JP2008309573A - Eddy current flaw detector and eddy current flaw detection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enhance inspection precision by discriminating whether a signal waveform is caused by the flaw of an object to be inspected being a measuring target or noise. <P>SOLUTION: The eddy current flaw detector is equipped with a normal probe 11 and a magnetic saturation type probe 13 equipped with a magnet to reduce the noise caused by a change in magnetic permeability and the signal waveforms obtained by scanning the same place by the normal probe 11 and the magnetic saturation type probe 13 are compared and analyzed by an analyzer 21 to discriminate whether the signal waveforms are caused by the flaw of the object to be inspected of the measuring target or noise. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an eddy current flaw detector and an eddy current flaw detection method.

  Eddy current testing (ECT) is known as a nondestructive inspection method for metals. This is because an eddy current is generated in the member to be measured by the magnetic flux generated by the ECT coil supplied with the excitation current, and a detection signal representing the magnetic flux generated by this eddy current is obtained as an output signal of the ECT coil. Since the detection signal reflects the position, shape, depth, etc. of the defect (scratch) of the subject, the subject is flawed based on the detection signal.

  As described above, the eddy current flaw detection method performs flaw detection by detecting changes in the strength and flow shape of the eddy current caused by defects in the subject. However, the strength and flow shape of such eddy currents are not only due to the above defects, but also due to changes in the electrical resistivity and permeability of the subject, changes in the impedance of the ECT coil caused by changes in the coil posture, etc. Change. Therefore, such a change in the impedance of the ECT coil becomes noise, and the detection accuracy of defects is lowered.

  In particular, when the subject is a magnetic material, or when a magnetic material is present in the vicinity of the subject, the permeability change appears as noise in the detection signal as shown in FIG. It is known to do. As a method of reducing such noise, as shown in FIG. 14C, a permanent magnet or an electromagnet 102 is disposed around the ECT coil 101 to reduce the influence of the permeability change, thereby reducing the influence from the ECT coil 101. Magnetic saturation type eddy current flaw detection methods that improve the S / N ratio of detection signals have been proposed.

  FIG. 14B shows the relationship between the magnetic field strength applied to the magnetic material and the magnetic permeability. The magnetic material has a magnetic permeability greater than that of the non-magnetic material when no magnetic field is applied, and the magnetic permeability increases as the applied magnetic field increases and reaches a peak point. When the applied magnetic field is further increased, the magnetic permeability of the magnetic material decreases and approaches the magnetic permeability of the non-magnetic material, and there is almost no difference between the magnetic permeability of the magnetic material and the magnetic permeability of the non-magnetic material. Is called magnetic saturation. That is, in the magnetic saturation state, the magnetic permeability difference of the subject 100 is small, and the change in eddy current due to the magnetic permeability difference is small. Therefore, if eddy current flaw detection is performed in the magnetic saturation state, the noise is small, Defect detection accuracy is improved.

As an example of such an eddy current flaw detection method using magnetic saturation, an example of an apparatus applied to flaw detection of a magnetic metal tube is disclosed in Japanese Patent Application Laid-Open No. 6-94682 “Electromagnetic induction flaw detection probe”.
JP-A-6-94682

  However, in the magnetic saturation type eddy current flaw detection method described above, if a strong magnet necessary to eliminate noise due to a change in permeability is used, an ECT equipped with a movable mechanism is used in the vicinity of a ferromagnetic material such as carbon steel. The sensor is difficult to move due to the strong magnetic force, and there is a situation that a realistic apparatus that can scan the subject cannot be configured.

  In addition, when applied to flaw detection inspection of dissimilar material joints by alloy welding, noise due to magnetic permeability change occurs at the dissimilar material boundary, but the noise may not be completely erased even with the strongest magnet was there. In addition, in a magnetic saturation type eddy current flaw detector in which the magnetic force of the magnet is slightly weakened in order to achieve a realistic apparatus configuration capable of scanning the subject, noise due to permeability change is not completely eliminated, and the detection signal It was necessary to determine whether the waveform was caused by a scratch on the subject or noise caused by a change in magnetic permeability.

  In addition to noise due to magnetic permeability change, for example, there is shape noise due to ECT sensor lift-off change or ECT sensor tilt change due to the subject surface shape, and this shape noise may show a signal waveform similar to a surface defect. Since there are many cases, it has been difficult to determine whether a detection signal waveform is caused by a surface defect of a subject or a shape noise.

  The present invention has been made to solve the above-described problem, and an object thereof is to provide an eddy current flaw detection apparatus and an eddy current flaw detection method capable of improving inspection accuracy.

In order to solve the above problems, the present invention employs the following means.
The present invention includes an eddy current flaw detection probe, a magnetic saturation type eddy current flaw detection probe that includes a magnet to reduce noise due to magnetic permeability change, and scans the same location by the eddy current flaw detection probe and the magnetic saturation type eddy current flaw detection probe. And analyzing means for comparing and analyzing the signal waveform obtained in step (b) to determine whether the signal waveform is caused by a flaw of the subject to be measured or caused by noise. An eddy current flaw detector is provided.

  According to the present invention, it is possible to determine whether a detection signal waveform is caused by a defect (scratch) of a subject or noise caused by a change in magnetic permeability, thereby improving inspection accuracy. It becomes possible.

  In the eddy current flaw detection apparatus, the magnet of the magnetic saturation type eddy current flaw detection probe has a magnetic force in a range in which a movable body including the eddy current flaw detection probe and the magnetic saturation type eddy current flaw detection probe can scan, or It is preferable to install with a lift-off within a scannable range.

  In the eddy current flaw detection apparatus, the analyzing means may be characterized in that the signal waveform of the eddy current flaw probe has an intrinsic defect, and the signal amplitude of the eddy current flaw probe is a signal amplitude of the magnetic saturation eddy current flaw probe. When the size exceeds a predetermined number of times, it may be determined that the noise is caused by a change in magnetic permeability.

  By doing so, even if the noise due to the permeability change is not completely erased, whether the detection signal waveform is caused by a defect (scratch) of the subject or the noise caused by the permeability change. This makes it possible to determine the accuracy of inspection.

  In the eddy current flaw detection apparatus, the analysis means shows that the signal waveform of the eddy current flaw probe has a characteristic of an opening defect, and the signal amplitude of the magnetic saturation eddy current flaw probe is the signal amplitude of the eddy current flaw probe. When the size exceeds a predetermined number of times, it may be determined that the noise is caused by shape noise due to a lift-off change or a tilt change of the eddy current flaw detection probe and the magnetic saturation eddy current flaw detection probe. .

  By doing in this way, regarding the opening defect and the shape noise having similar waveform characteristics of the detection signal, the detection signal waveform is caused by the opening defect (scratch) of the subject, or is caused by the shape noise. Therefore, it is possible to improve the inspection accuracy.

