Ultrasonic wave flaw detection device and ultrasonic wave flaw detection method

Abstract

To estimate a signal when a failed element is a normal element.SOLUTION: An ultrasonic wave flaw detection device detects defects inside an object using ultrasonic waves. The ultrasonic wave flaw detection device includes: a plurality of ultrasonic transmission elements that transmit ultrasonic waves toward the object; a plurality of ultrasonic reception elements that receive intensities of the ultrasonic waves reflected by the object as signals; and a data interpolation section that estimates a signal that is not actually received on the basis of the signals received by the ultrasonic reception elements. The data interpolation section acquires a neural network, which uses coordinates as input to output a corresponding signal, as a trained neural network, by training using coordinates of the plurality of ultrasonic reception elements and the signals received by the plurality of ultrasonic reception elements, and uses the trained neural network when there is a specific element confirmed to be faulty among the ultrasonic transmission elements or the ultrasonic reception elements to estimate a signal that would be received if the specific element has not failed.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to an ultrasonic flaw detection device and an ultrasonic flaw detection method. [Background technology]

[0002] In many industrial fields, there is a growing trend to use classifiers trained by machine learning methods to extract experience and knowledge from huge amounts of data and use this knowledge to automate processes. In particular, in ultrasonic flaw detection, which is used as a form of non-destructive testing in the manufacturing industry, many machine learning models based on neural networks, including deep learning, have been introduced, and efforts are underway to improve the efficiency of inspections that previously relied on human labor.

[0003] In addition to efforts to detect anomalies by applying deep learning to images of signals obtained by ultrasonic flaw detection, in recent years deep learning has also begun to be introduced to the signals themselves before they are imaged. For example, Patent Document 1 proposes a method for identifying detection targets such as defects using a machine learning model trained on 3D voxel data from a non-destructive testing device.

[0004] In parallel with the interest in machine learning, the number of elements used in ultrasonic flaw detection equipment is increasing. Instead of a single-element single probe, linear arrays in which elements are arranged in one dimension and matrix arrays in which elements are arranged in two dimensions have been developed, and the number of embedded elements is also steadily increasing. This is rapidly increasing the risk of malfunctions such as element degradation and failure.

[0005] Because ultrasound is generated by the physical vibration of the element caused by applying a high-frequency voltage to the piezoelectric element material, there are cases where the element physically deteriorates over many years of use or due to unexpected causes, causing the high-frequency ultrasonic signal to weaken, cases where tailing occurs due to deterioration of the damper that controls the element, cases where the ultrasonic signal does not generate at all due to element failure, etc. Cases where tailing occurs are when the signal continues to be output indefinitely, the signal shifts to the low-frequency side, and the time resolution deteriorates.

[0006] Furthermore, Patent Document 2 proposes a deep learning model in which an ultrasonic signal is used as input data and an appropriate corresponding filter control signal is used as output data. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Japanese Patent Application Publication No. 2023-30258 [Patent Document 2] Patent Publication No. 2021-159511 Summary of the Invention [Problem to be solved by the invention]

[0008] To mitigate the effects of a failed element, it is possible to check the flaw detection sensitivity of a standard test piece with each individual probe and adjust the gain to compensate for the loss of sensitivity due to deterioration, etc. However, there is a problem in that increasing the gain may result in an increase in noise, and it is not possible to correct for deterioration in all frequency bands.

[0009] Considering the recent trend of increasing numbers of elements, it is quite possible that the 3D voxel data will be degraded or contain missing values, which could prevent accurate flaw detection with the technology described in Patent Document 1. Furthermore, when using linear arrays or matrix arrays with a large number of elements, there is always the possibility that some elements will deteriorate or break down, which poses a problem of requiring a great deal of time to repair or remeasure the elements.

[0010] Furthermore, the deep learning model described in Patent Document 2 is trained on the assumption that the elements are normal, and therefore has the problem that if a faulty element is included, it may not be able to output an accurate filter control signal.

