JP5150302B2 - Ultrasonic inspection data evaluation apparatus and ultrasonic inspection data evaluation method - Google Patents

Description

  The present invention relates to an ultrasonic inspection data evaluation apparatus and ultrasonic inspection data for detecting a crack of an inspection object from ultrasonic inspection data obtained by transmitting an ultrasonic wave to the inspection object and receiving a reflected ultrasonic signal. It relates to the evaluation method.

  In general, an ultrasonic inspection apparatus transmits ultrasonic waves to an object to be inspected, receives a reflected ultrasonic signal from a crack or the like, and detects a crack or the like of the object to be inspected from the received ultrasonic inspection data. It is a device. Such an ultrasonic inspection apparatus is used, for example, for crack inspection of a reactor shroud. Ultrasonic inspection data often becomes complicated due to the influence of reflection signals from a plurality of ultrasonic propagation paths, noise, and the like, in addition to a reflection signal from a crack. Therefore, in order to detect a crack from this complicated ultrasonic inspection data, the present situation depends on the skill of a skilled inspector.

  An example of the ultrasonic inspection method will be described with reference to FIG. When inspecting the inspection object 21, a high-voltage pulse voltage is applied by the ultrasonic flaw detector 22, and an ultrasonic sensor 24 that transmits and receives the ultrasonic wave 23 is scanned on the inspection surface to observe changes in the ultrasonic signal 25. To do. If there is an echo that appears to be a reflection from the crack 26, the tip of the crack is identified from the echo change by scanning around that location. If the surface to be inspected is a simple shape such as a flat surface, the inspection range is narrow, and the assumed defect is a simple shape such as a slit, etc. There is a possibility that the defect can be detected with sufficient accuracy by a process of applying a time gate and a threshold to the data, for example.

  However, in the actual ultrasonic inspection, the inspection surface has unevenness, and the crack shape is also targeted for a complicated shape such as SCC (Stress Corrosion Cracking). There are many. For this reason, it is difficult to automate the inspection by a simple method as described above, and a skilled inspector detects defects while looking at the data manually. When the inspection range is wide, the probe is placed on a moving mechanism and scanned, and ultrasonic inspection data is collected together with the inspection position information. One-dimensional scanning can be performed electronically by arranging a plurality of probes side by side. By analyzing the ultrasonic signal obtained in this manner, the position and depth of the crack are obtained.

  In ultrasonic inspection, since the range that can be inspected by one inspection is narrow, in order to inspect a range having a certain area, the amount of ultrasonic inspection data becomes enormous, and data processing for obtaining defect data from the data The amount of work is also enormous.

  Therefore, in order to perform an ultrasonic inspection, a skilled worker needs to manually evaluate a huge amount of ultrasonic data. Moreover, in this evaluation work, since the evaluation varies depending on the skill of the inspector, the current situation is that a plurality of inspectors perform the evaluation a plurality of times, and improvement of inspection efficiency is desired.

As an apparatus for evaluating ultrasonic inspection data for increasing inspection efficiency, there is a device that devise a method of applying a time gate and determine the presence or absence of a crack using a neural network (see, for example, Patent Document 1).
JP-A-9-171006

In the conventional technique described above, there is a problem that it is difficult to perform data evaluation automatically and accurately due to noise or pseudo echo in the ultrasonic signal. In addition, as described above, the technique for determining the presence or absence of cracks using a neural network cannot identify the position of the crack end, etc. There was a problem that the inspector had to do it manually. In addition, this technique is intended for inspection at one point, and there is a problem that it is difficult to evaluate an inspection in a range having an area that requires scanning.
The present invention has been made to solve the above-described problems, and an ultrasonic inspection data evaluation apparatus capable of automatically detecting the presence / absence of a crack, the position of a crack end, and the like more accurately than before. It is intended to provide an ultrasonic inspection data evaluation method.

One aspect of the ultrasonic inspection data evaluation apparatus of the present invention, obtained from a plurality of inspection positions on the object to be inspected by transmitting ultrasonic waves, the ultrasonic inspection data obtained by receiving the reflected ultrasound signals for each inspection position Extracting a local maximum value for each of the ultrasonic inspection data, detecting continuity of the local maximum value, calculating a trajectory corresponding to the inspection position with respect to the local maximum value, and a feature amount calculating unit that forms a feature amount; A crack feature quantity database containing the maximum feature value of the reflected ultrasonic signal from the crack, data relating to the maximum value in the ultrasonic inspection data, and a crack received from the crack feature quantity database. A crack determination unit that compares and evaluates the feature value of the maximum value of the reflected ultrasonic signal to determine the presence / absence of a crack and the end position of the crack, and the crack determination unit has the trajectory Has two maxima at different levels And wherein the determined that Ri裂有come when assessed.