  In the eddy current flaw detection apparatus, the analysis means inputs a sample flaw detection signal obtained by actually measuring a standard sample with a known flaw depth by the eddy current flaw detection probe and the magnetic saturation type eddy current flaw detection probe, and has waveform characteristics. A feature value generation means for generating a feature value that is quantified, a correct data supply means for supplying known correct data as an amount representing the depth of the scratch of the standard sample, and the feature value and the correct data are input. Learning means for generating by evaluation an evaluation parameter that is a parameter for outputting a value with a sufficiently small error from the correct answer data using the feature amount, and the eddy current flaw detection probe and the subject to be measured A feature value generating means for generating a feature value similar to the sample flaw detection signal based on the actual flaw detection signal obtained by flaw detection with the magnetic saturation type eddy current flaw detection probe, and based on the measured flaw detection signal Based on the Ku features and said evaluation parameter corresponding to the feature quantity, may be provided with a evaluation result generating means for generating depth estimates of wounds.

  In this way, the influence of permeability change and the influence of material difference can be taken into account, so that the evaluation accuracy is markedly improved compared to the case where judgment is made based only on the signal of the eddy current flaw detection probe. It becomes possible to improve.

  In an eddy current flaw detection in which a subject includes two or more kinds of material regions, in addition to the eddy current flaw detection probe and the magnetic saturation eddy current flaw detection probe, a different material boundary position that is a boundary between different materials in the subject is detected. A position detection sensor for detecting a defect based on a signal waveform of the eddy current flaw detection probe or the magnetic saturation type eddy current flaw detection probe based on the position information of the position detection sensor. It is good.

  By doing in this way, it can classify | categorize correctly the position of the material which a detection signal has on the basis of the discriminated foreign material boundary position, and can perform more accurate determination and evaluation.

  In the eddy current flaw detection apparatus, the analysis unit determines a position of a different material boundary that is a boundary of different materials in the subject to be measured based on the signal value distribution of the magnetic saturation type eddy current flaw detection probe of the subject. The defect extraction and sizing based on the signal waveform by the eddy current flaw detection probe or the magnetic saturation type eddy current flaw detection probe may be performed based on the dissimilar material boundary position.

  In this way, without adding a new position sensor, it is possible to accurately classify which material the detection signal is on the basis of the determined different material boundary position, and perform more accurate determination / evaluation Can do.

  In the eddy current flaw detection apparatus, the magnetic saturation type eddy current flaw detection probe is sandwiched between a pair of magnets and the pair of magnets, and is attenuated with respect to the magnetic force at the center of the region sandwiched between the pair of magnets. It is good also as providing the some sensor arranged in parallel in the position where a rate is a predetermined permissible range.

  By doing in this way, a magnetic saturation type eddy current flaw detection probe can be realized also for multi-sensor type probes other than the internal search type or the penetration type.

  The present invention is an eddy current flaw detection method for an eddy current flaw detection apparatus comprising an eddy current flaw detection probe and a magnetic saturation type eddy current flaw detection probe that includes a magnet and reduces noise due to permeability change. By comparing and analyzing signal waveforms obtained by scanning the same location by the probe and the magnetic saturation type eddy current flaw detection probe, whether the signal waveform is caused by a flaw on the subject to be measured, or caused by noise There is provided an eddy current flaw detection method characterized by comprising an analysis step for discriminating whether or not.

  According to the present invention, it is possible to improve the inspection accuracy.

  Hereinafter, embodiments of an eddy current flaw detection apparatus and an eddy current flaw detection method according to the present invention will be described in detail with reference to the drawings.

[First Embodiment]
FIG. 1 is a configuration diagram of an eddy current flaw detector according to a first embodiment of the present invention. In the figure, the eddy current flaw detector according to this embodiment includes a normal probe 11, a magnetic saturation probe 13, a driving device 15, a flaw detector 17, a storage device 19, and an analyzer 21.
The normal type probe 11 is an eddy current flaw detection probe referred to in the claims, and is generally used in maintenance inspection of piping and the like by the eddy current flaw detection method. For example, a self-comparison type is used. The magnetic saturation type probe 13 is a magnetic saturation type eddy current flaw detection probe referred to in the claims, and a permanent magnet or an electromagnet is arranged around the ECT coil so as to reduce the influence of permeability change. It is.

  The driving device 15 moves the movable body 25 including the normal type probe 11 and the magnetic saturation type probe 13, and includes a position detector such as an encoder, and outputs a position signal to the flaw detector 17. Further, the flaw detector 17 excites the ECT coils of the normal probe 11 and the magnetic saturation probe 13, receives the output signals of the ECT coils as detection signals, and performs A / D conversion on these position signals and detection signals. Output to the recording device 19.

  The recording device 19 sequentially records the position signal and the detection signal from the flaw detector 17. The analyzer 21 performs evaluation by performing defect extraction and sizing based on the signal waveforms obtained by the normal probe 11 and the magnetic saturation probe 13. A characteristic is that a signal waveform obtained by scanning the same location by the magnetic probe 11 and the magnetic saturation probe 13 is comparatively analyzed to determine whether or not the signal waveform is caused by noise due to permeability change. There is.

  The eddy current flaw detection apparatus of the present embodiment can be applied to various types of destructive inspections, but in order to clarify the effect of the eddy current flaw detection apparatus of the present embodiment, here, as shown in FIG. A case where the present invention is applied to maintenance inspection of piping including the joint between the container 40 and the piping 50 in such a nuclear power plant will be described as an example.

  FIG. 2B is an enlarged view of the cross section of the joint portion between the pipe A of the portion A and the container 40 in FIG. The container 40 is made of ferromagnetic carbon steel 41 and has an inner surface covered with a SUS overlay 42. As SUS, a non-magnetic austenitic material is used, but it may be weakly magnetized by processing or welding. Further, a SUS base material 51 is used for the pipe 50. The joint between the container 40 and the pipe 50 is covered with an alloy buttering 52 and then joined by alloy welding. The alloy is non-magnetic. In the figure, 53 is an alloy metal.

  At the joint between the container 40 and the pipe 50 (particularly, portion B in the figure), as described in the background art, a magnetic permeability change appears as noise in the detection signal, and the SN ratio decreases. In order to cope with this, the magnetic saturation type probe 13 is used also in the present embodiment. However, if a strong magnet is used to completely eliminate the noise due to the magnetic permeability change, the movable body 25 is near the ferromagnetic body such as carbon steel. However, it becomes difficult to move, and a realistic apparatus capable of scanning the subject cannot be configured. Therefore, in the present embodiment, the magnet of the magnetic saturation type probe 13 is set to have a slightly weak magnetic force within a range that the movable body 25 can scan, or is installed with lift-off (larger lift-off) within the scanable range. As a result, the magnetism is weakened within the scannable range. In the maintenance inspection of the pipe including the joint between the container 40 and the pipe 50, the movable body 25 in which the normal type probe 11 and the magnetic saturation type probe 13 are juxtaposed is scanned in the axial direction and the circumferential direction.

