Steel plate hard spot detection and identification method combining multiple magnetic parameters

Abstract

The invention discloses a steel plate hard spot detection and identification method combining multiple magnetic parameters, which is characterized in that the change rule of a plurality of magnetic characteristic parameters of an incremental magnetic conductivity curve and a tangential magnetic field intensity signal along with hard spots, sensor lift-off and test piece residual magnetic field is combined to identify the interference of sensor lift-off fluctuation and test piece residual magnetic field on a detection signal; the detection probability of the magnetic detection technology on the hard spots of the steel plate is improved. Specifically, the invention relates to the implementation steps of the steel plate hard spot identification method based on multi-magnetic parameter scanning and detection probability imaging analysis, which are suitable for identifying lift-off fluctuation, characteristic magnetic parameters of a test piece residual magnetic field and a combined judgment rule thereof in a sensor scanning process. The method is suitable for non-destructive testing of hard spots in structures such as ferromagnetic steel plates, steel pipes and container wall plates.

Description

Technical Field

[0001] The present invention relates to a steel plate hard spot detection and identification method combining multiple magnetic parameters, belonging to the field of micro-magnetic non-destructive testing technology. The method analyzes multiple magnetic characteristic parameters in the incremental magnetic permeability curve and the tangential magnetic field strength signal, and studies the variation of these parameters with the hard spot, sensor lift-off height and specimen residual magnetic field, so as to effectively identify the interference of sensor lift-off fluctuation and specimen residual magnetic field on the detection signal. Background Art

[0002] Steel plates are important materials in industrial production, and their quality directly affects the performance and service life of the final product. During the production process of steel plates, hard spots may be formed on the surface or inside of the steel plates due to uneven material composition, improper heat treatment process or other external factors. Hard spots are areas with abnormal local hardness, and their hardness is significantly higher than that of the surrounding matrix material. The presence of hard spots will seriously affect the mechanical properties of steel plates, such as reducing toughness, increasing brittleness, and even causing the steel plates to break or fail during use. Therefore, hard spot detection is crucial. As an emerging method, incremental magnetic permeability detection technology has gradually been introduced into the detection of material surface hardness. Incremental magnetic permeability refers to the slight change in magnetic permeability of a material under the action of an alternating magnetic field, and its value is closely related to the microstructure and mechanical properties of the material. Existing methods use the sensitivity of incremental magnetic permeability to changes in surface hardness to achieve non-destructive detection of hard spots, but in actual applications, it is found that sensor lift-off jitter and specimen residual magnetic field will reduce the accuracy of hard spot detection. Therefore, it is necessary to propose a steel plate hard spot detection and identification method that combines multiple magnetic parameters to improve the accuracy of steel plate hard spot detection. Based on the multi-dimensional magnetic characteristic parameter analysis of the incremental permeability curve and the tangential magnetic field strength signal, this method systematically studies the dynamic change of magnetic characteristic parameters with hard spot defects, sensor lift-off height and specimen residual magnetic field. By extracting and quantifying the interference characteristics of sensor lift-off fluctuations and specimen residual magnetic field on the detection signal, effective identification and suppression of interference factors are achieved. Summary of the invention

[0003] The purpose of the present invention is to provide a method for detecting and identifying hard spots on steel plates by combining multiple magnetic parameters, identifying the interference of sensor lift-off fluctuations and residual magnetic fields of test pieces on detection signals, and improving the detection probability of hard spots on steel plates by magnetic detection technology.

[0004] To achieve the above purpose, this method mainly includes the following steps:

[0005] (1) Place the detection sensor on the surface of the pipeline steel plate and scan the surface of the steel plate. The sensor uses the AC bridge method to measure the original signal of the incremental magnetic permeability of the steel plate. The detection coil is close to the surface of the steel plate under test with the sensor, and the reference coil is far away from the test piece; the Hall element placed in the middle of the detection coil synchronously picks up the tangential magnetic field strength on the surface of the steel plate.

[0006] (2) By recording the change curves of various magnetic characteristic parameters during the sensor scanning process, and using the combination rules of the change laws of multiple curves, the recognition of the sensor lift-off distance fluctuation and the residual magnetic field in the specimen is realized, and the hard spot recognition result with the marked detection probability is given. The steps for determining the sensor lift-off distance fluctuation and the residual magnetic field in the specimen are as follows:

[0007] 1) Extract the bias voltage U from the tangential magnetic field intensity signal measured by the Hall element b and the signal peak-to-peak value U p , extract the peak value M from the incremental permeability curve p , respectively plot their change curves along the scanning path, and mark the coordinate windows where the amplitude is concave or convex in each curve;

[0008] 2) For the curves within the marked coordinate windows, determine the reasons for the curve changes according to the following combination rules:

[0009] (a) If both U p and M p show a concave trend and U b has no obvious change, it is a hard spot;

[0010] (b) If U p shows a convex trend, M p shows a concave trend and U b has no obvious change, it is the influence of sensor lift-off fluctuation;

[0011] (c) If both M p and U b show a concave trend, it is the influence of the residual magnetic field in the specimen.

