JP2024001224A - Defect estimation device, defect estimation method and program - Google Patents

Abstract

An object of the present invention is to estimate the three-dimensional position of internal defects in a fiber-reinforced plastic laminate. [Solution] A defect estimating device includes a feature receiving unit configured to receive an input of an image obtained by measuring feature data based on temperature changes occurring on the surface of a fiber-reinforced plastic laminate, and a feature data and internal defect The feature data received by the feature reception unit is input to the model that has learned the relationship with position information representing the three-dimensional position of the laminate, thereby estimating the three-dimensional position of the internal defect in the stack. A defect position estimator. The position information is information that divides each layer of the laminate into a predetermined number of meshes and indicates whether or not each mesh has an internal defect. [Selection diagram] Figure 5

Description

There is an application for exception to loss of novelty.

The present invention relates to a defect estimation device, a defect estimation method, and a program.

Carbon Fiber Reinforced Plastic (CFRP) is a composite material that uses resin as a base material and carbon fiber as a reinforcing material. Generally, CFRP is used by laminating prepregs whose strength and rigidity are reinforced in one direction. Hereinafter, CFRP having a laminate structure will be referred to as a "CFRP laminate."

In order to detect internal defects in CFRP laminates, non-destructive testing such as ultrasonic measurements (see Non-Patent Document 1) or radiographic techniques (see Non-Patent Document 2) has been conventionally performed. In addition, in order to solve the shortcomings of conventional non-destructive testing, research has been conducted, for example, on stress analysis using infrared thermography (see Non-Patent Document 3) and methods using machine learning models (see Non-Patent Document 4). There is.

Shuichi Mikami, Toshiyuki Oshima, Noboru Sugawara, Tomoyuki Yamazaki, "Quantitative evaluation of defect detection by ultrasonic flaw detection method by detailed analysis of echo waveforms", Proceedings of the Japan Society of Civil Engineers, vol. 501, pp. 103-112, 1994. Sultan, M., Worden, K., Pierce, S., Hickey, D., Staszewski,W., Dulieu-Barton, J., and Hodzic, A., "On impact damage detection and quantification for CFRP laminates", Mechanical Systems and Signal Processing, vol. 25, pp. 3135-3152, 2011. Sakagami, T., Mizokami, Y., Shiozawa, D., Fujimoto, T., Izumi, Y., Hanai, T., and Moriyama, A., "Verification of the repair effect for fatigue cracks in members of steel bridges based on thermoelastic stress measurement", Engineering Fracture Mechanics, vol. 183, pp. 1-12, 2017. Go Bei, Kyoichi Nishi, Tadashi Huang, Yumi Fujikawa, "Damage Identification of CFRP Laminated Materials Using Neural Networks and Experimental Data," Proceedings of the Japan Society of Mechanical Engineers, A edition, vol. 62, pp. 2338-2343, 1996.

However, the techniques described in Non-Patent Document 3 and Non-Patent Document 4 have a problem in that the positions of internal defects in the CFRP laminate cannot be specified in detail.

In view of the above technical problems, an object of the present invention is to estimate the three-dimensional position of an internal defect in a fiber-reinforced plastic laminate.

In order to solve the above problems, a defect estimating device according to one aspect of the present invention calculates the difference between the temperature distribution before a temperature change occurs on the surface of a fiber-reinforced plastic laminate and the temperature distribution after the temperature change occurs. The feature data reception unit is configured to receive input of feature data based on and a defect position estimating unit configured to estimate a three-dimensional position of an internal defect in the stack by inputting the following information. The position information is information that divides each layer of the laminate into a predetermined number of meshes and indicates whether or not each mesh has an internal defect.

According to one aspect of the present invention, it is possible to estimate the three-dimensional position of an internal defect in a fiber-reinforced plastic laminate.

FIG. 1 is a diagram showing an example of the overall configuration of a defect estimation system. 1 is a diagram showing an example of a hardware configuration of a computer. It is a diagram showing an example of the hardware configuration of a temperature measuring device. It is a figure showing an example of a CFRP laminate. 1 is a diagram illustrating an example of a functional configuration of a defect estimation system according to a first embodiment; FIG. It is a figure showing an example of a processing procedure of a learning method in a 1st embodiment. It is a figure showing an example of a processing procedure of an estimation method in a 1st embodiment. 7 is a diagram illustrating an example of a functional configuration of a defect estimation system in Modification 1. FIG. 7 is a diagram illustrating an example of a processing procedure of a learning method in Modification 1. FIG. 7 is a diagram illustrating an example of a processing procedure of an estimation method in Modification 1. FIG. It is a figure showing an example of functional composition of a defect estimation system in a 3rd embodiment. It is a figure showing a test result.

Embodiments of the present invention will be described below with reference to the accompanying drawings. Note that, in this specification and the drawings, components having substantially the same functional configuration are designated by the same reference numerals, thereby omitting redundant explanation.

[First embodiment]
Carbon fiber reinforced plastic (CFRP) has high specific modulus characteristics and is widely used in the space and aviation fields. Further, in the future, as the electrification of automobiles progresses, its use is expected to expand in the automobile field.

CFRP (CFRP laminate) having a laminate structure may have internal defects during use or manufacturing. Internal defects that occur during use include, for example, delamination, fiber breakage, and base material cracking. Internal defects that occur during manufacturing include, for example, foreign matter contamination.