As described above, when the magnetic force of the magnet of the magnetic saturation probe 13 is slightly weakened so as to achieve a realistic apparatus configuration capable of scanning the subject, noise due to permeability change is not completely erased, and inspection is performed. In order to improve accuracy, it is necessary to determine whether the detection signal waveform is caused by a defect of the subject or noise caused by a change in magnetic permeability.
Therefore, in the eddy current flaw detector according to the present embodiment, a signal waveform obtained by scanning the same location by the normal type probe 11 and the magnetic saturation type probe 13 is compared and analyzed, and the signal waveform is caused by noise due to permeability change. It is determined whether it is what to do.

Next, the determination method performed by the analyzer 21 will be described in detail with reference to the flowchart of FIG. In the eddy current flaw detection method, generally, a method of measuring a plurality of excitation frequencies is employed for the purpose of facilitating the feature classification of defects (scratches) based on a signal waveform. In this embodiment, a plurality of frequencies are also used. (For example, 100 [kHz], 200 [kHz], 400 [kHz], etc.) are measured.
Here, an example will be described in which an aperture defect signal is used as a standard signal, that is, calibration is performed so that the amplitude and phase angle of a predetermined aperture defect signal are constant for all flaw detection frequency waveforms.

  First, a normal sensor signal (... Sj ... Sk ...) is acquired by the normal probe 11 (step S111). Here, Sj is a detection signal at the excitation frequency Fj, Sk is a detection signal at the excitation frequency Fk, and Fj <Fk. Similarly, magnetic saturation type sensor signals (... S'j ... S'k ...) are acquired by the magnetic saturation type probe 13 (step S112). Here, S'j is a detection signal at the excitation frequency Fj, and S'k is a detection signal at the excitation frequency Fk.

  Next, whether or not the normal sensor signal by the normal probe 11 shows the inherent defect characteristic, specifically, the amplitude of the signal Sj when the flaw is detected at a low frequency of the flaw detection frequency is a high frequency of the flaw detection frequency. It is determined whether or not the amplitude of the signal Sk when the flaw is detected is larger (step S113). There are two types of defect (scratch) feature types: intrinsic defects that do not open in the scan plane and open defects that open in the scan plane. The characteristics of the signal waveform caused by noise due to magnetic permeability change are inherent defects. It is similar to the characteristics of the signal waveform caused by Further, the characteristic of the signal waveform due to the intrinsic defect has a characteristic of “low frequency signal amplitude> high frequency signal amplitude” as compared with the characteristic of the signal waveform due to the opening defect. A signal having a high possibility of being caused by an intrinsic defect or noise due to a magnetic permeability change is extracted.

  The determination in step S113 may be performed on the magnetic saturation type sensor signal by the magnetic saturation type probe 13. However, when the magnetic saturation type sensor signal is caused by noise due to a change in magnetic permeability, the magnetic saturation type sensor signal is reduced by magnetic saturation. Therefore, it is desirable to perform the normal sensor signal from the normal probe 11.

  If “low-frequency signal amplitude> high-frequency signal amplitude” in step S113, then the amplitude of the normal sensor signal SA by the normal probe 11 when flaw detection is performed at the same frequency is the magnetism by the magnetic saturation probe 13. It is determined whether or not the amplitude exceeds a predetermined number of times the amplitude of the saturation type sensor signal S′A (step S114). Here, the “predetermined number of times” varies depending on the setting of the magnetic force or lift-off magnitude of the magnet of the magnetic saturation probe 13, but is about 3 to 5 times.

In other words, if it is caused by noise due to permeability change, the magnetic saturation type sensor signal is reduced by magnetic saturation, so this judgment determines whether the signal waveform is caused by noise due to permeability change. it can. That is, when “normal sensor signal amplitude >> magnetic saturation type sensor signal amplitude”, it is determined that the signal is caused by noise due to magnetic permeability change (step S117), and “normal sensor signal amplitude >> magnetic saturation type sensor signal”. If it is not “amplitude”, it is determined that there is a possibility of a signal due to an intrinsic defect (step S116).
If “low frequency signal amplitude> high frequency signal amplitude” is not satisfied in step S113, the signal is determined to be a signal due to a factor other than an intrinsic defect or noise due to permeability change (step S115).

Next, the defect extraction based on the signal waveforms obtained by the normal probe 11 and the magnetic saturation probe 13 and evaluation by sizing performed in the analyzer 21 will be briefly described.
In general, in an inspection by an eddy current flaw detection method, a detection signal (voltage) is drawn and evaluated by a Lissajous figure on a complex plane having a real axis (horizontal axis) and an imaginary axis (vertical axis) as shown in FIG. This Lissajous figure is characterized by a magnitude (amplitude) and an inclination (phase angle θ) with respect to the real axis.

  FIG. 5 illustrates Lissajous figures of a normal sensor signal and a magnetic saturation sensor signal caused by noise due to intrinsic defects (axial defects, circumferential defects) and permeability change. FIG. 5 is an example of an ECT waveform when an aperture defect signal is used as a standard signal, that is, when calibration is performed so that the amplitude of a predetermined aperture defect signal has the same value and the phase angle is constant. In addition, the normal probe and the magnetic saturation type probe perform flaw detection with the same excitation frequency, and calibrate a signal obtained by detecting a standard defect with the same calibration condition under the same condition.

  As shown in the figure, for each of the normal sensor signal and the magnetic saturation type sensor signal, the “low frequency signal” may be caused by any of the intrinsic defects (axial defects, circumferential defects) and noise due to permeability change. There is a characteristic of “amplitude> high-frequency signal amplitude”. In addition, for each of the low frequency and the high frequency, “normal sensor signal amplitude≈magnetic saturation type sensor signal amplitude” when it is caused by an intrinsic defect (axial defect, circumferential defect), and any noise caused by permeability change. In the case of the above, “normal sensor signal amplitude >> magnetic saturation type sensor signal amplitude”. Therefore, from FIG. 5 as well, a two-step judgment is made as to whether or not “low-frequency signal amplitude> high-frequency signal amplitude” and “normal sensor signal amplitude >> magnetic saturation sensor signal amplitude”. Thus, it can be easily understood that it can be determined whether or not the signal waveform is caused by noise due to a change in magnetic permeability.