[0012] (3) Convert the magnetic characteristic parameter scanning distribution image in the incremental permeability curve into a detection probability distribution image through calculation, and realize the recognition of the hard spot area according to the given hard spot determination threshold. The specific implementation steps are as follows:

[0013] 1) Prepare a hard spot specimen, use the sensor to perform point-by-point detection along the path across the hard spot area, obtain the peak value M of the incremental permeability curve at each detection position p , and test the surface hardness H at some detection positions s ,

[0014] Construct the equation: M p = f(H s ) (1)

[0015] 2) Linearize D(M p ) = k·G(H s ) + b (2)

[0016] 3) Calculate the probability of detection curve POD(H s ) = Q(M p1 ) (3)

[0017] where Q(x) is the standard normal distribution and M p1 is the threshold value;

[0018] 4) Substitute the actually measured M p into Equation (2) to calculate the surface hardness of each point;

[0019] 5) Substitute the surface hardness and M p into the probability of detection curve to calculate the probability and obtain the imaging diagram;

[0020] 6) Set the threshold value and mark the hard spot area. Description of the Drawings

[0021] Figure 1 is the flow chart of the detection device and signal analysis.

[0022] In the figure: 1 is a U-shaped yoke, 2 is an exciting coil, 3 is the steel plate to be measured, 4 is a detection coil, 5 is a Hall element, and 6 is a reference coil.

[0023] Figure 2 is the schematic diagram of the detection signal waveform and characteristic quantity.

[0024] Figure 3 is the schematic diagram of the multi-magnetic parameter combination recognition method.

[0025] Figure 4 is the flow chart of the hard spot detection probability imaging. Detailed Embodiment

[0026] The following will combine the drawings and specific embodiments to elaborate in detail on an incremental magnetic permeability detection signal recognition method for steel plate hard spots provided by the present invention.

[0027] As Figure 1 shown, the incremental magnetic permeability detection system includes the control of high- and low-frequency synchronous excitation magnetic fields, the acquisition and storage of multi-channel data. The structure of the sensor is as Figure 1 shown, including a U-shaped yoke 1, an exciting coil 2, a detection coil 4, a Hall element 5, and a reference coil 6. The above U-shaped yoke 1 is arranged on the steel plate 3 to be measured, and the exciting coil 2 is wound around the U-shaped yoke 1; a detection coil 4 is arranged between the exciting coil 2 and the steel plate 3 to be measured; the above detection coil 4 is installed on the steel plate 3 to be measured, a Hall element 5 is arranged on the detection coil 4, and the reference coil 6 is arranged opposite to the detection coil 4;

[0028] During the process of superimposing the low-frequency magnetic field on the high-frequency magnetic field, the frequency of the low-frequency magnetic field can be set to 200 Hz, and the frequency of the high-frequency magnetic field can be set to 10 kHz. The frequency of the high-frequency magnetic field remains above 100 times that of the low-frequency magnetic field. To ensure the complete magnetization of the test piece, the steel plate 3 in the measured area, the low-frequency magnetization signal also needs to be transmitted to the low-frequency excitation coil 2 through a power amplifier. The incremental permeability signals and tangential magnetic field signals detected by the detection coil 4 and the Hall element are returned to the host computer software through a multi-channel acquisition board.

[0029] By recording the change curves of various magnetic characteristic parameters during the sensor scanning process and using the combination rules of the change laws of multiple curves, the recognition of the sensor lift-off distance fluctuation and the residual magnetic field in the test piece is realized, and the hard spot recognition result with the marked detection probability is given.

[0030] The steps for determining the sensor lift-off distance fluctuation and the residual magnetic field in the test piece are as follows:

[0031] (1) As Figure 2 shown, extract the bias voltage U b and the peak-to-peak voltage U p from the tangential magnetic field intensity signal measured by the Hall element, and extract the peak value M p from the incremental permeability curve. Then, plot their change curves along the scanning path respectively, and mark the coordinate windows where the amplitudes of the curves are concave or convex.

[0032] (2) As Figure 3 shown, for the curves within the marked coordinate windows, determine the reasons for the curve changes according to the following combination rules:

[0033] (a) If both U p and M p show a concave trend and U b has no obvious change, it is a hard spot;

[0034] (b) If U p shows a convex trend, M p shows a concave trend and U b has no obvious change, it is the influence of the sensor lift-off fluctuation;

[0035] (c) If both M p and U b show a concave trend, it is the influence of the residual magnetic field in the test piece.

