JP4029119B2 - Steel pipe stress diagnosis method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a method for nondestructively diagnosing stress generated by subsidence or formation fluctuation in a steel pipe embedded in the ground, using Barkhausen noise generated from the steel pipe.
[0002]
[Prior art]
Since steel pipes such as gas supply pipes and water pipes are buried in the ground, when subsidence occurs, bending stress is generated between steel pipe parts having different subsidence amounts. If the stress acts on the steel pipe for a long period of time, there is a risk of stress corrosion cracking, and if the stress is excessive, the steel pipe may be damaged. In particular, it is necessary to monitor the stress acting on the buried pipe so as to prevent this from occurring in the gas supply pipe and to confirm safety.
[0003]
For this purpose, a method is used to obtain a bending stress by opening a thin crack from the ground surface to the surface of the steel pipe, inserting a rod called a sinking rod into the resistance, and estimating the deformation of the steel pipe occurring in the ground from the sinking amount of the rod. Has been implemented conventionally. However, this method cannot measure the horizontal displacement of the steel pipe, and because the number of sinking bars is limited, the estimation accuracy of the deformation amount of the steel pipe is insufficient. was there.
[0004]
Therefore, a method has been proposed in which a magnetostrictive sensor (magnetic anisotropy sensor) using magnetostriction is directly applied to the surface of the steel pipe and the stress acting on the steel pipe is obtained from the output value. This measurement principle uses the fact that the magnetostriction is positive in steel materials such as iron, so that when stress acts on the steel pipe surface, the permeability increases in the tensile stress direction and decreases in the compressive stress direction. . For example, a stress release method for adjusting a value calculated by approximating a magnetostrictive sensor output measured around a steel pipe with a sine curve to a value that does not exceed a reference value for security or a minimum value (Japanese Patent Laid-Open No. 3-176630). No. 3) A method of linearly approximating the angular dependence of magnetostrictive sensor outputs in the vicinity of two stress neutral parts, and estimating the bending stress from the average value of the slopes of both (JP-A-3-176626), magnetostrictive sensor output Approximating the difference between SINθ and SINθ approximation by SIN2θ and estimating the flat stress from the amplitude value (Japanese Patent Laid-Open No. 3-176627), removing the measured value of the weld when measuring the ERW pipe with a magnetostrictive sensor Then, a correction method using COSθ and COS2θ (Japanese Patent Laid-Open No. 5-28158), an outer diameter at a position where the magnetostrictive sensor output is maximized and a position shifted by 90 ° therefrom are measured, and the flatness is obtained. Direction The method of correcting the maximum direction stress value (Japanese Patent Laid-Open No. 6-288842), incorporating the stress partially measured by a magnetostrictive sensor or the like into a simulation based on subsidence measurement, obtaining the maximum stress in the entire buried pipe, and determining the reference value A management method (Japanese Patent Application Laid-Open No. 9-242933), etc., that prevents it from exceeding is disclosed.
[0005]
However, when these methods are applied to an existing pipe, if the output value (initial value) at a predetermined measurement site is known before the pipe is buried, that is, with no external stress applied. There is no problem, but if it is unknown, the initial value varies depending on the location, so the stress cannot be obtained directly from the output value measured after embedding. Therefore, the stress has to be obtained using various correction methods.
[0006]
[Problems to be solved by the invention]
As described above, in the conventional method, when trying to obtain external stress acting on each part of the steel pipe where the initial value of the measurement part is unknown, a certain degree of hypothetical judgment is required. The decline of was inevitable.
[0007]
Therefore, the present invention provides a method for enabling stress diagnosis with sufficient accuracy even when the initial value before embedding is unknown by using a steel pipe with a controlled compressive residual stress applied to the surface. For the purpose.
[0008]
[Means for Solving the Problems]
The gist of the present invention is as follows.
(1) Using a magnetic head composed of an excitation head and a detection head as a diagnostic object, the measurement portion of the steel pipe is AC-excited by the excitation head, and the voltage signal induced in the detection head is frequency-separated. A steel pipe stress diagnostic method for detecting Barkhausen noise and diagnosing a stress value applied to the steel pipe,
Applying a predetermined compressive residual stress to the steel pipe surface in advance,
At a predetermined measurement site on the surface of the steel pipe, the effective voltage of Barkhausen noise measured by exciting in the axial direction of the steel pipe is V _L , and the effective value voltage of Barkhausen noise measured by exciting in the circumferential direction is V _C. In this case, from the value of (V _L −V _C ), using a calibration curve representing the relationship between the external stress and the effective value voltage of Barkhausen noise obtained in advance using the same member as the steel pipe, A stress diagnosis method for a steel pipe, wherein the stress value in the axial direction acting on a measurement site is obtained.
