JP2003050232A - Automatic calibration device for eddy current signal - Google Patents

Abstract

PROBLEM TO BE SOLVED: To automatically calibrate the phase and sensitivity of an eddy current signal even when the phase and amplitude are totally different between a sample signal 21 and an object signal 22. SOLUTION: The correlation (sample correlation) of mutual phase and amplitude of signal elements (1)-(4) of the sample signal 21, and the correlation (object correlation) of the mutual phase and amplitude of elements A-F of the object signal 22 are compared, and the corresponding signal elements of the sample signal 21 and object signal 22 are specified to match the sample signal 21 and object signal 22. Even when the phase and amplitude are totally different between the sample signal 21 and object signal 22, the sample signal 21 and object signal 22 are correctly matched to allow automatic calibration of phase and sensitivity of the eddy current signal.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an eddy current signal automatic calibration apparatus and method for automatically calibrating the phase and sensitivity of an actual eddy current signal based on a reference eddy current signal.

[0002]

2. Description of the Related Art The eddy current flaw detection method is generally used for nondestructive inspection of a metal pipe which is a magnetic material such as a heat transfer tube. In order to analyze the eddy current signal and perform non-destructive inspection, it is necessary to perform a calibration process to make the phase and amplitude of the sensor signal constant.
Therefore, the phase / sensitivity of the eddy current signal has been conventionally adjusted.

In order to adjust the phase and sensitivity of the eddy current signal,
Conventionally, a time series of eddy current signal elements for a test piece provided with scratches, holes, dents, etc., which are signal factors, is stored, and the eddy current signal element actually obtained through a sensor in the test piece (target The time series of the signal element) and the time series of the signal elements (sample signal elements) stored in advance are compared, and the sample signal element and the target signal element are matched to match the phase of the eddy current signal with respect to the target signal element. Adjusting the sensitivity is done.

Specifically, the time series (peak / peak) of the target signal element and the time series (peak / peak) of the sample signal element are compared in terms of amplitude, phase angle, time width, and the like. The time series of the target signal element and the time series of the sample signal element are matched by correlating the existing ones. Therefore,
By matching the time series of the target signal element and the time series of the sample signal element, it was possible to automatically calibrate the phase and sensitivity of the eddy current signal.

[0005]

However, in the conventional adjustment of the phase / sensitivity of the eddy current signal, the time series of the target signal element and the time series of the sample signal element are compared to determine the amplitude, phase angle, time width, etc. Matching was performed based on the similarity of. For this reason, if the amplitude, phase angle, time width, etc. of the target signal are significantly different from the sample signal (which can occur sufficiently before calibrating the sensor), it is considered that matching cannot be performed. Therefore, there is a possibility that automatic calibration of the phase / sensitivity of the eddy current signal cannot be performed.

The present invention has been made in view of the above situation. Even when the target signal and the sample signal have completely different phases and amplitudes, the two are correctly matched and the phase of the eddy current signal is adjusted. -It is an object to provide an automatic eddy current signal automatic calibration device and an automatic calibration method that enable automatic sensitivity calibration.

[0007]

An automatic eddy current signal calibrating apparatus of the present invention for achieving the above object stores a characteristic of a sample signal of a test piece, which is a target of sensitivity adjustment, as a time series of sample signal elements. Sample storage means, target signal extraction means for taking in a target signal to be analyzed and extracting it as a time series of target signal elements, and sample signal element correlation for deriving a sample correlation between the sample signal elements stored in the sample storage means The deriving means, the target signal element correlation deriving means for deriving the target correlation between the target signal elements extracted by the target signal extracting means, and the sample correlation and the target signal element correlation deriving means derived by the sample signal element correlation deriving means. An associating unit that associates the sample signal and the target signal based on the derived target correlation, and a signal based on the sample signal and the target signal that are associated by the associating unit. Characterized by comprising an automatic calibration means for automatically calibrating the situation.

