JPH03176627A - Estimating method for stress of depressed pipe - Google Patents

Abstract

PURPOSE:To enable estimation for the value of stress due to depressing the pipe generated on the corresponding straight part by approximating with a SIN2theta curve the deviation value which is obtained by subtracting a SINJtheta approximate value from the output of magnetostrictive sensor measured in the straight pipe part near the curved pipe part of a piping material, and utilizing a signal amplitude value of the approximated SIN2theta curve. CONSTITUTION:The outputs of magnetostrictive sensor in the state with stress of the depressed pipe shown in the figure (a), wherein each output of the magnetostrictive sensor for angles indicating the bearings on the pipe periphery are shown by (+) marks. The deviation values obtained by subtracting the SINtheta approximate values from the outputs of magnetostrictive sensor in the (a) are shown in the figure (b), wherein the deviation values for the same angles are shown by (+) marks. The deviation values and the SIN2theta approximate curve approximated thereto are shown in the figure (c), wherein each deviation values are shown by (+) marks and the SINtheta approximate curve is shown by a solid line. Thereby, the deviation values which are obtained by subtracting the SINtheta approximate values from the stress measuring values in the depressed pipe, are almost approximated by the SIN2theta curve varying periodically for every 1/2 turn of the magnetostrictive sensor when the magnetostrictive sensor makes one turn along the pipe peripheral direction.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention is applicable to the case where, for example, loop piping is carried out at the back of an abutment of a pipeline, and stress is generated near the weld between the curved pipe section and the straight pipe section. This invention relates to a method for estimating flattening stress in a tube, in which the stress caused by flattening that occurs in a straight pipe section near the curved tube section is estimated from the measured value of a magnetostrictive sensor.

[Prior Art] In addition to X-rays and ultrasonic waves, there are methods using magnetostrictive sensors to measure stress and residual stress in steel materials or steel structures. As a method for measuring the stress of magnetizable cylindrical materials such as round rods and pipes using this magnetostrictive sensor, there is a magnetostrictive stress measuring method disclosed in Japanese Patent Application No. 153622/1988, which was previously filed.

In the magnetostrictive stress measurement method, when a load is applied to a magnetic material, anisotropy occurs in the magnetic permeability, and the permeability in the direction of the load increases, while the permeability in the direction perpendicular to the direction of the load decreases, so the difference in magnetic permeability in both directions is measured. A magnetostrictive sensor with an excitation coil and a detection coil (
This method measures the direction and magnitude of principal stress by detecting it with a magnetic anisotropy sensor (also called a magnetic anisotropy sensor).

According to this measurement method, the measurement time for -1 point is lO~10
It only takes 0a+sec and is extremely convenient to handle.

However, since conventional magnetostrictive stress measurement methods are generally performed by bringing a magnetostrictive sensor into contact with a surface to be measured, the magnetic resistance at the contact surface varies greatly depending on the state of the surface to be measured. Therefore, there was a drawback that the measurement error became large.

Therefore, the idea of measuring in a non-contact state, that is, with the magnetostrictive sensor a certain distance away from the surface to be measured, has come up, but in this case, the magnetostrictive sensitivity decreases, so the settings of the magnetostrictive sensor must be adjusted. There was another problem that required some very delicate adjustments.

In the invention of the earlier application, the problem in the non-contact measurement was solved, and an apparatus was developed that could perform the magnetostrictive stress measurement method on cylindrical materials such as magnetizable round bars and pipes in a non-contact manner, and the m-determination method was developed. We have provided a method that uses a device to measure the stress distribution in the circumferential direction of a cylindrical material with higher accuracy than before.

FIG. 1 is a diagram explaining the magnetostrictive stress measurement method according to the previous application, and FIG. Stress - σ
indicates that it is working. In addition, the same figure (b) shows the cylindrical material 1.
perpendicular to the central axis of and at a certain distance i from the outer peripheral surface of
While maintaining the lift-off (gap) of Ih, rotate the magnetostrictive sensor 2 once along the circumferential direction in a clockwise direction from the uppermost point of the cylindrical material 1, that is, the angular position O''.
This figure shows a method of continuously measuring magnetostrictive signals detected by the magnetostrictive sensor 2 at each angular position between 0'' and 360''.

FIG. 2 is a diagram for explaining the SIN approximation method using the magnetostrictive stress 1N-1 regular method in FIG. Since the maximum tensile stress occurs at an angle of 0'' (i.e. directly above the cylindrical material 1) and the maximum compressive stress occurs at an angle of 180' (i.e. directly below the cylindrical material 1), the stress distribution approximates a SINθ curve. distribution.

FIG. 2(b) shows the actual value of stress measured by the strain gauge and the approximate value of SINθ when a load of -20 kg/- is applied to the cylindrical material. From this figure, the actual stress distribution and SIN
It can be seen that the curve is quite similar to the θ curve.