  As described above, in the eddy current flaw detector and the eddy current flaw detection method according to the present embodiment, the normal probe (eddy current flaw probe) 11 and the magnetic saturation probe (which includes a magnet and reduces noise due to a change in permeability) A magnetic saturation type eddy current flaw detection probe) 13, and an analysis device (analysis means) 21 compares and analyzes signal waveforms obtained by scanning the same location by the normal type probe 11 and the magnetic saturation type probe 13, It is determined whether the signal waveform is caused by a flaw on the subject to be measured or caused by noise.

  In the eddy current flaw detector and the eddy current flaw detection method of this embodiment, the normal type probe 11 or the magnetic saturation type probe 13 among the signals obtained by flaw detection under a plurality of frequency conditions by the analyzer (analyzing means) 21. The signal amplitude when the flaw is detected at a relatively low frequency is larger than the signal amplitude when the flaw is detected at a relatively high frequency, and the signal amplitude by the normal probe 11 when the flaw is detected at an arbitrary frequency is a magnetic saturation type. When the signal amplitude exceeds a predetermined number of times the signal amplitude by the probe 13, it is determined that the noise is caused by a change in magnetic permeability.

  Thereby, even when the noise due to the magnetic permeability change is not completely erased, it is possible to determine whether the detection signal waveform is caused by a defect (scratch) of the subject or the noise caused by the magnetic permeability change. It is possible to improve the inspection accuracy.

  Further, in the eddy current flaw detector according to the present embodiment, the magnet of the magnetic saturation probe 13 has a slightly weak magnetic force within the range that the movable body 25 can scan, or lift-off within the scanable range (more Install with a large lift-off). That is, since it can be determined whether or not the detection signal waveform is caused by noise due to permeability change, it is necessary to increase the strength of the magnet for magnetic saturation in the magnetic saturation probe 13 until the noise is completely removed. The device can be configured with a relatively weak magnetic force within a range in which the movable body 25 can scan even in the vicinity of the ferromagnetic material.

[Second Embodiment]
Next, an eddy current flaw detector and an eddy current flaw detection method according to a second embodiment of the present invention will be described. The configuration of the eddy current flaw detector of this embodiment is the same as that of the first embodiment (FIG. 1), and detailed description of each component will be omitted.
However, the analyzer 21 is evaluated by performing defect extraction and sizing based on the signal waveforms obtained by the normal probe 11 and the magnetic saturation probe 13 as in the first embodiment. A signal waveform obtained by scanning the same portion by the probe 11 and the magnetic saturation probe 13 is compared and analyzed, and the signal waveform is caused by shape noise due to a lift-off change or a tilt change of the normal probe 11 and the magnetic saturation probe 13. It is different in that it is determined whether or not it is.

  In addition to the noise due to the magnetic permeability change described in the first embodiment, there is a shape noise due to the lift-off change of the ECT sensor due to the shape of the subject surface and the inclination change of the ECT sensor. In order to reduce this shape noise, for example, methods such as a cross coil method, a self-comparison method, and a mutual induction method have been proposed, but the shape noise due to a sudden lift-off change or inclination change is completely eliminated. This shape noise often shows a signal waveform similar to that of an opening defect. Therefore, in order to improve inspection accuracy, is the detection signal waveform caused by the opening defect of the subject? Alternatively, it is necessary to determine whether it is caused by shape noise.

  Therefore, in the eddy current flaw detector of this embodiment, the signal waveforms obtained by scanning the same location by the normal probe 11 and the magnetic saturation probe 13 are compared and analyzed, and the signal waveforms are caused by shape noise. It is determined whether or not there is.

Next, the determination method performed by the analyzer 21 will be described in detail with reference to the flowchart of FIG.
First, a normal sensor signal (... Sj ... Sk ...) is acquired by the normal probe 11 (step S121). Here, Sj is a detection signal at the excitation frequency Fj, and Sk is a detection signal at the excitation frequency Fk. Similarly, magnetic saturation type sensor signals (... S′j... S′k...) Are acquired by the magnetic saturation type probe 13 (step S122). Here, S′j is a detection signal at the excitation frequency Fj, and S′k is a detection signal at the excitation frequency Fk.

  Next, it is determined whether or not the normal sensor signal from the normal probe 11 shows the characteristics of the signal waveform of the opening defect (step S123). Here, “showing the characteristics of the signal waveform of the opening defect” can be determined, for example, by calibrating under the same conditions as in the first embodiment, with Sj≈Sk and the phase angle of each frequency waveform being substantially equal. . The characteristics of the signal waveform due to the shape noise are similar to the characteristics of the signal waveform due to the opening defect. Based on the determination in step S123, a signal having a high possibility of being caused by an opening defect or shape noise is extracted. Note that the determination in step S123 may be performed on the magnetic saturation sensor signal from the magnetic saturation probe 13.

  When the characteristics of the signal waveform of the opening defect are shown in step S123, the amplitude of the magnetic saturation type sensor signal S′A by the magnetic saturation type probe 13 when flaw detection is performed at the same frequency is the normal sensor by the normal type probe 11. It is determined whether or not the amplitude exceeds a predetermined number of times the amplitude of the signal SA (step S124). Here, the “predetermined number of times” is about twice as large.

  When the detection signal is calibrated so that the standard defect signal waveform has a specified amplitude, when it is caused by an opening defect, the normal sensor signal waveform by the normal probe 11 and the magnetic saturation sensor signal by the magnetic saturation probe 13 are used. There is no significant difference from the waveform, but when it is caused by shape noise, the amplitude of the magnetic saturation sensor signal by the magnetic saturation probe 13 is larger than that of the normal sensor signal by the normal probe 11. There is. This is because the noise reduction method is designed with a coil arrangement that cancels the shape noise, but the magnetic saturation probe 13 with a magnet has a distorted eddy current distribution compared to the normal probe 11 without a magnet. Therefore, it is considered that the effect of canceling the shape noise is reduced.

Therefore, when “magnetic saturation type sensor signal amplitude >> normal sensor signal amplitude” in step S124, it is determined that the signal is caused by shape noise (step S127), and “magnetic saturation type sensor signal amplitude >> normal sensor signal”. If it is not “amplitude”, it is determined that there is a possibility of a signal due to an opening defect (step S126).
Further, when the feature of the signal waveform of the opening defect is not shown in step S123, it is determined that the signal is due to a factor other than the opening defect or shape noise (step S125).
FIG. 7 illustrates a Lissajous figure of a normal sensor signal and a magnetic saturation type sensor signal caused by an opening defect (axial defect, circumferential defect) and shape noise.