The sample correlation and the object correlation are characterized in that they are correlations of the phase angles of the signals. Further, the sample correlation and the target correlation are characterized in that they are correlations of signal amplitudes. Further, the sample correlation and the target correlation are characterized in that they are correlations of signal intervals. Further, the sample correlation and the object correlation are characterized in that they are the correlation of the phase angle and the amplitude of the signal. Further, the associating means is provided with a function of comparing the aspect ratio of the circumscribed rectangle of the Lissajous-displayed signal between the sample signal element and the target signal element. Further, the associating means adds the absolute value of the difference between the value of the phase difference of the target signal element and the value of the phase difference of the sample signal element in an arbitrary time series, and displays the time series in which the added value becomes the minimum. It is characterized by having a function of associating a sample signal with a target signal by making a determination.

In the associating means, the absolute value of the difference between the amplitude difference value of the target signal element and the amplitude difference value of the sample signal element is added in an arbitrary time series, and the added value becomes the minimum. It is characterized by having a function of associating a sample signal with a target signal by determining a time series. Further, the associating means adds the absolute value of the difference between the value of the phase difference of the target signal element and the value of the phase difference of the sample signal element in an arbitrary time series, and displays the time series in which the added value becomes the minimum. The function that correlates the sample signal and the target signal by making a judgment, and the absolute value of the difference between the amplitude difference value of the target signal element and the amplitude difference value of the sample signal element is added in an arbitrary time series, and the added value With the function of associating the sample signal with the target signal by determining the time series in which
It has a function of adding different weights to the phase angle determination result and the amplitude determination result.

The automatic eddy current signal calibration method of the present invention for achieving the above object stores the characteristics of the sample signal of the test piece, which is the target of the sensitivity adjustment, as a time series of the sample signal elements, and analyzes the object. The signal is captured and extracted as a time series of target signal elements, and the sample phase / amplitude correlation between the stored sample signal elements is calculated, and the target phase / amplitude correlation between the extracted target signal elements is calculated, and the sample phase / amplitude is calculated. The sample signal and the target signal are associated with each other based on the correlation and the target phase / amplitude correlation, and the signal condition is automatically calibrated based on the associated sample signal and target signal.

Further, the sample interval correlation between the sample signal elements and the target interval correlation between the target signal elements are obtained, and the sample interval signal and the target signal are associated with each other after adding a target interval correlation after a lapse of a predetermined time from the start of automatic calibration. Is characterized by. Also,
The aspect ratio of the circumscribed rectangle of the Lissajous-displayed signal is compared between the sample signal element and the target signal element, and the sample signal and the target signal are provisionally associated with each other.

[0012]

DETAILED DESCRIPTION OF THE INVENTION

FIG. 1 is a concept of a state where an inspection is carried out by a pipe inspection apparatus equipped with an eddy current signal automatic calibration apparatus according to an embodiment of the present invention, and FIG. 2 is a block configuration of the automatic calibration apparatus. , FIG. 3 shows the state of the sample signal and the target signal for the test piece, FIG. 4 shows the relationship between the phase angle and the signal element, FIG. 5 and FIG. 6 are flowcharts of the automatic calibration method, and FIG. 8 shows the concept of the lattice points of the sample signal and the target signal, and FIG. 8 shows the concept of the lattice points of the sample signal and the target signal in the evaluation value calculation.

As shown in FIG. 1, in the pipe inspection apparatus 1, an eddy current flaw detection probe 2, a signal transmission unit 3 and the like are linearly connected via a flexible member 4.
The pipe inspection device 1 moves axially in the pipe 6 supported by the pipe sheet 5, the flexible member 4 is bent at the bent portion of the pipe 6, and the entire pipe inspection device 1 is inserted along the bent pipe. It has become so.

A test piece 7 is arranged on the inlet side of the pipe 6 of the tube plate 5, and the pipe inspection apparatus 1 is passed through the test piece 7 which is preliminarily provided with scratches, holes and dents, and an eddy current flaw detection probe 2 is provided.
The eddy current signal's phase and sensitivity (condition) are automatically calibrated.
That is, the pipe inspection device 1 sends an eddy current signal (target signal) to the automatic calibration device 8 when passing through the test piece 7, and the automatic calibration device 8 stores the sample signal of the test piece 7.