[0009]
(2) The compression in the in-plane direction of the measurement site so that the effective voltage of the Barkhausen noise measured by exciting in the axial direction and the circumferential direction of the measurement site on the surface of the steel pipe before installation on the site is uniform. 2. The steel pipe stress diagnostic method according to item 1, wherein the steel pipe to which residual stress is applied is used.
[0010]
(3) The steel pipe stress diagnosis method according to (2) above, wherein the steel pipe to which the compressive residual stress is applied isotropically and uniformly in an in-plane direction of the measurement site is used.
[0011]
DETAILED DESCRIPTION OF THE INVENTION
Since the Barkhausen noise of a steel material changes according to the structure of external stress, crystal grain size, precipitates, dislocations, and the like, it was indispensable not to change the structure in order to diagnose the external stress. That is, even if an external stress acts on the steel material, when it is within the elastic range, there is no structural change, so Barkhausen noise depends only on the stress and reversibly changes with respect to the stress. However, if an external stress higher than the yield stress acts on the steel material and enters the plastic region, the structure changes due to the growth of dislocations and crystal rotation, so it is no longer possible to diagnose only the external stress. It will be possible.
[0012]
The inventors have made it possible to control the initial state of the residual stress at the measurement site so as to hardly cause a structural change even when the magnitude of the external stress becomes larger than the yield stress, In such a state, as a result of examining the relationship between stress and Barkhausen noise in detail, the present invention has been achieved. That is, the present inventors measured by exciting in the axial direction and circumferential direction with stress or strain when plastically deformed from the elastic region to the plastic region, and further in the plastic region to various strains. The relationship between the magnitude of Barkhausen noise was measured in detail. As a result, when the measurement site is plastically deformed and a tensile stress is newly applied to the sample to which compressive residual stress is applied in the in-plane direction, the effective voltage of the stress or strain and Barkhausen noise It has been found that the stress or strain range in which the linear correlation is established is much wider than that in the case where there is no compressive residual stress. Furthermore, the effective voltage (V _L ) of Barkhausen noise measured by exciting in the axial direction changes very sensitively according to tensile stress or tensile strain, whereas it hardly changes on the compression side. The effective voltage (V _C ) of Barkhausen noise measured with excitation in the circumferential direction hardly changes even if the stress or strain changes on both the tension side and the compression side. I found out.
[0013]
In ordinary ERW and seamless pipes, the structure and residual stress are different for each part of the steel pipe surface. Therefore, when actually measuring Barkhausen noise, the effective value of the same steel pipe is only a few centimeters different. The voltage will vary greatly. Therefore, it is necessary to manage the initial value for each measurement site, which is complicated in management. The present inventors apply a compressive residual stress of approximately the same magnitude in the in-plane direction of the surface of the ERW pipe or seamless pipe to measure the Barkhausen noise at any measurement site on the steel pipe surface. It was found that an effective voltage can be obtained. By applying this compressive residual stress isotropically to the surface, the initial value becomes isotropic, and it is possible to prevent deterioration of the diagnostic accuracy of stress no matter which direction external stress is applied from. become. However, when measuring Barkhausen noise at a predetermined measurement site around the steel pipe, even if it is possible to apply a compressive residual stress of uniform size to all of these measurement sites, labor-saving in the manufacturing process In view of productivity, it is desirable to simplify the process of applying residual stress as much as possible.
[0014]
The inventors of the present invention, for example, a method of causing small steel balls and ceramic particles such as air blast and shot blast to collide with a sample surface at high speed, polishing with a sander, etc. Was evaluated. As a result, if these treatments are repeated several times, almost uniform compressive residual stress is applied to all the measurement sites around the steel pipe. I found that the size was different. However, even when the magnitude of the compressive residual stress differs depending on the measurement site, it has been newly found that if the measurement site is the same, the magnitude of the compressive residual stress in the axial direction and the circumferential direction is almost the same.
[0015]
When attention is paid to one measurement site of a steel pipe to which compressive residual stress has been applied by the method described above, excitation is performed in the axial direction and circumferential direction in the absence of external stress before installation on site. The effective value voltage of the Barkhausen noise measured in this way is uniform (V _{L to} V _C , no external stress), but when an external bending stress is applied thereto, the Barkhausen noise measured by exciting in the axial direction ( V _L ) changes according to the stress, but Barkhausen noise (V _C ) measured by excitation in the circumferential direction hardly changes. Therefore, if V _L −V _C is obtained, only the effective value voltage generated by the external stress can be extracted regardless of the location of the measurement site. In consideration of measurement errors when detecting the actual effective voltage of Barkhausen noise, measurement is performed with excitation in the axial and circumferential directions in the absence of external stress before installation on site. If the difference in effective voltage (V _L −V _C ) of the Barkhausen noise is about 2 mV to 3 mV or less, V _L and V _C can be regarded as uniform.