  As shown in the figure, for each of the normal sensor signal and the magnetic saturation type sensor signal, the characteristics of similar signal waveforms are shown regardless of any inherent defects (axial defects, circumferential defects) and shape noise. ing. In addition, when each of the low frequency and the high frequency is caused by an opening defect (axial defect, circumferential defect), “magnetic saturation type sensor signal amplitude≈normal sensor signal amplitude”, which is caused by any of the shape noise In this case, “magnetic saturation type sensor signal amplitude >> normal sensor signal amplitude”. Therefore, from FIG. 7, the signal waveform is determined based on the two-stage determination of whether or not the characteristics of the signal waveform of the opening defect are exhibited and whether or not “magnetic saturation type sensor signal amplitude >> normal sensor signal amplitude”. It can be easily understood that it can be determined whether or not it is caused by shape noise.

  As described above, in the eddy current flaw detector and the eddy current flaw detection method according to the present embodiment, the normal probe (eddy current flaw probe) 11 and the magnetic saturation probe (which includes a magnet and reduces noise due to a change in permeability) A magnetic saturation type eddy current flaw detection probe) 13, and an analysis device (analysis means) 21 compares and analyzes signal waveforms obtained by scanning the same location by the normal type probe 11 and the magnetic saturation type probe 13, It is determined whether the signal waveform is caused by a flaw on the subject to be measured, or whether the signal waveform is caused by shape noise due to a lift-off change or a tilt change of the normal probe 11 and the magnetic saturation probe 13.

  In the eddy current flaw detector and the eddy current flaw detection method of the present embodiment, the signal waveform of the normal probe 11 or the magnetic saturation probe 13 indicates the characteristics of the opening defect by the analyzer (analyzing means) 21, and the magnetic When the signal amplitude by the saturation probe 13 exceeds a predetermined number of times the signal amplitude by the normal probe 11, the shape noise due to the lift-off change or inclination change of the normal probe 11 and the magnetic saturation probe 13 is a factor. It is determined that

  Thereby, with respect to the opening defect and the shape noise having similar waveform characteristics of the detection signal, it is determined whether the detection signal waveform is caused by the opening defect (scratch) of the subject or the shape noise. Therefore, the inspection accuracy can be improved.

[Third Embodiment]
Next, an eddy current flaw detector according to a third embodiment of the present invention will be described. The configuration of the eddy current flaw detector of this embodiment is the same as that of the first embodiment (FIG. 1), and detailed description of each component will be omitted.
However, the analysis device 21 is characterized in that a sample flaw detection signal obtained by actually measuring a standard sample with a known flaw depth using a normal probe and a magnetic saturation probe is input, and a feature value obtained by quantifying the waveform feature is generated. A quantity generating means, a correct data supplying means for supplying known correct data as a quantity representing the depth of the scratch on the standard sample, the feature quantity and the correct data are input, and an error between the correct answer data using the feature quantity A learning means for generating an evaluation parameter that is a parameter for outputting a sufficiently small value by learning, and an actual flaw detection signal obtained by flaw detection of a subject to be measured by the normal probe and the magnetic saturation probe, Based on the feature amount generating means for generating the same feature amount as the sample flaw detection signal, the feature amount based on the actually measured flaw detection signal and the evaluation parameter corresponding to the feature amount, Point is a configuration different to and a evaluation result generating means for generating an estimate is.

  In the analyzer 21 of the first embodiment, the phase angle θ is measured, and this phase angle θ is mapped to a characteristic curve indicating the relationship between the phase of the flaw detection signal and the depth of the flaw prepared in advance. Although the depth of the flaw was estimated, the relationship between the two may not necessarily correspond accurately, and the flaw detection accuracy may not be practically sufficient. This is because even if the depth of the scratch is the same, the phase angle θ differs depending on various factors such as the shape of the scratch (for example, length, width, etc.) and the relative positional relationship between the scratch and the coil. In particular, inspection accuracy is often a problem in the case of an internal flaw (a flaw on the inner peripheral surface side of the heat transfer tube) in which the rate of change in the depth of the flaw is small relative to the rate of change in phase of the flaw detection signal.

Therefore, in the eddy current flaw detector of the present embodiment, the “eddy current flaw detection signal evaluation method and apparatus” disclosed in Japanese Patent Laid-Open No. 2002-22708, which has already been proposed by the present inventors, is applied to the present invention. Improve the evaluation accuracy of the depth of flaws detected by flaw detection.
FIG. 8 and FIG. 9 are flowcharts for explaining an analysis and evaluation method for eddy current flaw detection signals in the analysis device 21 of the eddy current flaw detection apparatus of the present embodiment.

First, for each of the normal type probe 11 and the magnetic saturation type probe 13, the respective feature amounts generated by the processing from step S1 to step S3 in FIG. 9 are combined, and the processing up to step S4 is performed using the combined feature amount. The parameters for evaluation are generated in advance.
That is, a sample flaw detection signal obtained by actually measuring a standard sample which is a member to be measured whose depth of flaw is known is input (step S1), a predetermined filtering process (step S2) is performed, and the waveform features are digitized. A feature amount is generated (step S3). A feature amount for learning is generated by combining feature amounts obtained from signals detected by the normal probe 11 and the magnetic saturation probe 13. Here, since the feature quantity is a numeric string, to combine the feature quantities means to connect the numeric strings. Thereafter, referring to the known correct answer data (step S1) as an amount representing the depth of the scratch of the standard sample, the evaluation parameter, which is a parameter for outputting a value with a sufficiently small error from the correct answer data, is combined. It is generated by learning using the feature amount (step S4). As a result, the evaluation parameter A is generated. In addition, the process from step S1 to step S4 is repeated to generate an evaluation parameter B using other waveform features. The evaluation parameters A and B are generated for each type of scratch.

Next, for each of the normal type probe 11 and the magnetic saturation type probe 13, the feature quantity of the actually measured flaw detection signal based on the evaluation parameters A and B is generated from step S5 to step S8 in FIG. Thus, an evaluation feature amount is generated, and evaluation is performed by the processing from step 9 to step S11.
That is, an actual measurement flaw detection signal is input (step S5), a predetermined filtering process (step S6) is performed, and then the type of flaw is classified (step S7), and a feature quantity similar to the sample flaw detection signal is obtained for each classification. Generate (step S8). Further, based on the feature quantity based on the actually measured flaw detection signal and the evaluation parameter A corresponding to the feature quantity, data representing the depth of the flaw is generated (steps S9 and S10). The processing as described above is also performed for the evaluation parameter B. Each data representing the depth of the flaw obtained by the above-mentioned processing is compared and verified (step S11), and when the flaw depth represented by each data is within a predetermined range, the flaw depth is estimated. (Step S12).

  The analysis and evaluation of the actually measured flaw detection signal based on this evaluation parameter will be described as follows with reference to FIG. First, a normal sensor signal (... Sj... Sk ...) is acquired by the normal probe 11 (step S131). Similarly, a magnetic saturation sensor signal (... S'j. ) Is acquired (step S132).