[0011] The present disclosure has been made in consideration of the above circumstances, and provides an ultrasonic flaw detection device and an ultrasonic flaw detection method that can estimate a signal that would be generated if a faulty element were a normal element. [Means for solving the problem]

[0012] In order to solve the above problem, according to one aspect of the present disclosure, there is provided an ultrasonic flaw detection device that uses ultrasonic waves to detect internal defects of an object, the ultrasonic flaw detection device comprising: a plurality of ultrasonic transmitting elements that transmit ultrasonic waves toward the object; a plurality of ultrasonic receiving elements that receive signals representing the intensity of the ultrasonic waves reflected by the object; and a data interpolation unit that estimates signals that are not actually received based on the signals received by the ultrasonic receiving elements, wherein the data interpolation unit acquires a trained neural network that takes coordinates as input and corresponding signals as output by learning using the coordinates of the plurality of ultrasonic receiving elements and the signals received by the plurality of ultrasonic receiving elements, and when there is a specific element among the ultrasonic transmitting elements or ultrasonic receiving elements that is known to be faulty, the ultrasonic flaw detection device uses the trained neural network to estimate the signal that would have been received if the specific element had not been faulty.

[0013] In order to solve the above problem, according to one aspect of the present disclosure, there is provided an ultrasonic flaw detection method for detecting internal defects of an object using ultrasonic waves, the ultrasonic flaw detection method comprising: an ultrasonic flaw detection device having a plurality of ultrasonic transmitting elements that transmit ultrasonic waves toward the object, a plurality of ultrasonic receiving elements that receive signals representing the intensity of the ultrasonic waves reflected by the object as signals, and a data interpolation unit that estimates signals that are not actually received based on the signals received by the ultrasonic receiving elements; an acquisition step using the data interpolation unit to learn a neural network that inputs coordinates and outputs corresponding signals using the coordinates of the plurality of ultrasonic receiving elements and the signals received by the plurality of ultrasonic receiving elements, thereby acquiring it as a trained neural network; and an estimation step, when a specific element that is known to be faulty among the ultrasonic transmitting elements or ultrasonic receiving elements is present, using the trained neural network to estimate a signal that would have been received if the specific element had not failed. [Effects of the Invention]

[0014] According to the present disclosure, it is possible to estimate the signal that would have occurred if the failed element had been a normal element. [Brief explanation of the drawings]

[0015] [Figure 1] FIG. 2 is a block diagram showing the hardware configuration of the ultrasonic flaw detection device. [Figure 2] FIG. 2 is a plan view of an ultrasonic transmitting element array. [Figure 3] FIG. 2 is a plan view of an ultrasonic receiving element array. [Figure 4] FIG. 2 is a functional block diagram of the ultrasonic flaw detector. [Figure 5] FIG. 10 is a plan view showing an example of a specific position. [Figure 6] 10 is a flowchart of an ultrasonic flaw detection process. DETAILED DESCRIPTION OF THE INVENTION

[0016] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

[0017] 1 shows the hardware configuration of an ultrasonic flaw detector 10 according to the present disclosure. The ultrasonic flaw detector 10 is a device that detects defects inside an object using ultrasonic waves.

[0018] The target is an object to be inspected when the ultrasonic flaw detector 10 inspects the target for defects, and may be, for example, a strip-shaped steel plate such as a rolled steel plate or a slab in a steel mill, etc., but is not limited to this. The target may be made of other materials than iron, such as aluminum or titanium, and may have various shapes other than a flat strip-like shape, such as a columnar billet, a steel bar, a wire rod, a tubular steel pipe, or an H-shaped cross-section steel.

[0019] Furthermore, defects include, but are not limited to, scratches, inclusions, bubbles, and the like inside the object.

[0020] 1, the ultrasonic flaw detector 10 includes a controller 12. The controller 12 is configured as a device including a general computer.

[0021] The controller 12 includes a central processing unit (CPU) 12A, a read-only memory (ROM) 12B, a random access memory (RAM) 12C, and an input / output interface (I / O) 12D. The CPU 12A, ROM 12B, RAM 12C, and I / O 12D are connected to each other via a bus 12E. The bus 12E includes a control bus, an address bus, and a data bus.

[0022] Furthermore, the ultrasonic transmitting element array 14, the ultrasonic receiving element array 16, the operation unit 18, the display unit 20, the communication unit 22, and the storage unit 24 are connected to the I / O 12D.

[0023] FIG. 2 is a plan view of the ultrasonic transmitter array 14. As shown in FIG. 2, the ultrasonic transmitter array 14 has a configuration in which a plurality of ultrasonic transmitter elements 30 are arranged in M ​​rows and N columns. Note that M≧1 and N≧2. In the example of FIG. 2, M=5, N=7, and the ultrasonic receiver elements 32 are arranged two-dimensionally, but it is also possible to have a configuration in which M=1 and the ultrasonic receiver elements 32 are arranged one-dimensionally. It is also possible to have a configuration in which a single ultrasonic receiver element 32 scans one-dimensionally or two-dimensionally.