[Problems to be Solved by the Invention] In the SINθ approximation method using the magnetostrictive stress measurement method of pipe materials, which is shown in the above-mentioned Japanese Patent Application No. 153622/1982,
Although it is suitable for measuring bending stress in straight sections of pipe material, it is affected by the flattening of the bent section of straight pipes near the bent section of the pipe material.
Even though it is a straight pipe, the stress distribution does not form a clean SIN curve with respect to the angle θ in the circumferential direction of the pipe. Therefore, there was a problem in that the previously applied 5INo approximation method could not be applied to straight pipe sections near curved pipe sections.

Fig. 3(a) is a diagram showing the stress state when the pipe is flattened. In Fig. 3(a), the compressive stress -σ is 90° at the positions of 06 and 180'' in the circumferential direction of the pipe. and 270
It is shown that tensile stress +σ is acting at the position of °.

FIG. 3(b) is a diagram showing the output of the magnetostrictive sensor when the tube shown in FIG. 3(a) is in a state of flattened stress. This is the angle that indicates the direction of the sensor on the pipe circumference when the sensor is rotated once. The vertical axis is the output of the magnetostrictive sensor (in volts), and the measured value of the sensor is indicated by the cross mark in the figure.

FIG. 3(b) clearly shows the problem that the SINθ approximation method is not applicable near the curved portion of the pipe, even in the case of a straight pipe portion.

The present invention has been made to solve this problem, and based on the magnetostrictive sensor output measured in the straight pipe section near the bent pipe section of the pipe material, the stress value due to the flattening that occurs in the straight pipe section is estimated. The purpose of this study is to obtain a method for estimating flattened stress in pipes.

[Means for Solving the Problems] A method for estimating flat stress in a pipe according to the present invention uses a measuring device in which a magnetostrictive sensor moves relatively on the outer peripheral surface or inner peripheral surface of the pipe material in a non-contact state. In the method of estimating the stress distribution in the tube circumferential direction by approximating it with a SIN curve, the deviation value obtained by subtracting the SINθ approximate value from the magnetostrictive sensor output measured in the straight pipe section near the curved section of the pipe material is calculated as SIN2.
The pipe is approximated by a θ curve and includes a tube flattening stress estimating means for estimating a stress value due to flattening occurring in the corresponding straight pipe portion from the signal amplitude value of the approximated 5IN2θ curve.

[Operation] In this invention, the stress distribution in the circumferential direction of the tube is approximated by a SIN curve using a measuring device in which a magnetostrictive sensor moves relatively on the outer circumferential surface or the inner circumferential surface of the tube in a non-contact state. In this method, the deviation value obtained by subtracting the SINθ approximate value from the magnetostrictive sensor output measured in the straight pipe section near the bent pipe section by the tube flattening stress estimation means is approximated by the SIN2θ curve, and the deviation value obtained by subtracting the SINθ approximate value is Approximate S IN2θ
A stress value due to flattening occurring in the corresponding straight pipe portion is estimated from the signal amplitude value of the curve.

[Example] FIG. 4 is a block diagram of a pipe stress measuring device to which the pipe flat stress estimation method of the present invention is applied. In the figure, lO is a traveling device section, which includes a magnetic anisotropy sensor 11 and a traveling trolley 1.
Built-in 2. The magnetic anisotropy sensor 11 is a sensor for detecting magnetic anisotropy in the circumferential direction of a tube material in a non-contact manner, and includes, for example, an excitation coil and a detection coil that are perpendicular to each other, and a constant excitation current is applied to the excitation coil. is applied, and the magnetic anisotropy caused by the action of stress is detected as a voltage signal obtained from the detection coil. The traveling trolley 12 is a traveling mechanism that travels, for example, on rails and/or gears provided on the outer periphery of the tube, and moves the magnetic anisotropy sensor II in the circumferential direction of the tube to perform measurements. i3 is a magnetostriction measurement unit that supplies a constant current to the excitation coil of the magnetic anisotropy sensor 11, and at the same time amplifies the detection signal obtained from the detection coil of the sensor 1t, and converts it into a voltage signal proportional to the magnetic anisotropy. This is a magnetostriction measurement unit that outputs. 14 is a motor driver which supplies a travel drive signal to the traveling trolley I2 to cause it to travel;
An encoder signal is fed back as position information of the driving result. 15 h A/D converter, 16 is for example R3232C
17 is a personal computer 7
(hereinafter referred to as a personal computer), 18 is a data display section using a CRT or liquid crystal.