In one aspect of the ultrasonic inspection data evaluation method of the present invention, ultrasonic waves are transmitted from a plurality of inspection positions to an object to be inspected, and from the ultrasonic inspection data obtained by receiving a reflected ultrasonic signal for each inspection position , Extracting a local maximum value for each ultrasonic inspection data, detecting a continuity of the local maximum value, calculating a trajectory corresponding to the inspection position with respect to the local maximum value, and obtaining the characteristic amount; and the ultrasonic inspection data Compares and evaluates the data related to the local maximum value and the maximum value of the reflected ultrasonic signal from the crack stored in the crack feature database to determine the presence or absence of the crack and the end position of the crack. A crack determination step , wherein the crack determination step determines that there is a crack when it is evaluated that the trajectory has two maximum values at different levels .

  According to the present invention, the presence / absence of a crack, the position of a crack end, and the like can be automatically detected with higher accuracy than in the past.

  Embodiments of an ultrasonic inspection data evaluation apparatus and ultrasonic inspection data evaluation method according to the present invention will be described below with reference to the drawings. First, a first embodiment will be described with reference to FIG. In the present embodiment, the movement (trajectory) of the maximal value accompanying the scanning of the inspection position is used as the feature value of the maximal value.

  The ultrasonic inspection data evaluation apparatus according to the present embodiment extracts a local maximum value from the ultrasonic inspection data 1 at each inspection position, and detects continuity of the local maximum value with the moved inspection position as the local maximum value data. A feature amount calculation device 2 for obtaining a locus of a maximum value at the time of position scanning, a crack feature amount database 3 for storing a crack feature amount indicating a feature amount of each reflected ultrasonic signal when there is a crack, and a feature A crack determination device 4 that compares the output of the quantity calculation device 2 with the data in the crack feature database 3 to determine the presence or absence of a crack and identify the end position, and the crack depth from the identified end position A crack measuring device 5 for measuring the sheath position and a result display / storing device 6 for displaying and storing the measurement result are provided. These functions can be realized by a single computer.

  Next, a crack determination method based on the maximum value of the ultrasonic inspection data 1 will be described. Here, a case where the ultrasonic inspection is performed by scanning in one direction (one dimension) will be described. As shown in FIG. 8, this one-dimensional scanning is performed by a method of moving one ultrasonic sensor in one dimension, or by a phased array sensor in which a plurality of (for example, 256) ultrasonic sensors are arranged in a row. This can be done by a measuring method. An example of ultrasonic inspection data that changes according to scanning is shown in FIG. In FIG. 2, 10 graphs show the waveforms of reflected ultrasonic signals at 10 scanning positions of pos1-10, respectively. The vertical axis represents the ultrasonic level (%), and the horizontal axis represents time (μs). Is shown. In this case, the interval between the positions (scanning positions) of pos1-10 is about 1 mm or less. The reflected ultrasonic signal only needs to reflect and return one wave reflected from the inspection object, but the reflected ultrasonic signal actually collected becomes complicated data due to the influence of damping and the like. However, only the first wave, that is, the maximum value is meaningful.

  Although the data shown in FIG. 2 is a complex waveform, it is only necessary to focus on the maximum value as a meaningful wave. In FIG. 2, the maximum value in the vicinity of 5 μs is a reflected wave from the surface or the like, and is not used for crack determination. Assuming that the vicinity of 8 to 15 μs is a region where there is a possibility of cracking, pay attention only to the maximum value of the ultrasonic waveform that is the displayed ultrasonic inspection data, and depending on how the position of the maximum value changes according to the scan. Identify the end of the crack.

  In the case of the ultrasonic data shown in FIG. 2, points that are considered to be maximum values are indicated by circles. Among these, the points surrounded by dotted circles at the scanning positions pos2,3,4,8,10 do not appear when the scanning positions are pos1,5,6,7,9, and are smaller than 20% in level. Since it is a point, it can be determined as noise.

  On the other hand, since the points surrounded by a solid circle appear continuously at the scanning position of pos1-10, it can be determined that the ultrasonic signal is from the crack. Examining the movement of this maximum value, this maximum value first takes a maximum value at pos3 from pos1 to pos10, once decreases at pos4, and then the level of the ultrasonic signal continues to increase to pos10. % Is clearly different from noise. Although not shown in the figure, the level continues to increase after pos10, reaches the maximum value (83%) at pos14, and the signal level decreases at the subsequent position, making it indistinguishable from the signal from the surface to be inspected.