  Next, when a defect is extracted by the process of specifying the factor of the signal waveform shown in the first embodiment or the second embodiment (step S133), based on the evaluation parameter corresponding to the position and range of the defect. The feature amount based on the signal waveform is calculated by the procedure from step S6 to step S9 in FIG. 9 (steps S134 and S135). Note that, after calculating the feature values based on the sensor signals of the normal type probe 11 and the magnetic saturation type probe 13, the feature value obtained by combining these is calculated, but since each feature value is a numerical sequence, It is only necessary to connect numeric sequences in a prescribed order.

  Further, sizing (step S136) is performed in the same procedure as the procedure from step S10 to step S11 in FIG. 9, and an analysis result is obtained.

  As described above, in the eddy current flaw detector according to the present embodiment, a standard sample, which is a member to be measured whose flaw depth is known, is measured in advance using signals detected by a normal probe and a magnetic saturation probe. Since the evaluation parameters obtained by statistically processing the feature quantities obtained by quantifying the features effective for the evaluation based on the sample flaw detection signal obtained in this way are generated, and the analysis flaw detection signal is analyzed and evaluated based on the evaluation parameters, the normal type Compared with the case where the determination is made based only on the magnitude and phase angle of the flaw detection signal of only the probe, the evaluation accuracy can be significantly improved.

[Fourth Embodiment]
Next, an eddy current flaw detector according to a fourth embodiment of the present invention will be described. The eddy current flaw detector according to this embodiment has a configuration in which a position detection sensor is added to the configuration of the first embodiment (see FIG. 1), and detailed description of each component other than the position detection sensor is omitted.

  In the first to third embodiments, the position information is obtained by the position signal from the encoder of the driving device 15, whereas in the present embodiment, different materials are used for the normal probe 11 and the magnetic saturation probe 13. A position detection sensor for detecting the boundary position (for example, a standard comparison type pancake coil or the like) is juxtaposed to accurately detect the different material boundary position in the detection signals of the normal probe 11 and the magnetic saturation probe 13. . In addition, the application of the eddy current flaw detector of the present embodiment assumes a maintenance inspection (see FIG. 2) of the part including the joint between the container 40 and the pipe 50 in the nuclear power plant, as in the first embodiment. The different material boundary position is the boundary between alloy buttering and SUS overlay, and the boundary between the alloy weld and the SUS base material.

FIG. 10 is a flowchart for explaining an analysis and evaluation method for an eddy current flaw detection signal in the analysis device 21 of the eddy current flaw detection apparatus of the present embodiment. Also in this embodiment, in principle, the same analysis and evaluation as in the third embodiment is performed.
That is, a normal sensor signal (... Sj ... Sk ...) is acquired by the normal probe 11 (step S141), and similarly, a magnetic saturation type sensor signal (... S'j ... S'k ...) is obtained by the magnetic saturation probe 13. ) Is acquired (step S142), and a sensor signal (... S "A ...) for detecting the welding boundary is acquired by the position detection sensor (step S143).

  Next, when a defect is extracted by the process of specifying the factor of the signal waveform shown in the first embodiment or the second embodiment based on the determination criterion corresponding to the position and range of the evaluation target signal (step S145). Based on the evaluation parameter corresponding to the position and range of the defect, a feature amount based on the signal waveform is calculated, sizing (step S146), and an analysis result is obtained.

  In the example of FIG. 2, for example, in the example of FIG. 2, the region where the SUS overlay 42 is located on the inner surface, the alloy welded portion (alloy molten metal 53 and alloy buttering 52), the SUS base material portion 51, etc. There are different evaluation criteria for defects (scratches) depending on each part. That is, it is necessary to generate defect determination criteria and sizing evaluation parameters corresponding to the respective parts in advance.

  Based on the sensor signal for detecting the welding boundary of the position detection sensor, it is accurately determined which part the evaluation target signal is in (step S144), and based on the determined result, the evaluation standard for defects (scratches) corresponding to each part is used. Then, defect extraction (step S145) and sizing (step S146) are performed.

  As described above, the eddy current flaw detector according to the present embodiment further includes a position detection sensor that detects which part the evaluation target signal is in, and the analysis device (analysis means) 21 includes position information of the position detection sensor. Based on the above, defect extraction and sizing based on signal waveforms by the normal probe 11 and the magnetic saturation probe 13 are performed.

  As a result, it is possible to accurately determine whether or not the normal type probe 11 and the magnetic saturation type probe 13 are located at the different material boundary position, and the normal type probe 11 or the magnetic saturation type probe 13 is detected based on the determined different material boundary position. It is possible to accurately classify signals and perform more accurate determination / evaluation. In particular, even when the welding boundary is not a straight line or when there is an error in the position information of the scanning tool (such as the position signal of the encoder), it is possible to accurately grasp which part the detection signal belongs to.

[Fifth Embodiment]
Next, an eddy current flaw detector according to a fifth embodiment of the present invention will be described. The configuration of the eddy current flaw detector of this embodiment is the same as that of the first embodiment (FIG. 1), and detailed description of each component will be omitted.
However, the analyzer 21 determines the position of a different material boundary, which is a boundary of different materials in the subject to be measured, based on the signal value (zero point deviation) distribution of the magnetic saturation probe 13 of the subject, and the different material. The difference is that defect extraction and sizing based on the signal waveform by the normal probe 11 or the magnetic saturation probe 13 is performed based on the boundary position. That is, the different material boundary position is discriminated based on the magnetic saturation type sensor signal of the magnetic saturation type probe 13 by the analyzer 21 without newly adding a position detection sensor as in the fourth embodiment. is there.

  In the magnetic saturation type probe 13, the zero point (for example, the median value of the signal value) shifts due to the variation of the material of the subject. By examining the distribution of the zero point deviation, the dissimilar material boundary position (welding boundary) is accurately determined. It is possible to determine. In the case of a sensor method in which the drift coil reduction effect is present in the sensor itself, such as a cross coil method, it is difficult to determine the zero point deviation at the different material boundary position (welding boundary) by the normal probe 11.

FIG. 11 is a flowchart for explaining an analysis and evaluation method for an eddy current flaw detection signal in the analysis device 21 of the eddy current flaw detection apparatus of the present embodiment. Also in this embodiment, in principle, the same analysis and evaluation as in the third embodiment is performed.
That is, a normal sensor signal (... Sj ... Sk ...) is acquired by the normal probe 11 (step S151), and similarly, a magnetic saturation type sensor signal (... S'j ... S'k ...) is obtained by the magnetic saturation probe 13. ) Is acquired (step S152).