In the automatic calibrating device 8, the signal element (sample signal element) for scratches, holes, or dents in the sample signal and the signal element (target signal element) of the eddy current flaw detection probe 2 are matched, and the eddy current is detected. The phase / sensitivity of the eddy current signal of the current flaw detection probe 2 is adjusted. After the eddy current signal is automatically calibrated, the signal of the eddy current flaw detection probe 2 in the pipe 6 is analyzed to inspect the damage and the like in the pipe 6.

The automatic calibration device 8 will be described with reference to FIG.

As shown in the figure, a sample signal management device 11 is provided as a sample storage means for receiving the sample signal of the test piece 7 and storing the characteristics of the sample signal as a time series of sample signal elements. In addition, the eddy current signal of the eddy current flaw detection probe 2 is captured by the signal capturing device 12, the signal portion (peak / peak) is extracted by the extracting device 13, and the feature of the signal portion (target signal element A time series) is calculated (signal extraction means).

A provisional associating device 15 is provided to provisionally associate the target signal calculated by the feature calculating device 14 with the sample signal stored in the sample signal managing device 11. The result of the temporary association performed by the temporary association device 15 is the evaluation device 1
It is weighted by 6 and evaluated (detailed evaluation will be described later), and sent to the changing device 17. The changing device 17 changes the correspondence as necessary, and the element of the calibration signal is specified by the specifying device 18 to specify the correct correspondence (corresponding means).

The elements of the calibration signal identified by the identifying device 18 are sent to the signal calibrating device 19, and the signal calibrating device 19 detects the phase / sensitivity (situation) of the eddy current signal of the eddy current flaw detection probe 2.
Is automatically calibrated.

As shown in FIG. 3C, signal factors 7 to 7 such as scratches and holes are formed on the test piece 7, and FIG.
As shown in FIG. 5, the sample signal 21 includes elements (corresponding to the time series of sample signal elements) corresponding to the signal factors 7 to 7. On the other hand, the target signal 22 from the eddy current flaw detection probe 2 that has passed through the same test piece 7 is, as shown in FIG. 3B, signal factors 7 to 7 and noise factors A, B, C,
D, E, F are present.

The phase and sensitivity of the target signal 22 are adjusted by matching the sample signal 21 and the target signal 22. This time,
By specifying the corresponding element of each signal, the phase / sensitivity (phase angle or voltage of sensitivity) of the target signal 22 is adjusted and standardized. Therefore, in the automatic calibration device 8, the correlation between the elements of the sample signal 21 (sample correlation) is obtained, and the correlation between the elements A to F of the target signal 22 (target correlation) is obtained, and the sample correlation and the target correlation are obtained. Are compared to identify the corresponding elements of the sample signal 21 and the target signal 22, and the sample signal 21 and the target signal 22 are matched.

An example of matching will be described with reference to FIG.

As shown in FIG. 3 (e), in the element of the sample signal 21 corresponding to the signal factor 7, the amplitude L of the waveforms 23, 23 of the element displayed in Lissajous
, L (L / L), and the waveform 23
, 23 of the phase θ1 (sample correlation) is obtained. Further, as shown in FIG. 3 (a), in the element AB of the target signal 22 corresponding to the signal factor 7, there is a relationship between the amplitudes 1A and 1B of the waveform 24AB of the element AB displayed in the Lissajous manner (lA / lB). And the phase θ2 (symmetrical correlation) of the waveforms 24A and 24B is obtained.

Sample signal 21 with the same signal factor 7
Of the waveform 23 and the waveform 24AB of the target signal 22, regardless of the sizes of the waveforms 23 and 24 (L / L
) And (lA / lB) are approximately the same, and the phase θ
1 and θ2 are substantially the same. In this way, the same signal factor 7
Correspondences of the elements in the waveform 23 of the sample signal 21 and the waveform 24AB of the target signal 22 corresponding to are substantially the same. Therefore, the corresponding elements of the sample signal 21 and the target signal 22 are identified by comparing the sample correlation and the target correlation,
The sample signal 21 and the target signal 22 are matched.