In order to detect internal defects occurring in CFRP laminates, non-destructive inspections such as ultrasonic measurements (see Non-Patent Document 1) or radiographic techniques (see Non-Patent Document 2) have been conventionally performed. However, ultrasonic measurement requires a contact member such as water or oil, and the skill of the inspection engineer has a large effect on the accuracy of the inspection results. Furthermore, the radiographic method requires safety management such as setting of controlled areas and requires development time, resulting in large economic and time costs.

Therefore, stress analysis using infrared thermography (see Non-Patent Document 3) is attracting attention. Infrared thermography is a device that measures the distribution of infrared energy emitted from the surface of an object using an infrared sensor and converts it into temperature distribution. Non-Patent Document 3 discloses that defect detection using infrared thermography is useful for structures such as bridges. However, the technique disclosed in Non-Patent Document 3 cannot specify the details of the position of the internal defect.

Further, research is being conducted on a method using a machine learning model (see Non-Patent Document 4). In Non-Patent Document 4, a neural network for determining the position and amount of damage is created using first to third order natural frequencies as input data. However, in Non-Patent Document 4, the CFRP laminate is divided into ten parts in the longitudinal direction, and the position and amount of damage are determined in eight elements excluding the two elements at the tip. Therefore, in the technique disclosed in Non-Patent Document 4, the resolution of the output position information is low, and the details of the position of the internal defect cannot be specified.

The defect estimating device in this embodiment estimates the three-dimensional position of an internal defect in the CFRP laminate from an image obtained by measuring the surface temperature of the CFRP laminate using infrared thermography. For estimation, a convolutional neural network (CNN) is used that has learned the relationship between the surface temperature change distribution of the CFRP laminate and the three-dimensional position of an internal defect in the CFRP laminate.

According to the defect estimating device in this embodiment, it is possible to obtain an estimated value of positional information (three-dimensional data) of an internal defect that the CFRP laminate has from an image (two-dimensional data) representing the surface temperature of the CFRP laminate. . That is, it is possible to precisely estimate the three-dimensional position of an internal defect in the CFRP laminate.

<Overall configuration of defect estimation system>
First, the overall configuration of the defect estimation system in this embodiment will be described with reference to FIG. FIG. 1 is a block diagram showing an example of the overall configuration of a defect estimation system in this embodiment.

As shown in FIG. 1, a defect estimation system 100 in this embodiment includes a temperature measurement device 1 and a defect estimation device 2. The temperature measurement device 1 and the defect estimation device 2 are connected to enable data communication via a communication network N1 such as a LAN (Local Area Network) or the Internet.

The temperature measuring device 1 is an electronic device that measures the surface temperature of a CFRP laminate that is a learning target or an estimation target. The temperature measuring device 1 changes the surface temperature of the CFRP laminate and measures the surface temperature of the CFRP laminate using an infrared camera before and after the surface temperature changes. The temperature measuring device 1 also generates an image representing a change in surface temperature based on two images representing the distribution of surface temperature (hereinafter also referred to as "surface temperature images") measured before and after the surface temperature changes. (hereinafter also referred to as a "temperature change image").

The method of changing the surface temperature of the CFRP laminate differs depending on the shape and use of the CFRP laminate. In this embodiment, the surface temperature is changed by applying a fixed external force to the CFRP laminate. Specifically, both ends of a plate-shaped CFRP laminate are supported and pulled in opposite directions.

The method of changing the surface temperature of the CFRP laminate is not limited to the method of applying a fixed external force. For example, the surface temperature may be changed by applying heat to the CFRP laminate. In this case, what is necessary is to irradiate the surface of the CFRP laminate with infrared rays to raise the surface temperature, and then measure the temperature change over time.

The defect estimating device 2 is an information processing device such as a PC (Personal Computer), a workstation, or a server that estimates the position of an internal defect in the CFRP laminate. The defect estimating device 2 inputs feature data based on the temperature change image based on the temperature change image obtained by measuring the temperature change of the CFRP laminate to be studied and the position information of the internal defect input by the user. Learn an estimation model that outputs estimated values of position information of internal defects. Furthermore, the defect estimating device 2 uses the learned estimation model to estimate the position of the internal defect from a temperature change image obtained by measuring the temperature change of the CFRP laminate to be estimated.

Note that the overall configuration of the defect estimation system 100 shown in FIG. 1 is one example, and there may be various system configuration examples depending on the use and purpose. For example, the defect estimation device 2 may be realized by a plurality of computers, or may be realized as a cloud computing service.

<Hardware configuration of defect estimation system>
Next, the hardware configuration of the defect estimation system in this embodiment will be explained with reference to FIGS. 2 and 3.

≪Computer hardware configuration≫
The defect estimating device 2 in this embodiment is realized by, for example, a computer. FIG. 2 is a block diagram showing an example of the hardware configuration of the computer 500 in this embodiment.

As shown in FIG. 2, the computer 500 includes a CPU (Central Processing Unit) 501, a ROM (Read Only Memory) 502, a RAM (Random Access Memory) 503, an HDD (Hard Disk Drive) 504, an input device 505, It has a display device 506, a communication I/F (Interface) 507, and an external I/F 508. CPU501, ROM502, and RAM503 form what is called a computer. Each piece of hardware in the computer 500 is interconnected via a bus line 509. Note that the input device 505 and the display device 506 may be connected to an external I/F 508 for use.

The CPU 501 is an arithmetic unit that implements control and functions of the entire computer 500 by reading programs and data from a storage device such as the ROM 502 or the HDD 504 onto the RAM 503 and executing processing.