[0016]
In a normal gas conduit or the like, measurement is performed in a state where the internal pressure is applied to the steel pipe. However, since the internal pressure is known, the stress generated on the surface of the steel pipe by the internal pressure can be calculated dynamically. Therefore, in this case, when it is desired to obtain the external stress, the stress increase due to the internal pressure may be subtracted from the measured value.
[0017]
Next, the measurement procedure will be described.
The effective voltages V _L and V _{C of} Barkhausen noise are excited by excitation in the axial direction and circumferential direction at a plurality of predetermined locations around the surface of the steel pipe to which the compressive residual stress controlled at each measurement site is applied. Measure. At that time, considering a plane that includes the pipe axis center line of the steel pipe and intersects with a predetermined measurement site on the surface of the steel pipe, one of each plane is used as a reference plane, and each measurement site is a plane that includes the reference plane and each measurement site Is displayed at an angle between Any surface may be used as a reference surface. The relationship between these angles and (V _L −V _C ) is shown in a graph or table. Here, the normal number of measurement points is about several to several tens.
[0018]
Next, the maximum value of (V _L −V _C ) is obtained from this graph or table. This part is where the tensile stress acting in the axial direction is maximized. The maximum tensile stress value in the axial direction can be obtained using a calibration curve obtained in advance using the same steel type from the maximum value of the effective value voltage. Here, a calibration curve representing the relationship between stress and Barkhausen noise can be easily obtained by simultaneously measuring Barkhausen noise while applying stress to the same steel type member with a strain gauge attached. it can.
[0019]
Further, when this maximum tensile stress value is σ _max , the bending moment M can be obtained from M = Z × σ _max (where Z is a section modulus). When compressive residual stress corresponding to the yield stress is applied to the surface of the steel pipe, diagnosis can be performed with Barkhausen noise up to an external tensile stress equivalent to about twice the yield stress.
[0020]
In an electric resistance welded pipe, there are a welded portion and heat-affected portions on both sides of the welded portion, but bulk hazen noise may change greatly in these portions. Since these parts can be confirmed visually when coating or coating is not applied, they can be removed from the measurement site in advance. What is necessary is just to exclude the value corresponding to a site | part. In a welded part or heat-affected part, the effective value voltage changes discontinuously in a graph showing the relationship between the angle of the measured part and the effective voltage of Barkhausen noise. it can. If a known non-contact type magnetic head (Japanese Patent Laid-Open No. 7-174730) is used, measurement can be performed even from above the covering material.
[0021]
Since the Barkhausen noise generated from a deeper part of the sample is more attenuated, the voltage generated in the detection coil becomes smaller. This is called the skin depth effect. Assuming that the depth of the source where Barkhausen noise attenuates to 1 / e on the sample surface is d, d = (ρ / πfμ) ^1/2 , (ρ is electric resistance, f is the detection frequency of Barkhausen noise, μ Is expressed by permeability). The depth to which the residual stress is applied is preferably at least 0.5 d or more, where d is the detected depth. This is because if it is less than 0.5 d, the stress range in which the linear correlation between the two is reduced in the relationship between Barkhausen noise and stress or strain.
[0022]
When the present invention is actually used, a calibration curve indicating the relationship between the external stress in the member to be measured and the effective voltage of Barkhausen noise is measured in advance, and the effective value actually measured using this calibration curve is measured. The voltage difference (V _L −V _C ) may be converted into stress.
[0023]
【Example】
Hereinafter, the present invention will be specifically described with reference to examples.
A compressive residual stress was applied by performing a shot blasting treatment using a steel-based abrasive over the entire surface of the SGP300A steel pipe having an outer diameter of 318 mm, a wall thickness of 6.9 mm, and a length of 6000 mm. However, the strength of the shot blast treatment was positively changed depending on the part, and the magnitude of the compressive residual stress was changed depending on the location. Barkhausen noise was measured in a state where an arbitrary amount of external stress was applied by bending the steel pipe, and it was examined whether or not external stress could be diagnosed by the present invention.