As shown in FIG. 4, element A of the target signal 22.
When the degree of coincidence in the most overlapping state is evaluated by translating the line segment connecting the line segment that connects -F and the line segment that connects the element of the sample signal 21 to each other, for example, in the phase angle correlation, the element A of the target signal 22 is evaluated. , B, D, and F respectively correspond to elements of the sample signal 21 (solid line), and C and E of the target signal 22.
Can be determined to be an element of noise (dotted line). The same applies to the amplitude correlation. As a comparison in this case, it is also possible to judge the degree of coincidence by considering a learning value or the like in addition to the actual value.

The automatic calibration method will be described with reference to FIGS.

As shown in FIG. 5, when the test piece 7 is new, the test piece 7 is registered. The sample signal 21 of the test piece 7 is prepared in step S1, and the signal portion (peak / peak) of the sample signal 21 is extracted in step S2. In step S3, the phase angle and amplitude, which are the characteristic quantities of the signal portion (element), are calculated, and the characteristic quantities are stored in step S4. In this embodiment, the aspect ratio of the circumscribed rectangle of the Lissajous-displayed element is also calculated as the feature amount of the signal portion in step S3. Then, the stored characteristic amount is read in step S5. When the existing test piece 7 exists, the feature amount of the existing test piece 7 is read in step S5.

The steps S1 to S5 are processes of the sample signal management device 11.

In step S6, the target signal 22 which is the signal to be analyzed is read (signal capturing device 12), and in step S7 the signal portion (peak / peak) of the target signal 22 is extracted (extracting device 13). In step S8, the phase angle, the amplitude, and the aspect ratio of the circumscribed rectangle of the Lissajous-displayed element, which are the characteristic quantities of the signal portion (element), are calculated (feature calculation device 14). Test piece 7 calculated in step S3
The phase angle, the amplitude of the (sample signal 21) and the aspect ratio of the circumscribed rectangle of the Lissajous-displayed element and the phase angle of the target signal 22 calculated in step S8 and the aspect ratio of the circumscribed rectangle of the Lissajous-displayed element are It is sent to step S9.

In step S9, the elements of the sample signal 21 and the target signal 22 are provisionally associated with each other, and the candidates for association are narrowed down. For example, the sample signal 21 and the target signal 22
The aspect ratios of the elements are compared, and if the difference in the aspect ratio is greater than or equal to the reference, the elements are excluded from the association candidates. That is, as shown in FIG. 7, the vertical axis represents elements A to F of the target signal 22 (see FIG. 3) and the horizontal axis represents elements of the sample signal 21 (see FIG. 3).
In the case of, the difference in the aspect ratio at each intersection (each combination) is evaluated, and the combination deviates from the standard, that is, in the illustrated example,
The combinations of the elements and the elements D, the elements and the elements A and E, the elements and the elements C, and the elements and the elements B have a difference in the aspect ratio larger than the reference, and therefore are excluded from the association candidates.
This makes it possible to eliminate unnecessary comparison processing.

As shown in FIG. 6, in step S10,
The elements of the sample signal 21 and the target signal 22 are provisionally associated with each other based on the phase angle and the amplitude. That is, as shown in FIG. 4, the line segment connecting the elements A to F of the target signal 22 and the sample signal 2
The line segment connecting the elements 1 to 1 is translated, and the degree of coincidence in the most overlapping state is evaluated. The process of step S9 is omitted and the test piece 7 calculated in step S3 is
The phase angle and amplitude of the (sample signal 21) and the phase angle and amplitude of the target signal 22 calculated in step S8 are calculated in step S10.
It is also possible to send to.

Steps S9 and S10 are processing of the temporary associating device 15.

Next, in step S11, the difference .theta.k1 between the phase angles of the elements of the sample signal 21 and the difference .theta.k2 between the phase angles of the elements A to F of the target signal 22 are obtained, and (.theta.k1-.theta.k2).
The sum Σθ of the absolute values of is set as the evaluation value 1.