[0024] Fig. 3 is a plan view of the ultrasonic receiver array 16. As shown in Fig. 3, the ultrasonic receiver array 16 has a configuration in which a plurality of ultrasonic receiver elements 32 are arranged in m rows x n. Note that m ≥ 1 and n ≥ 2. In the example of Fig. 3, M = 5 and N = 7, and the ultrasonic receiver elements 32 are arranged two-dimensionally, but it is also possible to have a configuration in which m = 1 and the ultrasonic receiver elements 32 are arranged one-dimensionally. It is also possible to have a configuration in which a single ultrasonic receiver element 32 scans one-dimensionally or two-dimensionally.

[0025] 1, each ultrasonic transmitting element 30 of the ultrasonic transmitting element array 14 transmits an ultrasonic wave US toward an object X. Each ultrasonic receiving element 32 of the ultrasonic receiving array 16 receives the intensity of the ultrasonic wave reflected by the object X as a signal RV.

[0026] The operation unit 18 includes, for example, a mouse and a keyboard, and is a mechanism for operating the ultrasonic flaw detector 10 and performing input according to the operator's intentions.

[0027] The display unit 20 is configured by, for example, a liquid crystal display or the like, and is a mechanism for conveying the progress and final results of the processing performed by the ultrasonic flaw detector 10 to the operator by displaying them or the like.

[0028] The communication unit 22 is an interface for performing data communication with an external device such as an external server.

[0029] The storage unit 24 is configured with a non-volatile external storage device such as a hard disk, etc. As shown in Fig. 1, the storage unit 24 stores an ultrasonic flaw detection program 26, a trained neural network 27, and specific element information 28.

[0030] 4 is a block diagram showing the functional configuration of the CPU 12A of the ultrasonic flaw detector 10. As shown in FIG.

[0031] The CPU 12A reads and executes the ultrasonic flaw detection program 26 stored in the storage unit 24, thereby functioning as each of the functional units of a data interpolation unit 34 and a defect detection unit 36.

[0032] The data interpolation unit 34 is a functional unit that estimates signals that are not actually received based on signals received by the ultrasonic receiving elements 32. More specifically, the data interpolation unit 34 acquires a trained neural network 27 by training a neural network, which inputs coordinates and outputs corresponding signals, using the coordinates of the multiple ultrasonic receiving elements 32 and the signals received by the multiple ultrasonic receiving elements 32, and when there is a specific element among the ultrasonic transmitting elements 30 or the ultrasonic receiving elements 32 that is known to be faulty, the trained neural network 27 is used to estimate a signal that would have been received if the specific element had not failed.

[0033] Here, a failure of the ultrasonic transmitting element 30 refers to a state in which ultrasonic waves US are not output from the ultrasonic transmitting element 30, or a state in which ultrasonic waves US are output but not at the desired intensity. Also, a failure of the ultrasonic receiving element 32 refers to a state in which the ultrasonic receiving element 32 cannot receive normal reflected waves of ultrasonic US, or a state in which normal reflected waves of ultrasonic US can be received but not at the desired intensity.

[0034] The trained neural network 27 may be acquired by training a neural network that inputs time in addition to coordinates and outputs corresponding signals using the coordinates and time of the multiple ultrasonic receiving elements 32 and the signals received by the multiple ultrasonic receiving elements 32. In this case, if there is a specific element among the multiple ultrasonic receiving elements 32 that is known to be faulty, the coordinates and time of the specific element can be input into the trained neural network 27, thereby estimating the signal that would have been received by the specific element at that time if the specific element had not failed. Furthermore, if there is a specific element among the ultrasonic transmitting elements 30 that is known to be faulty, the coordinates and time of the ultrasonic receiving element 32 of interest can be input into the trained neural network 27, thereby estimating the signal that would have been received by the ultrasonic receiving element 32 of interest at that time if the specific element had not failed.

[0035] The trained neural network used here is called a neural field (NF). A neural field is a neural network that takes spatial coordinates and time as input and outputs a desired signal. The desired signal can be a scalar or a vector. Furthermore, the time of the spatial coordinates and time input to the neural field can be omitted.