  There is no local maximum at a point before Pos1. Assuming that the maximum point pos10 (pos14) of the maximum value is a reference point, the maximum value toward pos1 takes the minimum value at pos4, and the maximum value again at pos3. FIG. 3 shows such a locus of local maximum values with the vertical axis representing level (%) and the horizontal axis representing time (μs). It can be determined that the position of pos3 is the ultrasonic signal from the tip (end) of the crack from the movement (trajectory) of the maximum value.

  Thus, from a plurality of ultrasonic inspection data obtained by scanning the ultrasonic probe, it is possible to grasp the connection between the maximum values and identify the crack position and the like. Here, an example in which scanning is performed in one direction has been described, but when inspecting a range having an area, it is necessary to scan in two directions.

In the case of the above example, it can be determined that “there is a crack” from the following matters.
(1) There is a local maximum value of 1 (pos14)
(2) There is a continuous maximum (by scanning from pos10 to pos1)
(3) There is a maximum value 2 in the continuous change of the maximum value (pos3)
(4) The level of the local maximum 2 is low and is at the end of the continuous local maximum.
Since the movement of the maximum value differs depending on the shape and depth of the crack, it is considered that the trajectory shown in FIG. 3 and the trajectories from other cracks hardly coincide with each other in terms of distance and movement. Therefore, it is necessary to pattern the locus of the maximum value of the echo from the crack and identify the presence and depth of the crack from the degree of coincidence with the collected data.

  In this embodiment, the feature quantity for determining the presence / absence of the crack and the end position of the crack is made into a database, and as shown in FIG. 4, the feature quantity of the locus of the maximum value obtained by the following procedure is compared with the database. Automate the determination of cracks.

First, the feature quantity computing device 2 inputs ultrasonic data (101). Next, the feature quantity computing device 2 obtains maximum value data (maximum value and its position) at each inspection position from the inputted ultrasonic data as follows (102). That is, the feature quantity computing device 2 first obtains the maximum value within n points from each ultrasonic data Data (t). Here, for simplicity, t = n × N.
Data2 (t ′) = maximum value of Data (t) to Data (t + n) Next, thinning processing is performed at n points.
Data3 (N) = Data2 (N)
Next, the maximum value of the thinned data is detected.
When Data3 (N-1) <Data3 (N) and Data3 (N)> Data3 (N + 1)
Max_pos (i) = N
Max_lv (i) = Data3 (N) (i is the number of detected maximum values)
(Max_pos (i): position obtained as local maximum, Max_lv (i): maximum obtained)
By this processing, the maximum value data at each inspection position can be obtained. Since the thinning process is performed, the original data: Data (Max_pos (i) × n) and Max_lv (i) may be misaligned, and it is necessary to correct the position obtained as the maximum value.

Next, the feature quantity computing device 2 obtains the locus of the maximum value of the signal from the crack as follows (103). That is, the inspection position is moved by scanning, but when several local maximum values are detected at these two inspection positions, the local maximum signal from the same reflection source is detected. This is considered to be a signal from the same reflection source when a local maximum value is obtained and when some rules are met. An example of the rule is shown below.
(I) The case where the positional deviation between the position Max_pos1 (m) of the maximum value at pos1 and the position Max_pos2 (n) of the maximum value at pos2 is less than or equal to the threshold value.
(II) The same processing is repeated at a plurality of positions, and continuous maximum values are detected at inspection positions that are continuous at x points or more.
Since noise does not appear continuously, only the locus of the maximum value of the signal from the crack can be obtained by the above processing.

Next, the crack determination device 4 compares and calculates whether the locus of the maximum value (feature amount) calculated by the feature amount calculation device 2 matches the feature of the crack in the crack feature amount database 3. Then, the presence or absence of a crack is determined (104). First, the crack feature database 3 will be described. The crack feature database 3 quantifies the above evaluation method. For example,
(1) There is a local maximum value of 1 (pos14)
(2) There is a continuous maximum (by scanning from pos10 to pos1)
(3) There is a maximum value 2 in the continuous change of the maximum value (pos3)
(4) The level of the local maximum 2 is low and is at the end of the continuous local maximum.
When quantified, it becomes as follows.
(1) The maximum level of the locus of maximum value is a% or more (2) The number of points constituting the locus is b points or more (3) There are two local maximum values in the locus (4) Within c points from both ends of the locus There is a maximum value of level d% or less.
The numerical values a to d are fixed threshold values that vary depending on the ultrasonic inspection method. In addition, as a characteristic amount of cracks,
(5) A point where the level of the maximum value is e% or more continues for f points or more. (6) A condition is set such that the locus slope is g or more, and data satisfying the following conditions (a) and (b) is selected. It shall be recognized as a fissure.
(A) When the conditions of (1) to (4) are satisfied, there is a crack, and the position of the end is the position of the maximum value 2.
(B) When the conditions of (1), (2), and (5) are satisfied, there is a crack, and the position of the end is the point with the longest time.