For example, the elements of the sample signal 21 and the target signal 2
There are many combinations of the two elements, as shown in FIG. Considering here the difference θk1a between the phase angles of the elements of the sample signal 21 and the phase angle θk2a between the elements A and B of the target signal 22, (θk1−θk2) becomes (θk1a
−θk2a), and considering the difference θk1b between the phase angle between the elements of the sample signal 21 and the phase angle θb2b between the elements B and C of the target signal 22, (θk1−θk2) becomes (θk1
b − θk2b). In this way, (θk1−θk2) in each possible combination is obtained, and the sum Σθ of absolute values of (θk1−θk2) is set as the evaluation value 1.

Further, in step S12, the amplitude ratio rk1 between the elements of the sample signal 21 and the elements A of the target signal 22.
The amplitude ratio rk2 between F is calculated, and the sum Σr of the absolute values of (rk1 −rk2) is set as the evaluation value 2. At this time, the elements of the sum Σθ obtained in step S11 and the sum Σ obtained in step S12
The elements of r are equal.

For example, the elements of the sample signal 21 and the target signal 2
There are many combinations of the two elements, as shown in FIG. Among these, the amplitude ratio rk1a between the elements of the sample signal 21 and the amplitude ratio rk2a between the elements A and B of the target signal 22.
(Rk1−rk2) becomes (rk1a−rk2a), and considering the amplitude ratio rk1b between the elements of the sample signal 21 and the amplitude ratio rk2b between the elements B and C of the target signal 22,
(Rk1-rk2) becomes (rk1b-rk2b). In this way, (rk1-rk2) in each possible combination is obtained, and the absolute value Σr of (rk1-rk2) is set as the evaluation value 2.

After the evaluation value 1 is obtained in step S11 and the evaluation value 2 is obtained in step S12, the evaluation function is calculated in step S13. The evaluation function is calculated by adding the one obtained by multiplying the evaluation value 1 by the weight α and the one obtained by multiplying the evaluation value 2 by the weight β. The weights α and β are set depending on which of the phase and the amplitude is more important when the evaluation is performed. Weight α, β
May be a fixed value, or may be a variable value depending on the deterioration of the test piece 7 and the usage of the eddy current flaw detection probe 2.

The steps of S11, S12 and S13 are the processes of the evaluation device 16.

In step S14, it is determined whether or not there is a next combination of associations. If there is a next association, the operation proceeds to step S10 and the above-described processing is performed until there is no next association. repeat. Step S14
Is the processing of the changing device 17.

If it is determined in step S14 that there is no next correlation, the correlation having the smallest evaluation function is selected in step S15. When the evaluation function of the selected association and the reference value are compared in step S16 and it is determined that the evaluation function is greater than or equal to the reference value, it means that the sample signal 21 and the target signal 22 are in a state of correlation that cannot correspond to each other. Step S
At 17, it is determined that the sample signal 21 of the test piece 7 is defective. When it is determined in step S16 that the evaluation function is smaller than the reference value, the phases and amplitudes of the elements of the target signal 22 associated with the elements of the sample signal are calculated in step S18.

Steps S14, S15, S16, S17 and S18 are processes of the identifying device 16.

After the phases and amplitudes of the elements of the target signal 22 are calculated in step S18, the calibration parameters are calculated in step S19, that is, the phase change angle and the sensitivity voltage are calculated, and the target signal is calculated in step S20. 22 is calibrated based on the calibration parameters.

Steps S19 and S20 are the processes of the signal calibrating device 19.

As described above, the sample signal 21 and the correlation between the signal elements of the sample signal 21 (sample correlation) and the correlation between the elements A to F of the target signal 22 (target correlation) are compared to each other. The corresponding element of the target signal 22 is specified and the sample signal 21 and the target signal 22 are matched. Therefore, even if the phase and amplitude of the sample signal 21 and the target signal 22 are completely different, the sample signal 21 and the target signal 22 can be correctly matched, and the phase / sensitivity of the eddy current signal can be automatically calibrated.