[0024]
Barkhausen noise was measured as follows. Using a magnetic head consisting of an excitation head in which a 1000-turn enamel wire is wound around a U-shaped excitation core laminated with silicon steel plates, and a detection head in which a 500-turn enamel wire is wound around an acrylic bobbin having a cross-sectional area of 2 mm × 8 mm Then, excitation was performed in two directions, the axial direction and the circumferential direction, and the effective value voltages V _L and V _C at each measurement site were measured. The excitation frequency is 100 Hz, and the detection frequency is 10 kHz to 100 kHz. The case where (V _L -V _C ) was converted to stress using a calibration curve obtained in advance using the same steel type, and the case where _VL was converted to stress were examined as comparative examples.
[0025]
The measurement part of the steel pipe is a plane that includes the center line of the pipe and intersects with a predetermined measurement part on the surface of the steel pipe. One of the planes is used as a reference plane, and the measurement part is a plane including the reference plane and each measurement part. Displayed at an angle formed by In practice, an arbitrary surface was used as a reference surface, and 16 locations were measured over the circumference of the steel pipe at intervals of 22.5 °.
[0026]
The distribution in the depth direction of the residual stress magnitude on the surface of the steel pipe after the shot blasting was determined by an X-ray residual stress measurement method after etching a predetermined thickness from the surface in the plate thickness direction. As a result, at each measurement site, compressive residual stress of approximately the same magnitude is isotropically uniform in the plane at a depth of about 150 μm from the surface, but the absolute value of the compressive residual stress is It was confirmed that they are different.
[0027]
FIG. 1 is a characteristic diagram showing effective values V _L and V _C of Barkhausen noise measured by exciting in the axial direction and the circumferential direction at each measurement site.
Since the magnitude of the compressive residual stress at each measurement site is different, the effective value voltage varies. From this graph, it is difficult to determine which part of the steel pipe is subjected to the maximum tensile stress.
[0028]
FIG. 2 is a characteristic diagram showing V _L -V _C according to the present invention.
From this figure, it can be clearly seen that a tensile stress acts in the axial direction between the 23 ° portion and the 210 ° portion, and that the maximum tensile stress acts on the approximately 110 ° portion.
[0029]
In FIG. 2, it is increased by 4.5 mV at the maximum, but when this is converted into stress using a calibration curve obtained in advance, it becomes 210 MPa. This value almost coincided with the value obtained by converting the value of strain obtained by calculation by accurately measuring the curvature of the bent steel pipe into stress.
[0030]
Therefore, according to the present invention, it is understood that the stress can be diagnosed even when the compressive residual stress of each measurement site is different.
[0031]
【The invention's effect】
According to the present invention, the stress acting on the steel pipe can be accurately diagnosed by measuring the Barkhausen noise of the steel pipe having a compressive residual stress applied to the surface in advance at a predetermined site over the circumference of the steel pipe. Is possible. Since the present invention can be applied not only in the elastic region but also in the plastic region, by using the present invention, the steel pipe that is buried is monitored with the utmost care, and in some cases immediately. The diagnostic accuracy of stress in the plastic region where stress relief work must be performed is also greatly improved.
[Brief description of the drawings]
FIG. 1 is a (measured value) characteristic diagram showing a profile of an effective value voltage of Barkhausen noise.
FIG. 2 is a characteristic diagram showing a profile of an effective value voltage of Barkhausen noise (calculation result according to the present invention).

Claims (3)

Using a magnetic head composed of an excitation head and a detection head as a diagnostic object, AC excitation is performed on the measurement site of the steel pipe by the excitation head, and the voltage signal induced in the detection head is frequency-separated to generate Barkhausen noise. A stress diagnosis method for a steel pipe that detects a stress value applied to the steel pipe,
Applying a predetermined compressive residual stress to the steel pipe surface in advance,
At a predetermined measurement site on the surface of the steel pipe, the effective voltage of Barkhausen noise measured by exciting in the axial direction of the steel pipe is V _L , and the effective value voltage of Barkhausen noise measured by exciting in the circumferential direction is V _C. In this case, from the value of (V _L −V _C ), using a calibration curve representing the relationship between the external stress and the effective value voltage of Barkhausen noise obtained in advance using the same member as the steel pipe, A stress diagnosis method for a steel pipe, wherein the stress value in the axial direction acting on a measurement site is obtained. The compressive residual stress is applied in the in-plane direction of the measurement site so that the effective voltage of the Barkhausen noise measured by exciting in the axial direction and the circumferential direction of the measurement site on the surface of the steel pipe before installation on site is uniform. The stress diagnosis method for a steel pipe according to claim 1, wherein the given steel pipe is used. The steel pipe stress diagnosis method according to claim 2, wherein the steel pipe to which the compressive residual stress is applied isotropically and uniformly in an in-plane direction of the measurement site is used.

Images