[0036] Neural networks are generally used for image classification and object detection, while neural fields are used for applications such as estimating signals and states that arise from any physical phenomenon, or generating signals that complement missing data signals when generating an image.

[0037] When training a neural network, it is desirable to preferentially include signals transmitted and received from elements in the vicinity of a specific element in the training data. This is because signals from elements physically close to the specific element are likely to be similar to the signal of the specific element, and may be able to be estimated more accurately and easily than when the signal of the specific element is estimated using a trained neural network trained using signals from elements far from the specific element.

[0038] In this way, by inputting the coordinates of a specific element, which is an element that is known to be faulty, into the trained neural network 27, it is possible to estimate the signal that would have been received if the specific element had not failed. This makes it possible to generate high-resolution ultrasonic images and detect defects with high accuracy.

[0039] For example, as shown in FIG. 5, when there is a specific element 38 among the multiple ultrasonic receiving elements 32 that is known to be faulty, the data interpolation unit 34 can input the coordinates of the specific element 38 to the trained neural network 27, thereby estimating the signal that the specific element 38 would have received if the specific element 38 had not been faulty.

[0040] Furthermore, when there is a specific element among the multiple ultrasonic transmitting elements 30 that is known to be faulty, the data interpolation unit 34 may input the coordinates of the ultrasonic receiving element 32 of interest to the trained neural network 27, thereby estimating the signal that would have been received by the ultrasonic receiving element 32 of interest if the specific element had not failed.

[0041] Here, the ultrasonic receiving element 32 of interest is an ultrasonic receiving element 32 that may be adversely affected, such as being unable to receive the signal RV or having a reduced strength of the signal RV, due to reasons such as the ultrasonic US not being output or the output being reduced from the specific element when there is a specific element that is known to be faulty among the multiple ultrasonic transmitting elements 30. For example, when one ultrasonic transmitting element 30 among the multiple ultrasonic transmitting elements 30 is faulty, the effect of the failure often extends to the signals of all the ultrasonic receiving elements 32, and therefore the ultrasonic receiving element 32 of interest refers to one or more ultrasonic receiving elements that can be selected from all the ultrasonic receiving elements 32.

[0042] Specific element information 28 relating to specific elements is stored in advance in the storage unit 24. The specific element information 28 is updated every time a failure of an ultrasonic transmitting element 30 or an ultrasonic receiving element 32 is discovered. If a faulty ultrasonic receiving element 32 is present, the specific element information 28 includes information on the coordinates of the ultrasonic receiving element 32.

[0043] The data interpolation unit 34 refers to the specific element information 28 to identify a faulty ultrasonic transmitting element 30 or ultrasonic receiving element 32, and when a faulty ultrasonic transmitting element 30 or ultrasonic receiving element 32 is present, it can estimate a normal signal that would have been received if the element had not failed, eliminating the influence of the faulty element. Therefore, compared to a case where only signals that can be received by the ultrasonic receiving elements 32 (including the specific element) are used, it is possible to make a correct judgment with high accuracy without making a mistake in judgment about defects due to erroneous information from the faulty specific element. Furthermore, even in cases where an ultrasonic image is generated to determine internal defects of an object, as will be described later, it is possible to generate a high-resolution ultrasonic image compared to a case where an ultrasonic image is generated only from signals received by the ultrasonic receiving elements 32.

[0044] Note that if the maximum amplitude of the signal RV received by the ultrasonic receiving element 32 is equal to or less than a predetermined threshold, the ultrasonic receiving element 32 may be deemed to be faulty. Alternatively, the strength of the signal RV that can be transmitted by the ultrasonic transmitting element 30 may be measured individually, and if the maximum amplitude of the signal RV is equal to or less than a predetermined threshold, the ultrasonic transmitting element 30 may be deemed to be faulty. Furthermore, a minimum amplitude threshold may be set, and a process for determining that tailing exists when the minimum amplitude threshold is exceeded may be performed independently, or the determination of the minimum and maximum amplitudes may be combined. In this case, an ultrasonic receiving element 32 whose maximum amplitude of the signal RV does not satisfy the predetermined threshold condition is set as a specific element, and its coordinate information is stored in the storage unit 24 together with the specific element information 28.