Next, when a defect is extracted by the process of specifying the factor of the signal waveform shown in the first embodiment or the second embodiment based on the determination criterion corresponding to the position and range of the evaluation target signal (step S155). Based on the evaluation parameter corresponding to the position and range of the defect, a feature quantity based on the signal waveform is calculated, sizing is performed (step S156), and an analysis result is obtained.
Based on the signal S′B having one frequency among the flaw detection signals of the magnetic saturation sensor of the magnetic saturation probe 13, it is accurately determined which part the evaluation target signal is in (step S 154), and based on the determination result. The defect extraction (step S155) and sizing (step S156) processing is performed using the defect (scratch) evaluation criteria corresponding to each part.

  As described above, in the eddy current flaw detector according to the present embodiment, based on the zero point deviation distribution of the magnetic saturation probe 13 due to the variation of the subject material, the different material boundary that is the boundary of different materials in the subject to be measured. And the defect extraction and sizing based on the signal waveform by the normal type probe 11 or the magnetic saturation type probe 13 are performed based on the different material boundary position.

  Accordingly, it is possible to accurately determine whether or not the normal type probe 11 and the magnetic saturation type probe 13 are positioned at the different material boundary position without newly adding a position sensor, and the normal type probe is determined based on the determined different material boundary position. 11 or the signals detected by the magnetic saturation probe 13 can be accurately classified, and more accurate determination / evaluation can be performed. In particular, even when the welding boundary is not a straight line or when there is an error in the position information of the scanning tool (such as the position signal of the encoder), it is possible to accurately grasp which part the detection signal belongs to. Furthermore, since there is an effect of reducing excess noise as compared with the position detection sensor, it is possible to detect the dissimilar material boundary position with high accuracy.

[Sixth Embodiment]
Next, an eddy current flaw detector according to a sixth embodiment of the present invention will be described. The configuration of the eddy current flaw detector of this embodiment is the same as that of the first embodiment (FIG. 1), and detailed description of each component will be omitted.
However, with respect to the analyzer 21, the zero of the magnetic saturation probe 13 due to the variation of the electrical resistivity or permeability of the subject is obtained using a signal obtained by applying a preprocessing filter to the magnetic saturation sensor signal of the magnetic saturation probe 13. Based on the point deviation distribution, the position of a different material boundary, which is a boundary between different materials in the subject to be measured, is determined, and based on the different material boundary position, defect extraction based on the signal waveform by the normal probe 11 or the magnetic saturation probe 13 is performed. And sizing is different.

  In the eddy current flaw detector according to the fifth embodiment, when many signals other than the zero point fluctuation are generated, it is difficult to accurately determine the different material boundary position (welding boundary). By processing using a signal obtained by applying a pre-processing filter (smoothing filter) to the magnetic saturation type sensor signal, it is possible to easily determine the position of the different material boundary based on the zero point fluctuation.

  There are smoothing filters such as a low-pass filter, average value filter or median filter as pre-processing filters, but a low-pass filter is used when high-speed processing is required, and a median filter is used to reduce waveform changes. It may be selected according to the purpose of use.

FIG. 12 is a flowchart for explaining an analysis and evaluation method for an eddy current flaw detection signal in the analysis device 21 of the eddy current flaw detection apparatus of the present embodiment. Also in this embodiment, in principle, the same analysis and evaluation as in the third embodiment is performed.
That is, a normal sensor signal (... Sj ... Sk ...) is acquired by the normal probe 11 (step S161), and similarly, a magnetic saturation type sensor signal (... S'j ... S'k ...) is obtained by the magnetic saturation probe 13. ) Is acquired (step S162).

  Next, when a defect is extracted by the process of specifying the factor of the signal waveform shown in the first embodiment or the second embodiment based on the determination criterion corresponding to the position and range of the evaluation target signal (step S165). Based on the evaluation parameter corresponding to the position and range of the defect, a feature amount based on the signal waveform is calculated, sizing (step S166), and an analysis result is obtained.

  The magnetic saturation type sensor signal of the magnetic saturation type probe 13 is pre-processed by a smoothing filter (step S163), and the pre-processed magnetic saturation type sensor signal accurately determines which part the evaluation target signal is in. Determination (step S164) is performed, and defect extraction (step S165) and sizing (step S166) are performed using a defect (scratch) evaluation criterion corresponding to each part based on the determination result.

  As described above, in the eddy current flaw detector according to the present embodiment, a signal obtained by applying a preprocessing filter (smoothing filter) to the magnetic saturation sensor signal of the magnetic saturation probe 13 is used and based on the zero point deviation distribution. Since the different material boundary position determination is performed, in addition to the effect of the fifth embodiment, even when many signals other than the zero point variation are generated, the determination of the different material boundary position based on the zero point variation can be easily performed. Further, the different material boundary position can be accurately grasped using the signal of the flaw detection frequency for defect detection without adding the optimum flaw detection frequency for detection of the different material boundary position (welding boundary).

[Seventh Embodiment]
Next, an eddy current flaw detector according to a seventh embodiment of the present invention will be described. FIG. 13 is an external view of the magnetic saturation probe 13 in the eddy current flaw detector according to the present embodiment. In the figure, the magnetic saturation type probe 13 is sandwiched between a pair of magnets 60 and a pair of magnets 60, and the attenuation rate with respect to the magnetic force at the center of the region sandwiched between the pair of magnets 60 is a predetermined allowable value. It is the structure provided with the some sensors 61-63 arranged in parallel by the position of the range.

  Conventionally, a magnetic saturation type probe is composed of a combination of a pair of magnets for one sensor, and it has been difficult to dispose it at a close distance without a dead zone. That is, it has been difficult to manufacture a multi-sensor magnetic saturation type sensor using probes other than the internal search type and the penetration type. In other words, when there is an end in the arrangement of the multi-coil such as a flat plate type or a curved surface type, the influence of the magnet arranged in the adjacent coil is not uniform between the end and the center, so the sensor characteristics of the multi-coil are kept uniform. It was difficult.

  In addition, when magnets are arranged near the sensors of a multi-sensor probe, the signal characteristics of each sensor will be non-uniform unless the magnets are arranged so as to have the same relative position with respect to the position of each sensor. However, a magnetic saturation type probe can be configured even for a multi-sensor type probe by arranging the entire plurality of sensors between a pair of sufficiently long plate-shaped (or bar-shaped) magnets 60. Since the plate magnet 60 has a magnetic force drop near the end, the multi-coil is disposed at a position where the attenuation rate with respect to the central magnetic force is within an allowable range.

  As described above, according to the eddy current flaw detector of the present embodiment, the magnetic saturation probe 13 can be realized even for multi-sensor probes other than the internal search type and the penetration type. By applying to 1st Embodiment-6th Embodiment, the eddy current flaw detector which can perform a maintenance test | inspection at higher speed and with high precision is realizable.