The ROM 502 is an example of a nonvolatile semiconductor memory (storage device) that can retain programs and data even when the power is turned off. The ROM 502 functions as a main storage device that stores various programs, data, etc. necessary for the CPU 501 to execute various programs installed on the HDD 504 . Specifically, the ROM 502 stores data such as boot programs such as BIOS (Basic Input/Output System) and EFI (Extensible Firmware Interface) that are executed when the computer 500 is started, OS (Operating System) settings, and network settings. is stored.

The RAM 503 is an example of a volatile semiconductor memory (storage device) whose programs and data are erased when the power is turned off. The RAM 503 is, for example, DRAM (Dynamic Random Access Memory) or SRAM (Static Random Access Memory). The RAM 503 provides a work area where various programs installed on the HDD 504 are expanded when the CPU 501 executes them.

The HDD 504 is an example of a nonvolatile storage device that stores programs and data. The programs and data stored in the HDD 504 include an OS, which is basic software that controls the entire computer 500, and applications that provide various functions on the OS. Note that, instead of the HDD 504, the computer 500 may use a storage device (for example, SSD: Solid State Drive) that uses flash memory as a storage medium.

The input device 505 is a touch panel used by the user to input various signals, operation keys or buttons, a keyboard or mouse, a microphone used to input sound data such as voice, or the like.

The display device 506 includes a display such as a liquid crystal or organic EL (Electro-Luminescence) that displays a screen, a speaker that outputs sound data such as audio, and the like.

The communication I/F 507 is an interface that connects to a communication network and allows the computer 500 to perform data communication.

External I/F 508 is an interface with an external device. The external device includes a drive device 510 and the like.

The drive device 510 is a device for setting the recording medium 511. The recording medium 511 here includes a medium that records information optically, electrically, or magnetically, such as a CD-ROM, a flexible disk, and a magneto-optical disk. Further, the recording medium 511 may include a semiconductor memory or the like that electrically records information, such as a ROM or a flash memory. Thereby, the computer 500 can read and/or write to the recording medium 511 via the external I/F 508.

Note that the various programs installed on the HDD 504 are, for example, when the distributed recording medium 511 is set in the drive device 510 connected to the external I/F 508, and the various programs recorded on the recording medium 511 are read by the drive device 510. It is installed by Alternatively, various programs to be installed on the HDD 504 may be installed by being downloaded from a network different from the communication network via the communication I/F 507.

≪Hardware configuration of temperature measurement device≫
FIG. 3 is a block diagram showing an example of the hardware configuration of the temperature measuring device 1 in this embodiment. As shown in FIG. 3, the temperature measuring device 1 in this embodiment includes a control device 101, a pair of support sections 102, a drive control section 103, and an infrared camera 104.

The support section 102 includes a first support section 102A that supports one end of the CFRP laminate 1000, and a second support section 102B that supports the other end. The drive control section 103 controls the first support section 102A and the second support section 102B to move linearly along the line connecting the first support section 102A and the second support section 102B.

With the first support section 102A and the second support section 102B supporting both ends of the CFRP laminate 1000, the drive control section 103 moves the first support section 102A and the second support section 102B in mutually opposite directions. , tensile stress can be applied to the CFRP laminate 1000. This causes a temperature drop in the CFRP laminate 1000 that is proportional to the stress fluctuation, and the surface temperature of the CFRP laminate 1000 changes.

The infrared camera 104 receives infrared rays emitted from the surface of the CFRP laminate 1000, and generates a surface temperature image of the CFRP laminate 1000 through infrared analysis.

Control device 101 is realized by, for example, a computer. The control device 101 includes a CPU 501, a ROM 502, a RAM 503, an input device 505, a display device 506, and an external I/F 508. CPU501, ROM502, and RAM503 form what is called a computer. Each piece of hardware in the control device 101 is interconnected via a bus line 509. Note that the input device 505 and the display device 506 may be connected to an external I/F 508 for use.

The drive control unit 103 and the infrared camera 104 are connected to an external I/F 508 and are controlled by the control device 101. Further, the surface temperature image acquired by the infrared camera 104 is input to the control device 101 via the external I/F 508.

Here, the structure of the CFRP laminate in this embodiment will be explained with reference to FIG. 4. FIG. 4 is a conceptual diagram showing an example of a CFRP laminate in this embodiment.

As shown in FIG. 4, the CFRP laminate 1000 in this embodiment is constructed by laminating a plurality of prepregs 1001. In each prepreg 1001, the strength and rigidity of a base material 1002, which is a resin, is reinforced in one direction by reinforcing fibers 1003, which are carbon fibers. The CFRP laminate 1000 is configured to exhibit isotropy as a whole by rotating the fiber orientation in each layer by an equal angle when stacking the prepregs 1001.

Although FIG. 4 shows an example in which each prepreg is rotated by 90 degrees, it may be rotated by 30 degrees, 45 degrees, 60 degrees, etc., for example. Further, although FIG. 4 shows an example in which ten layers are laminated, the number of layers is not limited as long as it is two or more layers. In this embodiment, a CFRP laminate is assumed, but the types of reinforcing fibers and resin may be changed as appropriate.

<Functional configuration of defect estimation system>
Next, the functional configuration of the defect estimation system in this embodiment will be described with reference to FIG. 5. FIG. 5 is a block diagram showing an example of the functional configuration of the defect estimation system in this embodiment.

≪Functional configuration of temperature measurement device≫
As shown in FIG. 5, the temperature measurement device 1 in this embodiment includes a temperature change section 11, a temperature measurement section 12, and an image generation section 13.