[0045] The defect detection unit 36 ​​is a functional unit that detects defects inside the object X. Specifically, the defect detection unit 36 ​​detects defects inside the object X based on signals received by the ultrasonic receiving elements 32, or signals obtained by elements of the ultrasonic receiving elements 32 other than the specific element 38, and signals estimated by the data interpolation unit 34.

[0046] For example, when generating an ultrasound image using FMC / TFM (Full Matrix Capture and Total Focusing Method), which is one of the aperture synthesis methods, ultrasound US is transmitted from one ultrasound transmitting element 30 to the object X, and the intensities of the ultrasound reflected by the object X are received as signals RV by all ultrasound receiving elements 32. This process is performed the same number of times as the number of ultrasound transmitting elements 30. As a result, the number of signals RV obtained is the same as the number of ultrasound transmitting elements 30 multiplied by the number of ultrasound receiving elements 32. The defect detection unit 36 ​​then generates a three-dimensional ultrasound image of the object X based on the obtained signals RV obtained the same as the number of ultrasound transmitting elements 30 multiplied by the number of ultrasound receiving elements 32. At this time, if there are no faulty elements among the ultrasound transmitting elements 30 and the ultrasound receiving elements 32, the defect detection unit 36 ​​can generate a highly accurate ultrasound image based on normal reception signals received by the ultrasound receiving elements 32. On the other hand, even if either the ultrasonic transmitting element 30 or the ultrasonic receiving element 32 contains a specific element that is a faulty element, by using the ultrasonic flaw detection device 10 of this embodiment, it is possible to estimate the signal that would have been received by the target receiving signal element 32 if the specific element had not failed, and the defect detection unit 36 ​​can use the estimated signal instead of the signal from the specific element to generate a highly accurate ultrasonic image as if the specific element had returned to being a normal element.

[0047] That is, the defect detection unit 36 ​​generates an ultrasonic image of the object X based on signals received by elements other than the specified element 38, out of the number of signals equal to the number of ultrasonic transmitting elements 30 multiplied by the number of ultrasonic receiving elements 32, and signals estimated by the data interpolation unit 34, i.e., signals that would have been received if the specified element 38 had not failed, obtained by inputting the coordinates of the specified element 38 into the trained neural network 27. If an ultrasonic image is generated using the signal from a failed element as is, the generated pixel values ​​may differ from the actual values, making it appear as if a defect exists in a location where there is no defect. Furthermore, the signal from the failed element may reduce the signal-to-noise ratio of pixels that should capture the defect. By using the ultrasonic tip damage detection device 10 according to this embodiment, it is possible to restore the signal that would have been received if the specified element 38 had not failed, thereby enabling the defect to be imaged at its precise location with a higher signal-to-noise ratio.

[0048] The defect detection unit 36 ​​may perform known image processing such as binarization and noise removal on the generated ultrasonic image of the object X to detect the position, shape, etc. of the defect. Alternatively, a deep learning model different from the neural field used to estimate the ultrasonic signal may be separately prepared to determine the position and type of the defect from the ultrasonic image. Alternatively, the defect detection unit 36 ​​may output the generated ultrasonic image of the object X to the display unit 20 for display, allowing an operator to visually detect the defect.

[0049] Next, the processing of the ultrasonic flaw detection method executed by the CPU 12A of the ultrasonic flaw detection device 10 will be described. Fig. 6 shows a flowchart of the defect detection method according to the present disclosure. The processing of the ultrasonic flaw detection method shown in Fig. 6 is executed by the CPU 12 reading the ultrasonic flaw detection program stored in the storage unit 24 into the RAM 12C and executing it. Below, a case where an ultrasonic image is generated by the aperture synthesis method will be described.

[0050] In step S100, the CPU 12A causes one of the ultrasonic transmitting elements 30 to transmit an ultrasonic wave US.

[0051] In step S101, the CPU 12A acquires the signals RV received by all the ultrasonic receiving elements 32.

[0052] In step S102, the CPU 12A determines whether or not ultrasonic waves US have been transmitted from all of the ultrasonic transmitting elements 30. If ultrasonic waves US have not been transmitted from all of the ultrasonic transmitting elements 30 (N (No) in step S102), the process returns to step S100, and ultrasonic waves US are transmitted from the ultrasonic transmitting elements 30 that have not yet transmitted ultrasonic waves US. On the other hand, if ultrasonic waves US have been transmitted from all of the ultrasonic transmitting elements 30 (Y (Yes) in step S102), the process proceeds to step S103. In this way, the ultrasonic transmitting elements 30 are sequentially switched to transmit ultrasonic waves US, and the process of receiving signals RV from all of the ultrasonic receiving elements 32 is executed for the number of ultrasonic transmitting elements 30.