Then, the signal factor serving as a reference for the phase difference and the amplitude ratio is set as the factor that first exists in the test piece 7, and the characteristics of other signal factors are represented by the phase difference and the amplitude ratio with the signal. If the first factor signal is assumed, matching is performed based on the degree of similar maintenance of other signal factors (with the sample signal). When the matching is worse than the reference value, the correspondence of the first signal factor is changed by determining that the assumption of the first signal factor is wrong. As a result, when it is unclear which part of the test piece 7 is the entire signal length, the test piece 7 part is automatically cut out.

Therefore, when the sample signal 21 and the target signal 22 have completely different phases and amplitudes, the test piece 7 part to be analyzed can be automatically recognized with high accuracy. In addition, when the phase difference and amplitude ratio are similar and one of the multiple signal factors cannot be clearly determined, the signal factors are arranged and arranged (consider not only the signal waveform but also the signal positional relationship).
Can be used to automatically recognize the test piece 7 portion to be analyzed.

In the above-described embodiment, the phase and the amplitude are explained as the signal elements. However, when both the phase and the amplitude are used, the sample signal 21 and the target signal 22 are irrespective of the moving speed of the eddy current flaw detection probe 2. Therefore, it is possible to deal with the problem, and it is possible to suppress the influence even if the driving unit of the eddy current flaw detection probe 2 is deteriorated. It is also possible to use only one of the phase and the amplitude. When either one is used, the sample signal 21 and the target signal 22 can be associated with each other by simple control.

Further, the element interval can be applied as the correlation of the signal elements. By applying the element spacing, deterioration of the eddy current flaw detection probe 2 and the test piece 7
The sample signal 21 and the target signal 22 can be associated with each other regardless of the deterioration of When the element interval is applied, operation such as adding the element interval correlation after a lapse of a predetermined time from the start of automatic calibration can be performed.

By using the above-mentioned automatic calibration device to connect the inspection site and the analysis place through a network and monitor them, not only a fully automatic calibration process but also a semi-automatic process is possible.

[0051]

The automatic eddy current signal calibration apparatus of the present invention comprises:
Sample storage means for storing the characteristics of the sample signal of the test piece, which is the target of sensitivity adjustment, as a time series of sample signal elements,
Target signal extraction means for taking in the target signal to be analyzed and extracting it as a time series of target signal elements, sample signal element correlation deriving means for deriving the sample correlation between the sample signal elements stored in the sample storage means, and the target signal A target signal element correlation deriving means for deriving a target correlation between the target signal elements extracted by the extracting means, and a sample correlation derived by the sample signal element correlation deriving means and a target correlation derived by the target signal element correlation deriving means The sample signal is associated with the target signal based on the associating means, and the automatic calibration means for automatically calibrating the signal status based on the sample signal and the target signal associated by the associating means. By comparing the sample correlation of the signal and the target correlation of the target signal, the corresponding signal elements of the sample signal and the target signal are identified and the sample signal and the target signal are matched. It is possible. As a result, even if the sample signal and the target signal have completely different elements such as phase and amplitude, the sample signal and the target signal can be correctly matched, and the phase / sensitivity of the eddy current signal can be automatically calibrated.

Since the sample correlation and the target correlation are the correlations of the phase angles of the signals, the sample signal and the target signal can be correctly matched by simple control.

Further, since the sample correlation and the target correlation are the correlations of the amplitudes of the signals, the sample signal and the target signal can be correctly matched by simple control.

Further, since the sample correlation and the target correlation are correlations of signal intervals, the sample signal and the target signal can be associated with each other regardless of the deterioration of the sensor or the test piece.

Further, since the sample correlation and the target correlation are the correlation of the phase angle and the amplitude of the signal, the sample signal and the target signal can be associated with each other regardless of the moving speed of the sensor, and the driving means of the sensor is deteriorated. However, the influence can be suppressed.

Since the associating means is provided with a function of comparing the aspect ratio of the circumscribed rectangle of the Lissajous-displayed signal between the sample signal element and the target signal element, provisional association can be easily performed. Useless comparison processing can be eliminated.