The temperature change section 11 is realized by the support section 102 and the drive control section 103 shown in FIG. The temperature measuring section 12 is realized by an infrared camera 104 shown in FIG. The image generation unit 13 is realized by processing that causes the CPU 501 to execute a program loaded from the ROM 502 onto the RAM 503 shown in FIG.

The temperature change unit 11 changes the surface temperature of the CFRP laminate, which is a learning target and an estimation target, by causing the drive control unit 103 to drive the support unit 102 .

The temperature measurement unit 12 uses an infrared camera 104 to measure the surface temperature of the CFRP laminate, which is a learning target and an estimation target. Furthermore, the temperature measurement unit 12 generates a surface temperature image of the CFRP laminate. Furthermore, the temperature measurement section 12 sends the generated surface temperature image to the image generation section 13.

The image generation unit 13 generates a temperature change image based on two surface temperature images generated before and after the temperature change unit 11 changes the surface temperature of the CFRP laminate. Further, the image generation unit 13 transmits the generated temperature change image to the defect estimation device 2.

≪Functional configuration of defect estimation device≫
As shown in FIG. 5, the defect estimating device 2 in this embodiment includes an image receiving section 21, a position information receiving section 22, a feature calculation section 23, a model learning section 24, a model storage section 25, a defect position estimation It includes a section 26 and a result output section 27.

The image reception unit 21, the position information reception unit 22, the feature amount calculation unit 23, the model learning unit 24, the defect position estimation unit 26, and the result output unit 27 execute the program developed on the RAM 503 from the HDD 504 shown in FIG. is realized by the processing executed by the CPU 501. The model storage unit 25 is realized by the HDD 504 shown in FIG.

The image reception unit 21 receives the temperature change image from the temperature measurement device 1 . Further, the image reception unit 21 sends the temperature change image to the feature amount calculation unit 23.

The position information receiving unit 22 receives input of position information of an internal defect that a CFRP laminate that is a learning target has in response to a user's operation. Further, the position information receiving unit 22 sends the received position information of the internal defect to the model learning unit 24 .

The feature calculation unit 23 calculates feature data based on the temperature change image received from the image reception unit 21. The characteristic data in this embodiment is the surface principal stress sum distribution of the CFRP laminate.

When the feature value calculation unit 23 calculates feature data based on the temperature change image regarding the CFRP laminate to be studied, it sends the feature data to the model learning unit 24 . Furthermore, when the feature value calculation unit 23 calculates feature data based on the temperature change image regarding the CFRP laminate to be estimated, it sends the feature data to the defect position estimation unit 26 .

The model learning unit 24 generates learning data by associating the internal defect position information received by the position information receiving unit 22 with the feature data calculated by the feature value calculation unit 23. Furthermore, the model learning unit 24 uses the generated learning data to learn the estimated model.

The model storage unit 25 stores the learned estimated model generated by the model learning unit 24.

The defect position estimating unit 26 reads out the estimated model from the model storage unit 25 and inputs the feature data received from the feature calculation unit 23 into the estimated model, thereby obtaining position information of internal defects in the CFRP laminate to be estimated. Compute an estimate of .

The result output unit 27 outputs the internal defect estimation results to the display device 506 or the like. The estimation result includes the estimated value of the position information of the internal defect calculated by the defect position estimation unit 26.

<Defect estimation system processing procedure>
Next, the processing procedure of the defect estimation method executed by the defect estimation system in this embodiment will be explained with reference to FIGS. 6 and 7. The defect estimation method in this embodiment includes a learning method for learning an estimation model (see FIG. 6), and an estimation method for estimating the three-dimensional position of an internal defect using the learned estimation model (see FIG. 7).

≪Learning method≫
FIG. 6 is a flowchart showing an example of the processing procedure of the learning method in this embodiment.

In step S12-1, the temperature measurement unit 12 included in the temperature measurement device 1 measures the surface temperature of the CFRP laminate to be studied, before changing the surface temperature of the CFRP laminate. Next, the temperature measuring unit 12 generates a surface temperature image (hereinafter also referred to as "image before temperature change") representing the distribution of the measured surface temperature. Subsequently, the temperature measurement section 12 sends the image before temperature change to the image generation section 13.

In step S11, the temperature change unit 11 included in the temperature measuring device 1 changes the temperature of the CFRP laminate to be studied. Specifically, a predetermined fixed external force is applied to the CFRP laminate to be studied.

In step S12-2, the temperature measuring unit 12 included in the temperature measuring device 1 changes the surface temperature of the CFRP laminate to be studied, and then measures the surface temperature of the laminate. Next, the temperature measuring unit 12 generates a surface temperature image (hereinafter also referred to as a "temperature change image") representing the distribution of the measured surface temperature. Subsequently, the temperature measurement unit 12 sends the image after the temperature change to the image generation unit 13.

In step S13, the image generation unit 13 included in the temperature measurement device 1 receives the image before temperature change and the image after temperature change from the temperature measurement unit 12. Next, the image generation unit 13 generates a temperature change image representing a change in surface temperature of the CFRP laminate to be studied, based on the image before temperature change and the image after temperature change. Subsequently, the image generation unit 13 transmits the temperature change image to the defect estimation device 2.

Specifically, the image generation unit 13 calculates the difference between the image before the temperature change and the image after the temperature change. That is, for each pixel of each image, the surface temperature represented by the image before temperature change is subtracted from the surface temperature represented by the image after temperature change.

