KR101131996B1 - An Eddy Current Examination Method for the Outside Diameter Axial Cracks in Steam Generator Tubes Using Motorized Rotating Pancake Coil - Google Patents

Abstract

As a method of evaluating the size of the outer diameter crack by MRPC test,

Acquiring an eddy current test signal by an MRPC equipped with two pancake coils and one plus point coil using a predetermined frequency;

Analyzing the signal characteristics of each frequency from the acquired eddy current test signal to identify the outer diameter crack;

Creating a correction curve in a channel corresponding to a specific frequency of the plus point coil with respect to the identified outer diameter crack,

Measuring a depth from the point where the identified outer diameter crack starts to a point where the crack ends at regular intervals, and

Calculating an effective depth from the shape data of the measured length to depth.

Description

{An Eddy Current Examination Method for the Outside Diameter Axial Cracks in Steam Generator Tubes Using Motorized Rotating Pancake Coil}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a steam generator heat pipe, and more particularly, to a method of eddy current flaw detection for a motorized rotating pancake coil (MRPC) for outer shaft cracking of a steam generator heat pipe.

In general, in a nuclear power plant, a pressure boundary supporting an internal pressure of about 10 MPa higher than an external pressure is made of a nickel-chromium-iron alloy (Alloy 600, etc.), and is a thin tube having a diameter of about 20 mm and a thickness of about 1 mm. Generator heat pipe is used. However, since the steam generator heat pipe has a high incidence of corrosion cracking due to stress, it is necessary to periodically perform in-service inspection to repair the heat pipe in which cracks are detected. This is because, as the crack increases, the pressure resistance of the heat transfer tube decreases, and when the crack grows to a certain size or more, an accident such as a rupture or leakage of the heat transfer tube occurs. Therefore, to prevent this, regular inspections are being performed.

As the in-service inspection of the heat generator tube of the steam generator, an eddy current inspection nondestructive inspection technique is used. However, in order to detect cracks, MRPC (Motor Rotating Pancake Coil) eddy current flaw detection is mainly used.

In addition, in order to maintain high reliability of the steam generator as a nuclear power plant, the soundness evaluation procedure is periodically performed. The soundness evaluation procedure refers to a technology-based procedure for quantitatively verifying the integrity of a heat pipe. In performing the integrity evaluation of the steam generator, the ability to evaluate the crack size of the eddy current inspection is a very important variable. This is because the integrity evaluation of the heat pipe is based on the crack size and the growth rate of the size.

However, there is a lack of reliable database established for crack size of eddy current inspection. The reason is that the crack size evaluation ability of the inspection is influenced by various variables such as the inspection technique, the shape of the crack, the evaluator's skill, the evaluation procedure, and the signal quality. In addition, it is because it is difficult to secure enough cracks that are the same as the cracks generated in the steam generator.

The present invention has been made in view of the above circumstances, and the optimum crack size evaluation procedure and size evaluation performance of MRPC (Motorized Rotating Pancake Coil) eddy current flaw detection for the outer diameter crack of the heat transfer tube of the steam generator, And by defining the POD (Probability of Detection) performance, MRPC eddy current inspection method for the outer diameter crack of steam generator tube that can increase the value of the coefficient of crystal and reduce the amount of error by reducing the RMSE value. For the purpose of

The present invention has developed and implemented a multilateral evaluation program utilizing a retrograde steam generator in order to solve the problems of commonly used techniques and to obtain a reliable MRPC eddy current inspection performance. Specifically, as a large-scale program that utilizes various actual cracks of the decommissioning steam generator and observes the same inspection techniques, evaluators, and evaluation procedures as the on-site inspection, a database that simulates the inspection performance of the actual site was established. In addition, by conducting the destruction test in the radioactive material handling facility by passing the heat pipe, both the shape and size of the crack are utilized.

Among them, 39 representative outer diameter cracks were selected, and various evaluation procedures were applied, and ten different evaluation teams (A1 to A10) independently evaluated the size of the cracks, and the actual size confirmed by the fracture test. To compare with. At this time, the evaluation procedure applied is based on the evaluation procedure of the current field inspection, and six different evaluation procedures are developed by changing the combination of frequency, correction curve (based on signal amplitude or phase angle), and notch depth of correction specimen. It is applied.

