect,and the method of measuring yoke flux is proposed The spring slide rail structure is designed to solve the lifting problem of magnetic sensor.(2)MBN signal detection component: the application of inductive coil,Hall IC and TMR magnetoresistive sensor in MBN signal detection is discussed.It is found that the bandwidth and amplitude of Hall IC and TMR are much different from that of inductive coil,and the detection performance is inferior to that of inductive coil.At the same time,the signal conditioning circuit including high pass filter and multi-stage amplifier circuit is designed.(3)MBN signal stress characterization feature extraction: MBN signal data analysis software based on Lab Windows is designed to realize fast acquisition and display of MBN signal,filter amplification and stress characterization feature value extraction;peak to peak value,RMS,mean value,pulse number,peak to width ratio,marginal spectrum peak 1,marginal spectrum peak 2 and marginal spectrum energy of MBN original signal are extracted stepwise regression algorithm is used to select the eigenvalues,and it is found that marginal spectrum peak 2 has better linearity and higher sensitivity in stress characterization.