In step S21, the image receiving unit 21 included in the defect estimating device 2 receives a temperature change image regarding the CFRP laminate to be studied from the temperature measuring device 1. Next, the image reception unit 21 sends the temperature change image received from the temperature measurement device 1 to the feature quantity calculation unit 23.

In step S22, the position information receiving unit 22 included in the defect estimating device 2 receives input of position information representing the three-dimensional position of an internal defect in the CFRP laminate to be studied, in response to a user's operation. Next, the position information receiving unit 22 sends the received position information of the internal defect to the model learning unit 24 .

The position information of the internal defect in this embodiment includes the three-dimensional coordinates where the internal defect exists in the CFRP laminate and the size of the internal defect. Note that the size of the internal defect is the range in which the internal defect exists in each layer of the CFRP laminate.

Specifically, the internal defect position information is information in which each layer of the CFRP laminate is divided into a predetermined number of meshes, and a channel is set in each mesh to indicate whether or not an internal defect exists. For example, the channel is set to 1 when an internal defect exists, and 0 when there is no internal defect. That is, the position of each mesh represents a three-dimensional coordinate obtained by adding the depth of the layer to the two-dimensional coordinate seen from the surface of the CFRP laminate. Furthermore, the range in which the channel value is 1 in each layer represents the size of the internal defect.

Note that in this embodiment, the mesh is divided into orthogonal equal intervals. However, if a position where defects are likely to occur is known, the area near the position may be finely divided, and the other areas may be coarsely divided.

In step S<b>23 , the feature calculation unit 23 included in the defect estimation device 2 receives the temperature change image from the image reception unit 21 . Next, the feature calculation unit 23 calculates the surface principal stress sum distribution based on the temperature change image. Subsequently, the feature calculation section 23 sends the calculated surface principal stress sum distribution to the model learning section 24.

Specifically, the feature calculation unit 23 divides the temperature change image into a predetermined number of meshes, and calculates the amount of temperature change corresponding to each mesh, thereby obtaining the surface temperature change distribution. The amount of temperature change in the mesh may be the amount of temperature change at a representative point (such as the center) of the mesh, or may be the average amount of temperature change within the mesh. The mesh division method is the same as the internal defect position information.

Next, the feature calculation unit 23 calculates the surface principal stress sum distribution by applying Kelvin's theory (see Reference 1) to the surface temperature change distribution. Kelvin's theory gives the relationship between the amount of change in temperature ΔT due to the thermoelastic effect and the amount of change Δσ in the sum of principal stresses as ΔT = -kTΔσ. However, k is the thermoelastic coefficient and T is the absolute temperature.

[Reference 1] W. Thomson (Lord Kelvin), "On the Dynamical Theory of Heat", Trans. Roy. Soc., Vol.20, pp.261-283, 1853.

Steps S12-1 to S23 are repeatedly executed for each CFRP laminate to be studied. Thereby, the surface principal stress distribution and internal defect position information corresponding to each CFRP laminate to be studied are input to the model learning unit 24.

In step S24, the model learning unit 24 included in the defect estimating device 2 receives the position information of the internal defect from the position information receiving unit 22. Furthermore, the model learning section 24 receives the surface principal stress distribution from the feature calculation section 23.

Next, the model learning unit 24 generates learning data by associating the surface principal stress distribution with the position information of the internal defect. At this time, the model learning unit 24 may standardize the surface principal stress sum distribution included in the learning data with respect to the entire learning data.

Subsequently, the model learning unit 24 uses the generated learning data to learn the estimated model. The estimation model in this embodiment is a convolutional neural network that inputs the surface principal stress sum distribution of a CFRP laminate and outputs an estimated value of positional information of an internal defect that the CFRP laminate has.

A convolutional neural network is a special type of neural network used to process data that has a grid-like structure. Examples of data having a grid-like structure include time series data in which samples acquired at equal time intervals are arranged in one dimension, or image data in which pixels are arranged in two dimensions.

The convolutional neural network in this embodiment is configured as follows. The loss function uses binary cross error. Adam is used as the optimization method. At this time, the learning rate is set to 0.00003. The mini-batch size is 16 and the number of trainings is 1000. Note that regularization methods such as L2 regularization or batch normalization may not be used.

In step S25, the model learning unit 24 included in the defect estimation device 2 stores the learned estimation model in the model storage unit 25.

≪Estimation method≫
FIG. 7 is a flowchart showing an example of the processing procedure of the estimation method in this embodiment.

In step S12-1, the temperature measurement unit 12 included in the temperature measurement device 1 measures the surface temperature of the CFRP laminate to be estimated, before changing the surface temperature of the CFRP laminate. Next, the temperature measuring unit 12 generates a pre-temperature change image representing the distribution of the measured surface temperature. Subsequently, the temperature measurement section 12 sends the image before temperature change to the image generation section 13.

In step S11, the temperature change unit 11 included in the temperature measuring device 1 changes the temperature of the CFRP laminate to be estimated. Specifically, a fixed external force similar to that in the learning process is applied to the CFRP laminate to be estimated.

In step S12-2, the temperature measurement unit 12 included in the temperature measurement device 1 changes the surface temperature of the CFRP laminate to be estimated, and then measures the surface temperature of the CFRP laminate. Next, the temperature measuring unit 12 generates a post-temperature-change image representing the distribution of the measured surface temperature. Subsequently, the temperature measurement unit 12 sends the image after the temperature change to the image generation unit 13.