By statistically analyzing the evaluation size database compared to the actual size constructed by the above procedure, firstly, the optimal procedure is derived by comparing the superiority of different evaluation procedures, and secondly, the crack size capability of the optimal procedure is quantitatively presented. .

The evaluation size compared with the actual size is considered to have a primary correlation, and the superiority of each evaluation procedure is compared by comparing the statistical decision variable and the square root of the mean error.

The size of the outer diameter crack is obtained by applying the crack length, maximum depth, effective depth and effective length. All four crack size parameters consistently show that the specific evaluation procedure yields the best performance.

Core technologies for the soundness assessment of steam generators include crack detection probability (POD) and crack size evaluation capability as the performance of nondestructive testing.

In the present invention, in order to verify a high degree of reliability as a nuclear power plant, the integrity evaluation is carried out, and the performance of the non-destructive inspection, the growth rate of the crack, the reduction of the pressure resistance according to the crack size, the physical properties of the heat pipe material, etc. are all considered. Deterministic or probabilistic evaluation is being performed.

According to a first aspect of the present invention for achieving the above object, a method for evaluating the size of the outer diameter crack by MRPC inspection, MRPC equipped with two pancake coils and one plus point coil using a predetermined frequency Acquiring an eddy current test signal, and analyzing the signal characteristics of each frequency from the obtained eddy current test signal to identify the external shaft crack, and to the specific frequency of the plus point coil with respect to the identified external shaft crack. Creating a correction curve in the corresponding channel, measuring the depth from the point at which the identified outer diameter crack starts to the point at which the crack ends at regular intervals, and the effective depth from the shape data of the measured length to depth Comprising a step of calculating, wherein the correction curve, the axis of the depth of 100%, 60%, 20% of the thickness of the tube The outer celebrated crack sizing characterized in that the amplitude of the eddy currents created by the check signals obtained from the heat transfer tubes toward the notch machining is based are provided.

In addition, it is preferable to further include a step of estimating the actual crack size by substituting the size of the crack evaluated by the above steps into the following equation (1).

Where VR is the actual size, VM is the size evaluated by inspection, a0 and a1 are constants, and ε represents the error, respectively.

By the above steps, the magnitude of the error can be reduced by increasing the crystal coefficient value and reducing the RMSE value than the conventional inspection procedure.

The predetermined frequency is any one selected from 400 kHz, 300 kHz, 100 kHz, and 10 kHz, and the specific frequency is preferably 300 kHz.

Furthermore, according to the second aspect of the present invention, when performing the health assessment based on the MRPC test, a method for evaluating the POD indicating the probability that a crack of a certain size is detected or not detected in the test, Comprising the step of calculating as a function of the following formula (2), provided that the POD evaluation method characterized in that the maximum depth or effective depth is applied as the size of the crack.

(2)

(2)

Where x _i is the crack size and a and b are constants defining the logistic function.

In this case, the integrated POD reflects the average skill of all the evaluators, and the technical POD represents the POD when the evaluator's skill is perfect, i.e. as the technical limit of the MRPC test.

On the other hand, when the crack length is shorter than the predetermined threshold value, instead of the POD derived by the POD evaluation method as described above, it is preferable to calculate the POD using Equation (3) below considering both depth and length. Do.

(3)

(3)

Where L is length or effective length, and D is maximum depth or effective depth.

By calculating the POD by the above formula (3), the accuracy can be further improved (see Fig. 3).

According to the present invention, the optimum crack size evaluation procedure, size evaluation performance, and POD (Probability of Detection) performance of MRPC (Motorized Rotating Pancake Coil) eddy current inspection for outer diameter crack of the heat pipe of the steam generator By defining it, it is possible to reduce the magnitude of the error by increasing the value of the coefficient of determination than the existing inspection procedure and reducing the RMSE value.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

The performance of the eddy current flaw test for the heat generator tube of the steam generator is influenced by various variables such as the shape and size of the crack (outer / inner diameter, axial / circumferential direction), evaluator's skill, inspection technique, and signal quality.

The crack generated in the steam generator grows as the intergranular crack branches, the gap is closely adhered, and iron oxides are deposited around.

It is difficult to accurately simulate the performance of non-destructive testing on actual cracks because the electric discharge notch or fatigue cracks rapidly produced in the laboratory and corrosion cracking due to stress are smooth and clean and the crack gaps are not as dense as actual cracks. .