The operation shown in FIG. 4 will be explained. The stress in the circumferential direction of the pipe material;
To determine the PJ, for example, a rail or/and a gear (not shown) is attached to the outer circumferential surface of the pipe on a plane perpendicular to the central axis of the pipe, and the traveling device section 1o can run on this rail or/and gear via a holder. Attach to. Next, computer 1
7 is the motor driver 14 via the interface 1G.
The motor driver 14 causes the traveling device lO on the rail or/and gear to travel one rotation along the circumference of the pipe.During this traveling, the magnetic anisotropy sensor II (magnetostrictive sensor At each angular position between 0° and 3609 on the outer circumferential surface of the tube shown in FIG. After the signal is amplified, it is output, and the output is further quantized by the A/D converter 15 and supplied to the personal computer 17. The personal computer 17 causes the data display section 18 to display the sensor output value for each angle indicating the direction on the outer circumference of the tube material of the magnetic anisotropy sensor 11, and outputs a hard copy using a printer (not shown) if necessary. Based on the data displayed on the data display section 18 of the measuring device or the hard copy data output by the printer, the flat stress estimation process of the tube according to the present invention can be performed.

FIGS. 5(a) to 5(c) show the flattening stress S of the tube according to the present invention.
FIG. 3 is a diagram illustrating a method of approximation using an IN2θ curve.

FIG. 5(a) is a diagram showing the output of the magnetostrictive sensor in the state of flattening stress of the tube, and the output of each magnetostrictive sensor (unit: volt) is indicated by a 10 mark for each angle indicating the direction of the magnetostrictive sensor on the circumference of the tube. It shows.

Figure 5(b) shows the SIN from the magnetostrictive sensor output in Figure 5(a).
It is a diagram showing the deviation values obtained by subtracting the θ approximate value, and each deviation value (unit is volt) for the same angle as in the figure (a).
are each indicated by a ten mark.

FIG. 5(c) is a diagram showing the deviation value shown in FIG. 5(b) and a 5IN2θ approximate curve that approximates it.
For the same angle as in b), the cross marks indicate each deviation value (in volts), and the solid line indicates the 5IN2θ approximate curve. Fifth
From the stress measurement value in the flat state of the pipe, SINθ is shown in Figure (c).
It can be seen that the deviation value obtained by subtracting the approximate value can be approximately approximated by the SIN2θ curve that changes periodically every 172 rotations of the magnetostrictive sensor when the magnetostrictive sensor rotates once along the pipe circumferential direction. .

FIGS. 6(a) and 6(b) are diagrams for explaining the flat stress within the film and the 5IN2θ approximate curve according to the present invention.

FIG. 3(a) shows a general flattened stress state in which the direction of the maximum value of flattened stress differs from that in FIG. 3(a) by an angle φ. In addition, the figure (b) is the 5IN2θ approximation curve in the flat stress state of the figure (a), A-sin2(θ+
φ) is shown. Here, A is the maximum signal amplitude value, and φ is the angular difference between the flat axis and the reference axis (angle 0') of the magnetostrictive sensor.

Figure 7 is a diagram showing an example of each measurement position in a straight pipe section near a curved pipe.
This indicates that there is a fixed position where the pipe is installed on an abutment, etc., at a position mm apart, and measurement positions ■, ■, ■, and ■ are provided at the four positions shown in the diagram between this welding position and the fixed position. ing.

Figures 8(a) to (d) show each measurement position shown in Figure 7.
It is a figure which shows the magnetostrictive sensor output in - (■), respectively.

In the same figure, (a) is the measurement position, (b) is the measurement position ■, (c
) is the measurement position ■, (d) is the i'1llJ fixed position ■, and the measured values (units are volts) of the magnetostrictive sensor are shown in correspondence to the angle indicating the direction on the tube circumference of the horizontal axis. It is indicated by ten marks.

Figures 9(a) to (d) show the results of approximating the deviation values obtained by subtracting the SINθ approximate value from each magnetostrictive sensor output in Figures 8(a) to (d) using the SIN2θ curve. It is a diagram. (a) to (d) of the same figure indicate the above deviation values with respect to the same angle as the horizontal axis of Fig. 8 at measurement positions ■ to ■, respectively, with 10 marks, and the solid line is a 5IN2θ approximation that approximates this deviation value. Each curve is shown.

Figures 10(a) to (d) are diagrams showing the results of approximating the deviation values obtained by subtracting the 5IN2θ approximate value from each magnetostrictive sensor output in Figures 8(a) to (d) using a SINθ curve. It is. (a) to (d) of the same figure indicate the above-mentioned deviation values for the same angle as the horizontal axis of Fig. 8 at the measurement positions ■ to ■, respectively, with 10 marks, and the solid line indicates the SIN θ that approximates this deviation value. Each approximate curve is shown.

Figures 9(a) to (d) and Figures 10(a) to (d)
As can be seen from Figure 9(a) and (b), there is a 5IN2θ component due to the influence of flattening at measurement positions ■ and ■ in the straight pipe near the curved pipe. It is clear that the value (deviation value between the measured value and the SINθ approximate value) can be sufficiently approximated by the 5IN2θ approximate curve.