  The crack determination device 4 compares and calculates whether the feature amount calculated by the feature amount calculation device 2 matches the crack feature in the crack feature amount database 3, and determines the presence or absence of a crack. . For example, in the case of the locus shown in FIG. 3 described above, since the collected data satisfies the conditions (1) to (4), there is a crack in this data.

  If a crack exists, the position of the end of the next crack is obtained (105). In the above case, it is determined that the end of the crack has a maximum value of 2. If it is determined that there is no crack, the result display / storage device 6 displays and stores the result (107).

Next, the crack measuring device 5 automatically measures the depth and position of the crack based on the fact that the end of the crack has a maximum value of 2 and the flaw detection conditions (106). For example, when the time at position pos2 is t, the inspection object is a steel material having a sound speed of 5600 m / s, and the flaw detection angle is set to 45 degrees, the crack depth can be obtained by the following calculation.
t (μs) × 5600 (m / s) × sin 45 (degrees) × 0.5

  The result display / storage device 6 displays and stores the above result and presents the information to the inspector (107). In the above example, the case where the crack feature value is used as data as the crack feature value database 3 has been described. However, the feature value when there is no crack can also be used as the data. By the above processing, the presence / absence and depth of a crack can be automatically measured from ultrasonic data.

  Next, a second embodiment of the ultrasonic inspection data evaluation apparatus according to the present invention will be described with reference to FIG. The basic configuration of the second embodiment is the same as that of the first embodiment described above, and corresponding portions are denoted by the same reference numerals, and redundant description is omitted. In the second embodiment, after the crack is automatically determined, it is possible to intervene the judgment of the inspector in the determination result by the crack manual determination device 7, and the crack feature value database 3 is manually updated. A feature quantity data registration device 8 is provided.

  In the first embodiment described above, the crack determination is automated, but the reflected ultrasonic wave from the crack differs depending on the shape and depth of the crack, and is determined by the information stored in the crack feature database 3. It may not be possible or it may be misjudged. For this reason, in the second embodiment, the result of automatic determination is presented to the inspector, and by making this correctable, the determination can be made manually to prevent erroneous determination. At this time, by displaying the crack end position and the locus of the maximum value obtained by the method shown in the first embodiment, it is possible to easily determine whether the determination is erroneous. Further, if there is an erroneous determination, it is possible to improve the detection accuracy by examining the reason and updating the crack feature amount database 3. The crack feature quantity database 3 can be updated by the feature quantity data registration device 8 at the judgment of the inspector.

  Next, a third embodiment of the ultrasonic inspection data evaluation apparatus according to the present invention will be described with reference to FIG. In the first embodiment, one-dimensional scanning ultrasonic flaw detection is targeted, but in the third embodiment, two-dimensional scanning is targeted. In the third embodiment, in addition to the configuration of the first embodiment, a crack confirmation device 9 that determines the presence or absence of a crack using a crack determination result from ultrasonic inspection data at adjacent scanning positions. It has.

  In the ultrasonic inspection, first, scanning in the x direction is performed, and then scanning in the x direction is performed by moving in the y direction. For scanning in the x direction, the presence or absence of a crack is automatically determined by the same method as in the first embodiment. Let posx = x1 be the crack position from the ultrasonic inspection data obtained by scanning in the y direction with posy = y1. Next, the position of the crack is detected in the same manner from ultrasonic inspection data obtained by scanning in the x direction at positions ysy, y2, y3, y4.

  As a result, it is assumed that cracks are detected at posy = 2, 3, 4, 8, 13, 14 as shown in FIG. 7, and two cracks are detected at posy = 13. It is considered that the crack has a three-dimensional shape and is detected at continuous scanning positions. The crack confirmation device 9 checks whether or not the result obtained by the crack measurement device is correct by checking whether a crack is continuously detected at the adjacent scanning position in the y direction.