Therefore, in order to evaluate the performance of inspections for actual cracks, a research program utilizing a retrograde steam generator is carried out, and the acquisition and evaluation of the inspection signals all apply the same techniques, evaluation personnel and procedures as the actual plant inspections. .

In order to check the shape and size of the actual cracks and to compare them with the test results, a plurality of heat pipes were drawn and fractured at the radioactive material handling facility.

The crack size was recorded with the maximum depth and length, and the effective length and effective depth derived from the crack shape were also reflected. After calculating the burst pressures for each of the rectangular cracks that intersect in the irregular shape of the actual cracks, the depth and length of the rectangular crack with the smallest burst pressure were defined as the effective depth and the effective length. It is considered that the bursting pressure of the rectangular cracks conservatively represents the bursting pressure of the actual cracks.

The optimum procedure and performance of crack size evaluation are derived through the following process.

First, six different evaluation procedure parameters are selected, which are based on the current inspection techniques of the site, but which modify the frequency and notch correction curves. (The current inspection technology corresponds to ③.)

① 400 kHz: signal amplitude correction curve; Calibration specimen notch depth (100, 60, 40)

② 400kHz: signal amplitude correction curve; Calibration specimen notch depth (100, 60, 20)

③ 300kHz: signal amplitude correction curve; Calibration specimen notch depth (100, 60, 40)

④ 300 kHz: signal amplitude correction curve; Calibration specimen notch depth (100, 60, 20)

⑤ 400kHz: phase angle correction curve; Calibration specimen notch depth (100, 60, 40)

⑥ 300kHz: phase angle correction curve; Calibration specimen notch depth (100, 60, 40)

The cracks of 20 regression steam generators and 39 heat pipes were independently sized by 10 teams of 2 people each. At this time, all six evaluation procedure specifications are applied individually.

The estimated size and actual size are analyzed by linear regression equation (1) below.

Where VR is the actual size, VM is the size evaluated by inspection, a0 and a1 are constants, and ε represents the error, respectively.

Constants and data necessary for the above formula (1) can be presented as shown in Table 1. All data presented in Table 1 are the values derived from the evaluation procedure ④ above.

TABLE 1

Next, the statistical determination coefficient (r2 coefficient of determination) and RMSE (Root Mean Square Error) values are obtained from the regression results of the six test techniques, and the performance is compared. In this case, the closer to 1 the coefficient of determination satisfies the perfect linear relationship, and the smaller the RMSE value, the smaller the error between the regression equation and the actual value.

Next, since the evaluation procedure ④ consistently shows the largest r2 value and the smallest RMSE value in all size variables, the evaluation procedure ④ is selected as the optimal evaluation procedure. The regression results of the evaluation procedure ④ are shown in Figs. 1A to 1D. 1A to 1D are diagrams showing the regression relationship between the size evaluated by applying the optimum evaluation procedure and the actual size checked by the fracture test.

Next, POD performance is derived through the following procedure.

First, the MRPC test signals for the retrograde steam generator tubes are evaluated independently by nine different teams. In this case, each team consists of three members: the primary, the secondary, and the final evaluator (team leader).

Then, when comparing the results of the non-destructive test evaluation with the actual cracks identified by the failure test, all the ambiguous cases are subjected to technical discussion by the expert team, and finally, the size of the 75 outer diameter cracks and the detection by each team are constructed in the database table. .

Crack detection by 9 teams is indicated by 1 if detected and 0 if not detected. The POD database is binary data with only zeros and ones.

To derive a continuous POD function from this, we use the log-logistic function of equation (2).

(2)

(2)

Here, x _i represents the size of the crack, a, b represents a constant defining a logistic function, respectively.

Constants and data necessary for the calculation of Equation (2) can be presented as shown in Table 2 and Table 3.

Table 2. Integrated POD

Table 3. Technical POD

In this case, the integrated POD reflects the average skill of all the evaluators, and the technical POD represents the POD when the evaluator's skill is perfect, i.e. as the technical limit of the MRPC test.

Since the general linear regression method cannot be applied to the regression equation of Equation (2), the maximum likelihood method is applied.

The evaluation results for each of the nine teams can be derived as a POD function, in which case the POD function is a continuous function approaching 0 when the crack is small and approaching 1 when the crack is large. In the case where the size of the crack is the maximum depth or the effective depth, a statistically excellent POD function is defined, but it is difficult to give a statistical meaning to the POD function based on the length or the effective length.