1 is a configuration diagram of an eddy current flaw detector according to a first embodiment of the present invention. It is explanatory drawing of piping containing the joint of the container and piping to which the eddy current flaw detector of the 1st Embodiment of this invention is applied. It is a flowchart explaining the determination method in the analyzer of the eddy current flaw detector of the 1st Embodiment of this invention. It is explanatory drawing which illustrates a Lissajous figure. It is explanatory drawing which illustrates the Lissajous figure of the normal sensor signal and magnetic saturation type | mold sensor signal resulting from the intrinsic defect (an axial defect, a circumferential defect) and the noise by a magnetic permeability change. It is a flowchart explaining the determination method in the analyzer of the eddy current flaw detector of the 2nd Embodiment of this invention. It is explanatory drawing which illustrates the Lissajous figure of the normal sensor signal and magnetic saturation type sensor signal resulting from an opening defect (an axial defect, a circumferential defect) and shape noise. It is a flowchart explaining the analysis evaluation method of the eddy current test signal in the analyzer of the eddy current test apparatus of the 3rd Embodiment of this invention. It is a flowchart explaining the analysis evaluation method of the eddy current test signal in the analyzer of the eddy current test apparatus of the 3rd Embodiment of this invention. It is a flowchart explaining the analysis evaluation method of the eddy current test signal in the analyzer of the eddy current test apparatus of the 4th Embodiment of this invention. It is a flowchart explaining the analysis evaluation method of the eddy current test signal in the analyzer of the eddy current test apparatus of the 5th Embodiment of this invention. It is a flowchart explaining the analysis evaluation method of the eddy current test signal in the analyzer of the eddy current test apparatus of the 6th Embodiment of this invention. It is an external view of the magnetic saturation type | mold probe in the eddy current flaw detector of the 7th Embodiment of this invention. It is explanatory drawing explaining the conventional magnetic saturation type eddy current flaw detection method.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Normal type probe 13 Magnetic saturation type probe 15 Drive device 17 Flaw detector 19 Memory | storage device 21 Analyzer 40 Container 41 Carbon steel 42 SUS overlay 50 Piping 51 SUS base material 52 Alloy buttering 53 Alloy molten metal 60 Magnets 61-63 Sensor

Claims (9)

An eddy current testing probe,
A magnetic saturation type eddy current flaw detection probe that includes a magnet and reduces noise due to magnetic permeability change;
By comparing and analyzing signal waveforms obtained by scanning the same location by the eddy current flaw detection probe and the magnetic saturation type eddy current flaw detection probe, whether the signal waveform is caused by a flaw on the subject to be measured or noise An eddy current flaw detector comprising: an analyzing means for discriminating whether or not it is caused by   The magnet of the magnetic saturation type eddy current flaw detection probe has a magnetic force that can be scanned by the movable body including the eddy current flaw detection probe and the magnetic saturation type eddy current flaw detection probe, or is lifted off within a scannable range. The eddy current flaw detector according to claim 1 installed.   In the analyzing means, the signal waveform of the eddy current flaw probe shows a characteristic of an inherent defect, and the signal amplitude of the eddy current flaw probe exceeds a predetermined number of times the signal amplitude of the magnetic saturation eddy current flaw probe. When this is the case, the eddy current flaw detector according to claim 1, wherein the eddy current flaw detector is determined to be caused by noise due to a change in magnetic permeability.   In the analyzing means, the signal waveform obtained by the eddy current flaw detection probe is characterized by an opening defect, and the signal amplitude obtained by the magnetic saturation type eddy current flaw probe exceeds a predetermined number of times the signal amplitude obtained by the eddy current flaw detection probe. The eddy according to claim 1 or 2, wherein it is determined that a shape noise due to a lift-off change or a tilt change of the eddy current flaw detection probe and the magnetic saturation eddy current flaw detection probe is a factor. Current flaw detector. The analysis means includes
Generates a feature value by generating a sample value obtained by quantifying the waveform feature by inputting a sample flaw detection signal obtained by actually measuring a standard sample with a known flaw depth by the eddy current flaw detection probe and the magnetic saturation type eddy current flaw detection probe. Means,
Correct data supply means for supplying known correct data as a quantity representing the depth of the scratch of the standard sample;
Learning means for inputting the feature quantity and correct answer data, and generating an evaluation parameter, which is a parameter for outputting a value having a sufficiently small error from the correct answer data using the feature quantity, by learning;
A feature amount generating means for generating a feature amount similar to a sample flaw detection signal based on an actual flaw detection signal obtained by flaw detection of a subject to be measured by the eddy current flaw detection probe and the magnetic saturation type eddy current flaw detection probe;
5. An evaluation result generating means for generating an estimated value of a flaw depth based on a feature quantity based on the actually measured flaw detection signal and the evaluation parameter corresponding to the feature quantity. The eddy current flaw detector described. In an eddy current flaw detection in which a subject includes two or more kinds of material regions, in addition to the eddy current flaw detection probe and the magnetic saturation eddy current flaw detection probe, a different material boundary position that is a boundary between different materials in the subject is detected. A position detection sensor for
The said analysis means performs the defect extraction and sizing based on the signal waveform by the said eddy current flaw detection probe or the said magnetic saturation type eddy current flaw detection probe based on the position information of the said position detection sensor. The eddy current flaw detector described in 1.   The analysis means determines the position of a different material boundary that is a boundary of different materials in the subject to be measured based on the signal value distribution of the magnetic saturation type eddy current flaw detection probe of the subject, and based on the different material boundary position, The eddy current flaw detector according to any one of claims 1 to 5, wherein defect extraction and sizing are performed based on a signal waveform by the eddy current flaw detection probe or the magnetic saturation type eddy current flaw detection probe.   The magnetic saturation type eddy current flaw detection probe is sandwiched between a pair of magnets and the pair of magnets, and the attenuation rate with respect to the magnetic force at the center of the region sandwiched between the pair of magnets is within a predetermined allowable range. The eddy current flaw detector according to any one of claims 1 to 7, further comprising a plurality of sensors arranged side by side. An eddy current flaw detection method for an eddy current flaw detection apparatus comprising an eddy current flaw detection probe and a magnetic saturation type eddy current flaw detection probe that includes a magnet and reduces noise due to magnetic permeability change,
By comparing and analyzing signal waveforms obtained by scanning the same location by the eddy current flaw detection probe and the magnetic saturation type eddy current flaw detection probe, whether the signal waveform is caused by a flaw on the subject to be measured or noise An eddy current flaw detection method including an analysis step for discriminating whether it is caused by a fault.

Images