[0053] In step S103, the CPU 12A acquires the trained neural network 27 stored in the storage unit 24 by reading it into the RAM 12C.

[0054] In step S104, the CPU 12A determines whether any ultrasonic receiving element 32 is faulty. Specifically, it determines whether any ultrasonic receiving element 32 has a maximum amplitude of the signal RV of the ultrasonic receiving element 32 received in step S101 that is equal to or less than a predetermined threshold. Note that a process for determining the presence or absence of trailing by setting a minimum amplitude threshold may be provided independently, or these processes may be combined. If any ultrasonic receiving element 32 has a maximum amplitude of the signal RV that is equal to or less than a predetermined threshold (Y (Yes) in step S104), the process proceeds to step S105. On the other hand, if no ultrasonic receiving element 32 has a maximum amplitude of the signal RV that is equal to or less than a predetermined threshold (N (No) in step S104), the process proceeds to step S107. That is, if the maximum amplitude of the signal RV is greater than the threshold, it is determined that there is no fault, and if the maximum amplitude of the signal RV is equal to or less than the threshold, it is determined that there is a fault. Furthermore, if a minimum amplitude threshold is set, it is determined that there is a fault if the minimum amplitude is equal to or greater than the threshold.

[0055] In step S105, the CPU 12A sets the ultrasonic receiving elements 32 whose maximum amplitude of the signal RV is equal to or less than a predetermined threshold as specific elements, and stores the coordinate information in the storage unit 24 together with the specific element information .

[0056] In step S106, the CPU 12A refers to the specific element information 28, inputs the coordinates of the specific element 38 into the trained neural network 27, and executes a process for acquiring the signal output from the trained neural network 27 that would have been received if the specific element 38 had not failed, the process being repeated for each specific element 38.

[0057] In step S107, if there is no specific element, the CPU 12A generates an ultrasound image of the object X based on the signals RV acquired in the processing of step S101 for the number of ultrasonic transmitting elements 30 × the number of ultrasonic receiving elements 32. On the other hand, if there is a specific element, the CPU 12A generates an ultrasound image of the object X based on the signals RV received by elements other than the specific element 38 out of the signals RV acquired by executing the processing of step S101 for the number of ultrasonic transmitting elements 30 × the number of ultrasonic receiving elements 32, and on the signals acquired in step S106 that would have been received if each of the specific elements 38 had not failed.

[0058] In step S108, the CPU 12A performs known image processing, including binarization processing, on the ultrasonic image generated in step S105 to detect the position, shape, etc. of the defect.

[0059] In step S109, the CPU 12A outputs the detection results such as the position and shape of the defect detected in step S106 to the display unit 20 to display them.

[0060] As described above, in this embodiment, if an ultrasonic receiving element 32 is faulty, the coordinates of the ultrasonic receiving element 32 known to be faulty are input to the trained neural network 27. If an ultrasonic transmitting element 30 is faulty, the coordinates of the ultrasonic receiving element 32 of interest are input to the trained neural network 27. By estimating and restoring the signal that would have been received if a specific element 38 had not failed, an ultrasonic image can be generated. Therefore, defects can be imaged accurately and with a higher signal-to-noise ratio than when an ultrasonic image is generated by mixing signals from the faulty element 38 with signals received by the ultrasonic receiving element 32. Furthermore, when using a linear array or matrix array with a large number of elements for flaw detection, high-precision flaw detection is possible even if some elements are faulty. Therefore, ultrasonic flaw detection can be continued without interrupting operations, even in continuous production lines such as those in the steel industry. Furthermore, by avoiding the need to repair or replace elements, the cost of the ultrasonic flaw detection device 10 can be reduced.

[0061] Although the present embodiment has been described above, the present disclosure is not limited to the above-described embodiments, and various modifications and applications are possible within the scope of the gist of the present disclosure.

[0062] Furthermore, the configuration of the ultrasonic flaw detection device 10 described in the above embodiment (see Figure 1) is one example, and it goes without saying that unnecessary parts may be deleted or new parts may be added within the scope of the present disclosure.