Next, the technical POD and integrated POD are derived by integrating the evaluation results of nine teams into one. The technical POD represents the limitation of the inspection technique that can be expected when the evaluator's skill is perfect, and if one of the nine teams detects it, it is detected. On the other hand, the integrated POD is a POD function derived by integrating all nine team POD data, which means the performance that can be expected from any evaluator with average skills.

FIG. 2 is a diagram showing an integrated POD curve of MRPC eddy current inspection, which shows a representative integrated POD curve of a continuous POD function derived using the function of Equation (2).

The size parameter of the crack that dominates the POD function is the maximum or effective depth, but the crack length may also affect the POD somewhat. In order to simultaneously consider the effect of the length and depth of the crack on the POD, a POD contour can be drawn as shown in FIG. 3. 3 is a diagram showing the effective depth and the effective length POD contour of the MRPC eddy current inspection, which shows the POD contour derived using the two-variable logistic function of Equation (3). From these POD contours, it can be seen that when the crack length is short (particularly 10 mm or less), the length can also have a significant effect.

(3)

(3)

Where L is the length or the effective length, and D represents the maximum depth or the effective depth, respectively.

By calculating the POD using the function of Equation (3), the accuracy can be further improved (see Fig. 3). Constants and data necessary for the calculation of Equation (3) can be presented as shown in Table 4.

Table 4

Next, a method for evaluating the outer diameter crack size according to the present invention will be described with reference to FIG. 4. 4 is a flowchart illustrating a method for evaluating the outer diameter crack size according to the present invention.

The method for evaluating the outer diameter crack size of the present invention is a method for evaluating the size of the outer diameter crack by MRPC inspection, wherein two pancake coils and one plus point coil are mounted using frequencies of 400 kHz, 300 kHz, 100 kHz, and 10 kHz. An eddy current test signal is acquired by the MRPC (step S10), and the outer diameter crack is checked by analyzing signal characteristics of each frequency from the obtained eddy current test signal (step S11).

Subsequently, a correction curve is generated in the channel corresponding to 300 kHz of the plus point coil with respect to the identified outer diameter shaft crack (step S12), and from the point where the identified outer diameter shaft crack is started to the point where the crack ends at regular intervals. The depth is measured (step S13), and the effective depth is calculated from the shape data of the measured depth to depth (step S14).

Finally, the actual crack size is estimated by substituting the size of the crack evaluated by the above steps into the above formula (1) (step S15).

The correction curve is generated based on the amplitude of the eddy current test signal obtained from the heat transfer tube machined with the axial notches of 100%, 60%, and 20% depth to the thickness of the tube.

According to the method for evaluating the outer diameter crack size of the present invention as described above, the magnitude of the error can be reduced by increasing the crystal coefficient value and reducing the RMSE value from the existing inspection procedure.

The present invention can be immediately used for on-site inspection and soundness evaluation procedure. The size evaluation performance data suggested by the present invention can be used as a key element in the soundness evaluation procedure, and is a key input data when developing a soundness evaluation computer program. In addition to deterministic evaluation techniques, various probabilistic evaluation techniques can be used. It can also be used as a basis for developing eddy current inspection technology.

1A to 1D are diagrams showing the regression relationship between the size evaluated by applying the optimum evaluation procedure and the actual size confirmed by the fracture test.

Figure 2 shows the integrated POD curve of MRPC eddy current inspection.

3 is a diagram showing the effective depth and the effective length POD contour of the MRPC eddy current inspection.

4 is a flowchart illustrating a method for evaluating the outer diameter crack size according to the present invention.

Claims (5)

delete delete delete When performing soundness assessments based on Motorized Rotating Pancake Coil (MRPC) tests, Probability of Detection (POD) indicates the probability that a crack of some size will be detected or not detected in the test. As a method of evaluation, Computing the detectability (POD) as a function of the following formula (2), Detectable (POD) evaluation method characterized in that the maximum depth or effective depth is applied as the size of the crack.

(2) Where x _i is the crack size and a and b are constants defining the logistic function. 5. The method of claim 4, When the crack length is shorter than the predetermined threshold value, instead of the detectability POD derived by the detectability POD method as described above, the following equation (3) considering both depth and length is used. A method for evaluating detectability (POD), characterized by calculating a detectability (POD).

(3) Where L is length or effective length, and D is maximum depth or effective depth.

Images