In the present invention, the flattening stress in the straight pipe section near the curved pipe is actually measured using a strain gauge, etc., and a calibration curve (not shown) is created by correlating the measured value with the 5IN2θ approximate value. . Therefore, the 5IN2θ approximate value can be calculated from the output of the magnetostrictive sensor near the curved pipe, and the stress associated with the corresponding flattening of the pipe can be estimated from this calculated value and the calibration curve.

In addition, the effect of flattening becomes smaller as you move away from the curved pipe.
At aFJ fixed positions ■ and ■ of the straight pipe section, see Figure 1O (
As shown in C) and (d), it can be seen that the deviation value (deviation value between the measured value and the 5IN2θ approximate value) can be approximated by the SINθ approximate curve. This is because the bending stress component has a SINθ component.

As in the case of estimating the flattened stress, the stress in the straight pipe section slightly distant from the curved pipe is actually measured using a strain gauge, etc., and a calibration curve (
(not shown) is created in advance, the SINθ approximate value can be calculated from the output of the magnetostrictive sensor in the straight pipe section, and the corresponding bending stress can be estimated from this calculated value and the calibration curve.

Further, in the above embodiment, an example was shown in which the magnetostrictive sensor runs on the outer circumferential surface of the tube without contact, but the magnetostrictive sensor may similarly run on the inner circumference of the tube in a non-contact manner. Further, in this case, the magnetostrictive sensor may be fixed by rotating the tube material about its central axis without causing the magnetostrictive sensor to travel. In either case, the magnetostrictive sensor and the tube only need to move relative to each other, and the same effect can be obtained by fixing one and moving the other.

[Effects of the Invention] As described above, according to the present invention, the magnetostrictive sensor moves on the outer peripheral surface or the inner peripheral surface of the pipe material in a non-contact state.
The stress distribution in the circumferential direction of the tube material is determined by using a
In the method of estimation by approximation using an IN curve, the total ? The deviation value obtained by subtracting the +I11 magnetostrictive sensor output SINθ approximate value is S IN
The 2θtIII line is approximated, and the stress value due to deviation occurring in the corresponding straight pipe portion can be estimated from the f3 amplitude value of the approximated 5IN2θ curve, so SIN approximation using the conventional magnetostrictive sensor The oblate horn, which could not be measured by the method, can be measured by a total of A111, and the effect of expanding the range of the magnetostrictive sensor by a total of 11111j can be obtained.

[Brief explanation of drawings]

Figures 1 (a) and (b) are diagrams explaining the magnetostrictive stress ApJ standard method according to the prior application, and Figures 2 (a) and (b) are diagrams explaining the SIN approximation note using the magnetostrictive stress Al11 constant widening in Figure 1. figure,
Figure 3(a) is a diagram showing the stress state when the tube is flattened;
FIG. 3(b) is a diagram showing the magnetostrictive sensor output in the state of FIG. 3(a), FIG. 4 is a block diagram of a tube stress measuring device to which the tube flat stress estimation method of the present invention is applied, and FIG. (a)
~(c) are diagrams illustrating a method of approximating the flat stress of a pipe according to the present invention by a SIN2θ curve, and Figures 6(a) and (
b) is a diagram explaining the flat stress in the membrane and the 5IN2θ approximate curve according to the present invention; FIG. 7 is a diagram showing an example of each measurement position in a straight pipe section near a curved pipe; )~(d)
9 shows the output of the magnetostrictive sensor at each JIIJ fixed position shown in FIG. 7, and FIGS.
SI from each magnetostrictive sensor output in figures (a) to (d)
Figures 10 (a) to (d) show the results of subtracting the Nθ approximation value and approximating the result by a 5IN2θ curve.
5I from each magnetostrictive sensor output in figures (a) to (d)
FIG. 7 is a diagram showing the result of subtracting the N2θ approximation value and approximating the result by a SINθ curve. In the figure, 1 is a cylindrical material, 2 is a magnetostrictive sensor, 10 is a traveling device section, 11 is a magnetic anisotropy sensor, 12 is a traveling trolley,
Is 13 magnetostrictive? tFJ constant part, 14 is motor driver, 1
5 is an A/D converter, 1B is an interface, 17 is a personal computer, and 18 is a data display section.

Claims (1)

[Claims] Using a measuring device in which a magnetostrictive sensor moves relatively on the outer circumferential surface or inner circumferential surface of the tube material in a non-contact state, the stress distribution in the circumferential direction of the tube material is estimated by approximating it with a SIN curve. In the method of
A method for estimating flattening stress in a pipe, comprising: approximating an IN2θ curve, and estimating a stress value due to flattening occurring in the corresponding straight pipe portion from a signal amplitude value of the approximated SIN2θ curve.