[0063] Furthermore, in the present embodiment, a case has been described in which the technology of the present disclosure is applied to an ultrasonic flaw detector, but the technology of the present disclosure can also be applied to other non-destructive testing devices.

[0064] An ultrasonic flaw detection device according to aspect 1 of the present disclosure is an ultrasonic flaw detection device that uses ultrasonic waves to detect internal defects of an object, and includes a plurality of ultrasonic transmitting elements that transmit ultrasonic waves toward the object, a plurality of ultrasonic receiving elements that receive signals representing the intensity of the ultrasonic waves reflected by the object, and a data interpolation unit that estimates signals that are not actually received based on the signals received by the ultrasonic receiving elements. The data interpolation unit acquires a trained neural network by learning a neural network that takes coordinates as input and corresponding signals as output using the coordinates of the plurality of ultrasonic receiving elements and the signals received by the plurality of ultrasonic receiving elements. When a specific element that is known to be faulty is present among the ultrasonic transmitting elements or ultrasonic receiving elements, the trained neural network is used to estimate the signal that would have been received if the specific element had not failed.

[0065] In the ultrasonic flaw detection device of aspect 2 of the present disclosure, in the above-mentioned aspect 1, when there is a specific element among the ultrasonic receiving elements that is an element that is known to be faulty, the data interpolation unit inputs the coordinates of the specific element into the trained neural network, thereby estimating the signal that the specific element would have received if the specific element had not been faulty.

[0066] In the ultrasonic flaw detection device of aspect 3 of the present disclosure, in the above-mentioned aspect 1 or 2, when there is a specific element among the ultrasonic transmitting elements that is an element that is known to be faulty, the data interpolation unit inputs the coordinates of the ultrasonic receiving element of interest to the trained neural network, thereby estimating the signal that would have been received by the ultrasonic receiving element of interest if the specific element had not failed.

[0067] The ultrasonic flaw detection device according to aspect 4 of the present disclosure is, in any of aspects 1 to 3 above, a defect detection unit that detects defects inside the object, and the defect detection unit detects defects inside the object based on signals received by the ultrasonic receiving elements, or signals obtained by elements other than a specific element among the ultrasonic receiving elements, and signals estimated by the data interpolation unit.

[0068] An ultrasonic flaw detection method according to aspect 5 of the present disclosure is an ultrasonic flaw detection method for detecting internal defects of an object using ultrasonic waves, which uses an ultrasonic flaw detection device having a plurality of ultrasonic transmitting elements that transmit ultrasonic waves toward the object, a plurality of ultrasonic receiving elements that receive signals representing the intensity of the ultrasonic waves reflected by the object, and a data interpolation unit that estimates signals that are not actually received based on the signals received by the ultrasonic receiving elements, and includes an acquisition step in which, using the data interpolation unit, a neural network that inputs coordinates and outputs corresponding signals is trained using the coordinates of the plurality of ultrasonic receiving elements and the signals received by the plurality of ultrasonic receiving elements to acquire it as a trained neural network; and an estimation step in which, when a specific element that is known to be faulty among the ultrasonic transmitting elements or ultrasonic receiving elements is present, the trained neural network is used to estimate a signal that would have been received if the specific element had not failed.

[0069] In the ultrasonic flaw detection method according to aspect 6 of the present disclosure, in the above-mentioned aspect 5, when there is a specific element among the ultrasonic receiving elements that is an element that is known to be faulty, the data interpolation unit inputs the coordinates of the specific element into the trained neural network, thereby estimating the signal that the specific element would have received if the specific element had not been faulty.

[0070] In the ultrasonic flaw detection method according to aspect 7 of the present disclosure, in the above-mentioned aspect 5 or 6, when there is a specific element among the ultrasonic transmitting elements that is an element that is known to be faulty, the data interpolation unit inputs the coordinates of the ultrasonic receiving element of interest to the trained neural network, thereby estimating the signal that would have been received by the ultrasonic receiving element of interest if the specific element had not failed.