  The data detected by posy = 2, 3, 4 is an ultrasonic signal from one crack, and it can be considered that the crack is accurately detected. The data detected at posy = 8 is determined to be a false detection because no crack is detected on both sides of posy = 7 and posy = 9. For posy = 13, a crack was detected at posx = 1,7, but at the adjacent pos14, a crack was detected only at posx = 1, and the signal from the crack is considered to be posx = 1 for posy13 The data of posx = 7 at posy = 13 can be considered as erroneous determination. Thus, the ultrasonic inspection data scanned two-dimensionally can be used, and the presence or absence of a crack can be detected with high accuracy.

The figure which shows the structure of the ultrasonic inspection data evaluation apparatus of 1st Embodiment. The figure which shows the example of an ultrasonic signal. The figure which shows the locus | trajectory of the calculated | required maximum value. The flowchart which shows the procedure of ultrasonic inspection data evaluation. The figure which shows the structure of the ultrasonic inspection data evaluation apparatus of 2nd Embodiment. The figure which shows the structure of the ultrasonic inspection data evaluation apparatus of 3rd Embodiment. The figure which shows the example of the crack position detected by secondary scanning. The figure which shows an ultrasonic inspection typically.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Ultrasonic inspection data, 2 ... Feature-value calculating apparatus, 3 ... Crack feature-value database, 4 ... Crack determination apparatus, 5 ... Crack measuring device, 6 ... Result display and preservation | save apparatus, 7 ... Crack manual determination Device, 8 ... Feature value data registration device, 9 ... Crack confirmation device.

Claims (6)

Ultrasound is obtained by transmitting ultrasonic waves from a plurality of inspection positions to the object to be inspected, and receiving reflected ultrasonic signals for each inspection position , and extracting the maximum value for each ultrasonic inspection data. , Feature amount calculation means for detecting a continuity of the maximum value and calculating a trajectory corresponding to the inspection position for the maximum value ,
A crack feature database containing the maximum feature value of the reflected ultrasonic signal from the crack;
The presence / absence of a crack and the end of the crack are compared and evaluated by comparing the data relating to the maximum value in the ultrasonic inspection data with the feature value of the maximum value of the reflected ultrasonic signal from the crack stored in the crack feature value database. Crack determining means for determining the position of the part ,
The ultrasonic inspection data evaluation apparatus , wherein the crack determination means determines that there is a crack when the locus is evaluated to have two maximum values at different levels . In the ultrasonic examination data evaluation device according to claim 1,
The ultrasonic inspection data evaluation apparatus according to claim 1, wherein the crack determination means determines that there is a crack when a maximum value having a high level among the two maximum values of the trajectory is a predetermined level or more . In the ultrasonic examination data evaluation device according to claim 1 or 2,
Ultrasonic inspection data evaluation characterized by comprising crack measuring means for measuring the depth of the crack based on an inspection position corresponding to a maximum value having a low level among the two maximum values of the trajectory. apparatus. In the ultrasonic inspection data evaluation device according to any one of claims 1 to 3,
The crack determination means is used when an inspection position where a low maximum value exists among the two maximum values of the trajectory is within a predetermined number of inspection positions from the end of the trajectory. An ultrasonic inspection data evaluation apparatus characterized by determining that there is a crack . In the ultrasonic examination data evaluation device according to any one of claims 1 to 4,
When the object is scanned in two dimensions, the crack detection information of the first position scanned in one dimension at the first position and the adjacent position scanned in one dimension at the position adjacent to the first position If it is determined from the crack detection information that there is a crack at the first position and it is not determined that there is a crack adjacent to the first position, there is a crack at the first position. An ultrasonic inspection data evaluation apparatus comprising a crack confirmation means that makes an erroneous determination . From the ultrasonic inspection data obtained by transmitting ultrasonic waves to the object to be inspected from a plurality of inspection positions and receiving the reflected ultrasonic signal for each inspection position , extract the maximum value for each ultrasonic inspection data, Detecting the continuity of the maximum value, calculating a trajectory corresponding to the inspection position for the maximum value, and making it a feature amount ;
The presence / absence of a crack and the end of the crack are compared and evaluated by comparing the data relating to the maximum value in the ultrasonic inspection data with the feature value of the maximum value of the reflected ultrasonic signal from the crack stored in the crack feature value database. A crack determination step for determining the position ,
An ultrasonic inspection data evaluation method characterized in that, in the crack determination step, it is determined that there is a crack when it is evaluated that the locus has two maximum values at different levels .

Images