In step S13, the image generation unit 13 included in the temperature measurement device 1 receives the image before temperature change and the image after temperature change from the temperature measurement unit 12. Next, the image generation unit 13 generates a temperature change image representing a surface temperature change of the CFRP laminate to be estimated, based on the image before temperature change and the image after temperature change. Subsequently, the image generation unit 13 transmits the temperature change image to the defect estimation device 2.

In step S21, the image receiving unit 21 included in the defect estimating device 2 receives a temperature change image regarding the CFRP laminate to be estimated from the temperature measuring device 1. Next, the image reception unit 21 sends the temperature change image received from the temperature measurement device 1 to the feature quantity calculation unit 23.

In step S<b>23 , the feature calculation unit 23 included in the defect estimation device 2 receives the temperature change image from the image reception unit 21 . Next, the feature calculation unit 23 calculates the surface principal stress sum distribution based on the temperature change image. Subsequently, the feature calculation section 23 sends the calculated surface principal stress sum distribution to the model learning section 24.

In step S26, the defect position estimation unit 26 included in the defect estimation device 2 receives the surface principal stress sum distribution from the feature value calculation unit 23. Next, the defect position estimation section 26 reads the estimated model from the model storage section 25. Subsequently, the defect position estimating unit 26 inputs the received surface principal stress sum distribution into the read estimation model to calculate an estimated value of the position information of the internal defect of the CFRP laminate to be estimated. Then, the defect position estimation section 26 sends the estimated value of the position information of the internal defect to the result output section 27 .

In step S27, the result output unit 27 included in the defect estimating device 2 receives the estimated value of the internal defect position information from the defect position estimating unit 26. Next, the result output unit 27 outputs the estimation result of the three-dimensional position of the internal defect to the display device 506 or the like. The estimation result includes the estimated value of the position information of the internal defect calculated by the defect position estimation unit 26.

<Effects of the first embodiment>
The defect estimation system in this embodiment receives as input the surface principal stress sum distribution of the CFRP laminate to be learned, and learns an estimation model that outputs an estimated value of position information representing the three-dimensional position of an internal defect. In addition, the defect estimation system in this embodiment inputs the surface principal stress sum distribution based on an image representing a change in surface temperature of a CFRP laminate to be estimated into a trained estimation model. The three-dimensional position of the internal defect is estimated. Therefore, according to the defect estimation system in this embodiment, it is possible to estimate the three-dimensional position of an internal defect that the CFRP laminate has.

Further, the position information in this embodiment includes the depth of the layer in which the internal defect exists in the CFRP laminate and the range in which the internal defect exists in the layer. Therefore, according to the defect estimation system in this embodiment, the three-dimensional coordinates and size of internal defects in the CFRP laminate can be precisely estimated.

[Second embodiment]
The defect estimating device 2 in the first embodiment learned an estimation model using the surface principal stress sum distribution as feature data, and estimated the three-dimensional position of the internal defect. The defect estimating device 2 in the second embodiment is configured to use surface temperature change distribution as feature data.

<Functional configuration of defect estimation system>
The feature calculation unit 23 in this embodiment calculates the surface temperature change distribution based on the temperature change image received from the image reception unit 21. The feature quantity calculation unit 23 in the first embodiment calculated the surface temperature change distribution based on the temperature change image, and calculated the surface principal stress sum distribution by applying Kelvin's theory to the surface temperature change distribution. The feature calculation unit 23 in this embodiment uses the surface temperature change distribution before applying Kelvin's theory as feature data.

The model learning unit 24 in this embodiment generates learning data by associating the internal defect position information received by the position information receiving unit 22 with the surface temperature change distribution calculated by the feature value calculation unit 23. In addition, the model learning unit 24 in this embodiment uses the generated learning data to create a convolution neural network that inputs the surface temperature change distribution of the CFRP laminate and outputs an estimated value of position information of an internal defect that the laminate has. Learn the network (estimation model).

The defect position estimation unit 26 in this embodiment reads the estimation model from the model storage unit 25 and inputs the surface temperature change distribution received from the feature value calculation unit 23 into the estimation model, thereby determining the CFRP laminate to be estimated. Calculate an estimate of the location information of the internal defect.

<Effects of the second embodiment>
The defect estimation system in this embodiment uses a surface temperature change distribution calculated based on a temperature change image to learn an estimation model and estimate the position of an internal defect. Thereby, there is no need to calculate the surface principal stress sum distribution from the surface temperature change distribution, so the processing speed of the feature value calculation unit 23 is improved.

[Modification 1]
In the first and second embodiments, the defect estimating device 2 calculates feature data (i.e., surface principal stress sum distribution or surface temperature change distribution) from the temperature change image generated by the temperature measurement device 1. Configured. In the first modification, the temperature measurement device 1 is configured to calculate characteristic data based on the image before the temperature change and the image after the temperature change and send it to the defect estimation device 2.

Hereinafter, this modified example will be explained focusing on the differences from the first embodiment. Note that the configuration of this modification can be similarly applied to the second embodiment. When applied to the second embodiment, the surface principal stress sum distribution in this modification may be read as the surface temperature change distribution.

<Functional configuration of defect estimation system>
First, the functional configuration of the defect estimation system in this modification will be described with reference to FIG. 8. FIG. 8 is a block diagram showing an example of the functional configuration of the defect estimation system in this modification.