In the associating means, the absolute value of the difference between the phase difference value of the target signal element and the phase difference value of the sample signal element is added in an arbitrary time series, and the added value becomes the minimum. Since the function of associating the sample signal with the target signal is provided by determining the time series, the test piece to be analyzed can be automatically recognized with high accuracy by simple control.

In the associating means, the absolute value of the difference between the amplitude difference value of the target signal element and the amplitude difference value of the sample signal element is added in an arbitrary time series, and the added value becomes the minimum. Since the function of associating the sample signal with the target signal is provided by determining the time series, the test piece to be analyzed can be automatically recognized with high accuracy by simple control.

In the associating means, the absolute value of the difference between the phase difference value of the target signal element and the phase difference value of the sample signal element is added in an arbitrary time series, and the added value becomes the minimum. A function that correlates the sample signal with the target signal by determining the time series, and adds the absolute value of the difference between the amplitude difference value of the target signal element and the amplitude difference value of the sample signal element in any time series. , A function of associating the sample signal with the target signal by determining the time series in which the added value is the minimum, and a function of adding different weights to the determination result of the phase angle and the determination result of the amplitude are provided. Therefore, the test piece to be analyzed can be automatically recognized with high accuracy by emphasizing necessary elements.

In the automatic calibration method of the present invention, the characteristics of the sample signal of the test piece, which is the target of the sensitivity adjustment, are stored as a time series of sample signal elements, and the target signal to be analyzed is taken in to obtain the time series of the target signal elements. Based on the sample phase / amplitude correlation and the target phase / amplitude correlation, the target phase / amplitude correlation between the extracted target signal elements is found, as well as the sample phase / amplitude correlation between the extracted and stored sample signal elements is found. Since the sample signal and the target signal are associated with each other and the signal status is automatically calibrated based on the associated sample signal and target signal, the sample correlation of the sample signal is compared with the target correlation of the target signal. As a result, the corresponding signal elements of the sample signal and the target signal can be specified to match the sample signal and the target signal. As a result, even if the sample signal and the target signal have completely different elements such as phase and amplitude, the sample signal and the target signal can be correctly matched, and the phase / sensitivity of the eddy current signal can be automatically calibrated.

Then, the sample interval correlation between the sample signal elements and the object interval correlation between the target signal elements are obtained, and the sample interval and the target signal are associated with each other after the lapse of a predetermined time from the start of automatic calibration. Therefore, the sample signal and the target signal can be associated with each other regardless of the deterioration of the sensor or the test piece.

Further, the aspect ratio of the circumscribed rectangle of the Lissajous-displayed signal is compared between the sample signal element and the target signal element, and the sample signal and the target signal are provisionally associated with each other. It is possible to eliminate unnecessary comparison processing.

[Brief description of drawings]

FIG. 1 is a conceptual diagram of a state in which an inspection is performed by a pipe inspection device including an automatic eddy current signal calibrating device according to an embodiment of the present invention.

FIG. 2 is a block configuration diagram of an automatic calibration device.

FIG. 3 is an explanatory view of a situation of a sample signal and a target signal for a test piece.

FIG. 4 is a graph showing a relationship between a phase angle and a signal element.

FIG. 5 is a flowchart of an automatic calibration method.

FIG. 6 is a flowchart of an automatic calibration method.

FIG. 7 is a conceptual diagram of lattice points of a sample signal and a target signal in temporary association.

FIG. 8 is a conceptual diagram of lattice points of a sample signal and a target signal in evaluation value calculation.