[0071] The ultrasonic flaw detection device according to aspect 8 of the present disclosure, in any of aspects 5 to 7 above, has a defect detection step of using a defect detection unit to detect defects inside the object, and the defect detection unit detects defects inside the object based on signals received by the ultrasonic receiving elements, or signals obtained by elements other than a specific element among the ultrasonic receiving elements, and signals estimated by the data interpolation unit. [Explanation of symbols]

[0072] 10 Ultrasonic flaw detection equipment 12 Controllers 14 Ultrasonic transmitting element array 16 ultrasonic receiving element array 18 Control section 20 Display section 22 Communications Department 24 Memory section 26 Ultrasonic Flaw Inspection Program 27 Trained Neural Networks 28 Specific element information 30 ultrasonic transmitting element 32 ultrasonic receiving element 34 Data Interpolation Unit 36 Defect detection section 38 Specific elements RV signal US ultrasound X Object

Claims

1. An ultrasonic flaw detector that uses ultrasonic waves to detect internal defects of an object, a plurality of ultrasonic transmitting elements that transmit ultrasonic waves toward the target; a plurality of ultrasonic receiving elements that receive signals representing the intensities of ultrasonic waves reflected by the object; a data interpolation unit that estimates signals that are not actually received based on signals received by the ultrasonic receiving elements; and The data interpolation unit A neural network that receives coordinates as input and outputs corresponding signals is learned using the coordinates of the plurality of ultrasonic receiving elements and the signals received by the plurality of ultrasonic receiving elements, thereby obtaining a learned neural network; An ultrasonic flaw detection device that, when there is a specific element among the ultrasonic transmitting elements or ultrasonic receiving elements that is known to be faulty, uses the trained neural network to estimate the signal that would have been received if the specific element had not failed.

2. When there is a specific element among the ultrasonic receiving elements that is an element that is known to be faulty, The data interpolation unit 2. The ultrasonic flaw detection device according to claim 1, wherein the coordinates of the specific element are input to the trained neural network to estimate the signal that would have been received by the specific element if the specific element had not failed.

3. When there is a specific element among the ultrasonic transmitting elements that is an element that is known to be faulty, The data interpolation unit 2. The ultrasonic flaw detection device of claim 1, wherein the coordinates of a target ultrasonic receiving element are input to the trained neural network to estimate the signal that would have been received by the target ultrasonic receiving element if the specific element had not failed.

4. a defect detection unit that detects defects inside the object, The defect detection unit An ultrasonic flaw detection device according to any one of claims 1 to 3, which detects internal defects of the object based on signals received by the ultrasonic receiving elements, or signals obtained by elements other than a specific element among the ultrasonic receiving elements, and signals estimated by the data interpolation unit.

5. An ultrasonic flaw detection method for detecting internal defects of an object using ultrasonic waves, a plurality of ultrasonic transmitting elements that transmit ultrasonic waves toward the target; a plurality of ultrasonic receiving elements that receive signals representing the intensities of ultrasonic waves reflected by the object; a data interpolation unit that estimates signals that are not actually received based on signals received by the ultrasonic receiving elements; Using an ultrasonic flaw detector having Using the data interpolation unit, An acquisition step of acquiring a trained neural network by learning a neural network that receives coordinates as input and outputs corresponding signals using the coordinates of the plurality of ultrasonic receiving elements and the signals received by the plurality of ultrasonic receiving elements; an estimation step of estimating, when a specific element that is an element that is known to be faulty among the ultrasonic transmitting elements or the ultrasonic receiving elements, a signal that would have been received if the specific element had not failed by using the trained neural network; The ultrasonic flaw detection method has the following features.

6. When there is a specific element among the ultrasonic receiving elements that is an element that is known to be faulty, The data interpolation unit 6. The ultrasonic flaw detection method according to claim 5, wherein the coordinates of the specific element are input to the trained neural network to estimate a signal that would have been received by the specific element if the specific element had not failed.

7. When there is a specific element among the ultrasonic transmitting elements that is an element that is known to be faulty, The data interpolation unit 6. The ultrasonic flaw detection method according to claim 5, wherein the coordinates of a target ultrasonic receiving element are input to the trained neural network, thereby estimating a signal that would have been received by the target ultrasonic receiving element if the specific element had not failed.

8. a defect detection step of detecting a defect inside the object using a defect detection unit, The defect detection unit The ultrasonic flaw detection method according to any one of claims 5 to 7, wherein an internal defect of the object is detected based on a signal received by the ultrasonic receiving element, or a signal obtained by an element other than a specific element among the ultrasonic receiving elements, and a signal estimated by the data interpolation unit.

Images