≪Functional configuration of temperature measurement device≫
As shown in FIG. 8, the temperature measurement device 1 in this modification includes a temperature change section 11 and a temperature measurement section 12, and further includes a feature value calculation section 14, as in the first embodiment. Note that the temperature measuring device 1 in this modification does not include the image generation unit 13 that the temperature measuring device 1 in the first embodiment had.

The feature calculation unit 14 calculates the surface principal stress sum distribution based on two surface temperature images generated before and after the temperature change unit 11 changes the surface temperature of the CFRP laminate. Further, the feature calculation unit 14 transmits the calculated surface principal stress sum distribution to the defect estimation device 2.

≪Functional configuration of defect estimation device≫
As shown in FIG. 8, the defect estimating device 2 in this modified example includes a position information receiving unit 22, a model learning unit 24, a model storage unit 25, a defect position estimating unit 26, and A result output section 27 is provided, and a feature amount reception section 31 is provided in place of the image reception section 21. Note that the defect estimating device 2 in this modification does not include the feature value calculation unit 23 that the defect estimating device 2 in the first embodiment had.

The feature receiving unit 31 receives the surface principal stress sum distribution from the temperature measuring device 1 . When the feature receiving unit 31 receives the surface principal stress sum distribution regarding the CFRP laminate to be studied, it sends the surface principal stress sum distribution to the model learning unit 24 . Further, when the feature receiving unit 31 receives the surface principal stress sum distribution regarding the CFRP laminate to be estimated, it sends the surface principal stress sum distribution to the defect position estimating unit 26 .

<Defect estimation system processing procedure>
Next, the processing procedure of the defect estimation method executed by the defect estimation system in this modification will be described with reference to FIGS. 9 and 10.

≪Learning method≫
FIG. 9 is a flowchart showing an example of the processing procedure of the learning method in this modification.

In step S<b>14 , the feature calculation unit 14 included in the temperature measurement device 1 receives the image before temperature change and the image after temperature change from the temperature measurement unit 12 . Next, the feature calculation unit 14 calculates the surface principal stress sum distribution of the CFRP laminate to be studied, based on the image before the temperature change and the image after the temperature change. Then, the feature calculation unit 14 transmits the calculated surface principal stress sum distribution to the defect estimation device 2.

In step S31, the feature receiving unit 31 included in the defect estimating device 2 receives the surface principal stress sum distribution regarding the CFRP laminate to be learned from the temperature measuring device 1. Next, the feature receiving unit 31 sends the surface principal stress sum distribution received from the temperature measuring device 1 to the model learning unit 24.

≪Estimation method≫
FIG. 10 is a flowchart showing an example of the processing procedure of the estimation method in this modification.

In step S<b>14 , the feature calculation unit 14 included in the temperature measurement device 1 receives the image before temperature change and the image after temperature change from the temperature measurement unit 12 . Next, the feature calculation unit 14 calculates the surface principal stress sum distribution of the CFRP laminate to be estimated, based on the image before the temperature change and the image after the temperature change. Then, the feature calculation unit 14 transmits the calculated surface principal stress sum distribution to the defect estimation device 2.

In step S31, the feature receiving unit 31 included in the defect estimating device 2 receives the surface principal stress sum distribution regarding the CFRP laminate to be estimated from the temperature measuring device 1. Next, the feature receiving unit 31 sends the surface principal stress sum distribution received from the temperature measuring device 1 to the defect position estimating unit 26.

<Effects of modification 1>
In the defect estimation system in this embodiment, an image acquisition device calculates a surface principal stress sum distribution and transmits it to a defect estimation device. An image acquisition device that can be equipped with a program that calculates the surface principal stress sum distribution from infrared thermography has been realized. By combining with such an image acquisition device, the load on the defect estimation device to calculate feature data can be reduced.

[Third embodiment]
The defect estimating device 2 in the first embodiment generated learning data using a temperature change image obtained by actually measuring a change in surface temperature of a CFRP laminate as a learning target. The defect estimating device 2 in the third embodiment is configured to generate learning data through numerical analysis without measuring changes in surface temperature of the CFRP laminate.

<Functional configuration of defect estimation system>
The functional configuration of the defect estimation system in this embodiment will be described with reference to FIG. 11. FIG. 11 is a block diagram showing an example of the functional configuration of the defect estimation system in this embodiment.

As shown in FIG. 11, the defect estimation system according to the present embodiment differs from the defect estimation system according to the first embodiment in the following points. First, the defect estimating device 2 does not include the position information receiving section 22. Second, the defect estimation device 2 includes a learning data generation section 28 .

The learning data generation unit 28 calculates the surface principal stress sum distribution in each of a plurality of CFRP laminates having different internal defects by numerical analysis based on the finite element method (FEM). Further, the learning data generation unit 28 generates learning data by associating the position information of internal defects of each CFRP laminate with the calculated surface principal stress sum distribution. Further, the learning data generation section 28 sends the generated learning data to the model learning section 24.

The learning data generation unit 28 may include the surface principal stress sum distribution in the CFRP laminate having no internal defects in the learning data. By including learning data regarding CFRP laminates that do not have internal defects, it is possible to improve estimation accuracy when internal defects are small.

Specifically, the learning data generation unit 28 calculates the surface principal stress sum distribution as follows. First, a CFRP laminate to be studied is discretized into a finite element mesh. Next, the element stiffness matrix and equivalent nodal force vector are calculated for each element. Next, the element stiffness matrix and equivalent nodal force vector for each element are combined into the global coordinate system. Next, the simultaneous linear equations are solved and the nodal displacements are calculated. Then, strain and stress for each element are calculated from the nodal displacements.