[Explanation of symbols]

1 Piping inspection device 2 Eddy current flaw detection probe 3 Signal transmission unit 4 Flexible member 5 tube sheet 6 piping 7 test pieces 8 Automatic calibration device 11 Sample signal management device 12 Signal capture device 13 Extractor 14 Feature calculation device 15 Temporary association device 16 Evaluation device 17 Change device 18 Signal calibration device

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Terumi Takahama             1-1 1-1 Wadasaki-cho, Hyogo-ku, Kobe-shi, Hyogo             No. Mitsubishi Heavy Industries, Ltd.Kobe Shipyard F term (reference) 2G053 AA11 AB21 CB26 CC01 CC02                       CC07

Claims (12)

[Claims] 1. A sample storage means for storing characteristics of a sample signal of a test piece, which is a target of sensitivity adjustment, as a time series of sample signal elements, and a target signal to be analyzed is taken in and extracted as a time series of target signal elements. Target signal extracting means, sample signal element correlation deriving means for deriving the sample correlation between the sample signal elements stored in the sample storage means, and target for deriving the target correlation between the target signal elements extracted by the target signal extracting means Signal element correlation deriving means, associating means for associating the sample signal and the target signal based on the sample correlation derived by the sample signal element correlation deriving means and the target correlation derived by the target signal element correlation deriving means, An automatic calibration of an eddy current signal, comprising: an automatic calibration means for automatically calibrating a signal condition based on a sample signal and a target signal associated by the associating means. apparatus. 2. The automatic eddy current signal calibrating apparatus according to claim 1, wherein the sample correlation and the target correlation are correlations of phase angles of signals. 3. The eddy current signal automatic calibration device according to claim 1, wherein the sample correlation and the target correlation are correlations of signal amplitudes. 4. The eddy current signal automatic calibration device according to claim 1, wherein the sample correlation and the target correlation are correlations of signal intervals. 5. The automatic eddy current signal calibrating apparatus according to claim 1, wherein the sample correlation and the target correlation are correlations of a phase angle and an amplitude of the signal. 6. The associating means according to claim 1, wherein the associating means has a function of comparing an aspect ratio of a circumscribed rectangle of a Lissajous-displayed signal between a sample signal element and a target signal element. An automatic eddy current signal calibration device, which is provided. 7. The associating means according to claim 2,
By adding the absolute value of the difference between the phase difference value of the target signal element and the phase difference value of the sample signal element in an arbitrary time series, and determining the time series when the added value is the minimum, the sample signal and target An automatic eddy current signal calibration device having a function of correlating signals. 8. The associating means according to claim 3,
By adding the absolute value of the difference between the amplitude difference value of the target signal element and the amplitude difference value of the sample signal element in an arbitrary time series, and determining the time series in which the added value is the minimum, the sample signal and target An automatic eddy current signal calibration device having a function of correlating signals. 9. The associating means according to claim 5,
By adding the absolute value of the difference between the phase difference value of the target signal element and the phase difference value of the sample signal element in an arbitrary time series, and determining the time series when the added value is the minimum, the sample signal and target The function of associating signals and the absolute value of the difference between the amplitude difference value of the target signal element and the amplitude difference value of the sample signal element are added in an arbitrary time series, and the time series with the smallest added value is calculated. The function of associating the sample signal with the target signal by making a judgment, and the function of adding different weights to the judgment result of the phase angle and the judgment result of the amplitude are provided. Automatic calibration device. 10. A characteristic of a sample signal of a test piece, which is a target of sensitivity adjustment, is stored as a time series of sample signal elements,
The target signal to be analyzed is taken in and extracted as a time series of target signal elements, and the sample phase / amplitude correlation between the stored sample signal elements and the target phase / amplitude correlation between the extracted target signal elements are obtained, A vortex characterized by associating a sample signal with a target signal based on the sample phase / amplitude correlation and the target phase / amplitude correlation, and automatically calibrating the signal status based on the associated sample signal and target signal. Automatic calibration method for current signals. 11. The sample signal and target signal according to claim 10, wherein a sample interval correlation between sample signal elements and a target interval correlation between target signal elements are obtained, and the target interval correlation is added after a predetermined time has elapsed from the start of automatic calibration. A method for automatically calibrating eddy current signals, characterized in that 12. The aspect ratio of a circumscribed rectangle of a Lissajous-displayed signal according to claim 10 or 11, is compared between a sample signal element and a target signal element, and the sample signal and the target signal are provisionally associated with each other. Characteristic eddy current signal automatic calibration method.