[0036] (3) Convert the magnetic characteristic parameter scanning distribution image in the incremental permeability curve into a detection probability distribution image through calculation. According to the given hard spot determination threshold, the hard spot area recognition is realized. The specific implementation steps are as follows:

[0037] 1) Prepare a hardened spot specimen, and use a sensor to perform point-by-point detection along the path across the hardened spot area to obtain the peak value M of the incremental permeability curve at each detection position p , and test the surface hardness H at some detection positions s ,

[0038] Construct the equation: M p = f(H s ) (1)

[0039] 2) Linearize M p = k·G(H s ) + b (2)

[0040] 3) Calculate the detection probability curve POD(H s ) = Q(M p1 ); (3)

[0041] where Q(x) is the standard normal distribution and M p1 is the threshold value;

[0042] 4) Substitute the actually measured M p into Equation (2) to calculate the surface hardness of each point;

[0043] 5) Substitute the surface hardness and M p into the detection probability curve to calculate the probability and obtain the imaging diagram;

[0044] 6) Set the threshold value and mark the hardened spot area, and the result is as Figure 4 shown.

Claims

1. A method for detecting and identifying hard spots on a steel plate with combined multiple magnetic parameters, characterized in that, The sensor includes a U-shaped yoke, an exciting coil, a detection coil, a Hall element, and a reference coil. The above U-shaped yoke is arranged on the steel plate to be measured, and the exciting coil is wound around the U-shaped yoke. A detection coil is arranged between the exciting coil and the steel plate to be measured. The above detection coil is installed on the steel plate to be measured, and a Hall element is arranged on the detection coil. The reference coil and the detection coil are arranged opposite to each other. The sensor simultaneously detects the incremental permeability curve and the tangential magnetic field intensity signal in the steel plate. By recording the change curves of various magnetic characteristic parameters during the scanning process of the sensor and using the combination rules of the change laws of multiple curves, the identification of the lift-off distance fluctuation of the sensor and the residual magnetic field in the specimen is realized, and the hard spot identification result with the marked detection probability is given. Among them, the steps for determining the lift-off distance fluctuation of the sensor and the residual magnetic field in the specimen are as follows: (1) Extract the bias voltage U from the tangential magnetic field intensity signal measured by the Hall element b and the signal peak-to-peak value U p , extract the peak value M from the incremental permeability curve p , respectively plot their variation curves along the scanning path, and mark the coordinate windows where the amplitudes of the curves are concave or convex; (2) For the curves within the coordinate window of the mark, determine the reasons for the curve changes according to the following combination rules: a) If U p and M p both show a concave trend and U b has no obvious change, it is a hard spot; b) If U p shows a convex upward trend, and M p shows a concave downward trend and U b has no obvious change, it is the influence of the sensor lift-off fluctuation; c) If M p and U b show a downward concave trend simultaneously, it is the influence of the residual magnetic field in the test piece.

2. A method for detecting and identifying hard spots on a steel plate with combined multiple magnetic parameters according to claim 1, characterized in that, In the hard spot identification of the marked detection probability, the scanned distribution image of the magnetic characteristic parameters in the incremental permeability curve is calculated and converted into a detection probability distribution image, and the hard spot area identification is realized based on the given hard spot determination threshold. The specific implementation steps are as follows: (1) Prepare a hardened spot specimen, and use a sensor to perform point-by-point detection along the path across the hardened spot area to obtain the peak value M of the incremental permeability curve at each detection position p , and test the surface hardness H at some detection positions s , and construct the equation: M p = f(H s )(1) (2) Linearize D(M p ) = k·G(H s ) + b (2) (3) Calculate the probability of detection curve POD(H s ) = Q(M p1 ) (3) where Q(x) is the standard normal distribution and M p1 is the threshold value; (4) Substitute the actually measured M p into Equation (2) to calculate the surface hardness of each point; (5) Substitute the surface hardness and M p into the detection probability curve, calculate the probability, and obtain the imaging diagram; (6) Set the threshold and mark the hard spot area.

3. A method for detecting and identifying hard spots on a steel plate with combined multiple magnetic parameters according to claim 1, characterized in that, During the process of superimposing the low-frequency magnetic field and the high-frequency magnetic field, the frequency of the low-frequency magnetic field can be set to 200 Hz, and the frequency of the high-frequency magnetic field can be set to 10 kHz. The frequency of the high-frequency magnetic field is kept above 100 times the frequency of the low-frequency magnetic field. To ensure the complete magnetization of the specimen steel plate in the measured area, the low-frequency magnetization signal also needs to be transmitted to the low-frequency excitation coil through a power amplifier. The incremental permeability signal and the tangential magnetic field signal detected by the detection coil and the Hall element are returned to the upper computer software through a multi-channel acquisition board.

Images