Note that the element stiffness matrix and the overall stiffness matrix can be solved by discretizing the principle of virtual work at nodes. Further, the equivalent nodal force vector can be derived by integrating the surface stress vector by area (see Reference 2).

[Reference 2] Onate, E. "Structural Analysis with the Finite Element Method", Artes Graficas Torres S.L., 2009.

<Effects of the third embodiment>
The defect estimation system in this embodiment is configured to learn an estimation model using learning data generated by numerical analysis. In order to prepare learning data through actual measurements, it is necessary to collect CFRP laminates having various internal defects and measure temperature changes for each with the temperature measuring device 1. Therefore, it takes a huge amount of time to collect learning data.

If learning data is generated by numerical analysis, it becomes possible to learn the estimation model without actually measuring the surface temperature change of the CFRP laminate. Therefore, according to the defect estimating device 2 in this embodiment, the time and economic costs for learning the estimation model can be significantly reduced.

[Test results]
An evaluation test was conducted to evaluate the performance of the defect estimation system in one embodiment. In this test, the surface principal stress sum distribution was calculated for the CFRP laminate to be studied by numerical analysis based on the finite element method described in the third embodiment. Furthermore, for the CFRP laminate to be estimated, the surface principal stress sum distribution was calculated by numerical analysis based on the finite element method.

Specifically, 2496 laminate models with different positions and sizes of internal defects and one laminate model without internal defects were prepared. Among these, one laminate model without internal defects and 1996 laminate models with internal defects were used as learning data, and the remaining 500 laminate models were used as verification data.

The laminate model used in this experiment was a 10-layer laminate with a length of 140 mm, a width of 50 mm, and a thickness of 2 mm. In this laminate, the size varies in the range of 10 to 70 mm in length and 10 to 50 mm in width, and the internal defect exists in any of the 2nd to 9th layers while changing the center coordinates. A body model was generated.

Under the above test conditions, the three-dimensional position of the internal defect was accurately estimated in 497 of the 500 sets of verification data. Even for the three sets for which estimation failed, the outline shape and layer number of the internal defect could almost be identified.

FIG. 12 is a diagram showing the test results. FIG. 12(A) is a diagram illustrating three of the successful estimation cases. FIG. 12(B) is a diagram showing three failure examples in which estimation has failed. In each test result, the upper row is the estimated result, and the lower row is the correct answer. Each rectangle corresponding to channels 1 to 8 represents the presence or absence of internal defects in each mesh in the 2nd to 9th layers. Note that 9 channels are channels added so that the total sum in the stacking direction is 1. A thin region indicates that there is an internal defect, and a dark region indicates that there is no internal defect.

As shown in FIG. 12A, it can be seen that in the successful example, the range of internal defects and the depth of the layer are accurately estimated. As shown in FIG. 12(B), it can be seen that even in the case of failure, the range of internal defects and the depth of the layer can be roughly estimated.

[Application example]
The defect estimation system in the embodiment described above can be used at a CFRP manufacturing facility or a maintenance inspection facility for products using CFRP.

In CFRP manufacturing facilities, when prepregs are laminated, there is a risk that foreign matter may get mixed between the layers, resulting in defective products. Since contamination with foreign matter cannot be determined by appearance, all finished products are inspected by non-destructive testing such as ultrasonic measurement or radiography. However, as described above, these inspection methods have large economic and time costs, which is one reason for reducing productivity.

The defect estimation system in one embodiment can be used for primary screening of defective products. In other words, internal defects are estimated for the entire quantity of finished products using a defect estimation system, and the conventional Destructive testing can be done. As a result, the number of times conventional non-destructive testing, which is costly, can be significantly reduced.

Furthermore, according to the defect estimation system in one embodiment, it is possible to estimate the detailed three-dimensional position of an internal defect. good. Therefore, by applying the defect estimation system in one embodiment to a CFRP manufacturing facility, it is possible to inspect finished products for defects at low cost.

In products using CFRP, damage such as delamination, fiber breakage, and base material cracking occurs due to vibrations or collisions with foreign objects during use. Therefore, at maintenance and inspection facilities, all members made of CFRP are subjected to non-destructive testing such as ultrasonic measurement. As mentioned above, ultrasonic measurement requires a contact member, and the test results are influenced by the skill of the test technician.

As in the case of a manufacturing facility, if the defect estimation system in one embodiment performs a primary inspection on the entire product, it is possible to extract locations where internal defects are likely to exist. Further, in the ultrasonic inspection, it is only necessary to focus the ultrasonic inspection on positions where there is a high possibility that an internal defect exists. Therefore, by applying the defect estimation system in one embodiment to a maintenance inspection facility for products using CFRP, damage analysis of products can be performed at low cost.

[supplement]
Although the embodiments of the present invention have been described in detail above, the present invention is not limited to these embodiments, and various modifications or variations can be made within the scope of the gist of the present invention as described in the claims. Changes are possible.

100 Defect estimation system 1 Temperature measurement device 2 Defect estimation device 3 User terminal 11 Temperature change unit 12 Temperature measurement unit 13 Image generation unit 21 Image reception unit 22 Position information reception unit 23 Feature amount calculation unit 24 Model learning unit 25 Model storage unit 26 Defect position estimation unit 27 Result output unit 28 Learning data generation unit 31 Feature reception unit

Claims (10)

a feature reception unit configured to receive input of feature data based on a difference between a temperature distribution before a temperature change occurs on the surface of the fiber-reinforced plastic laminate and a temperature distribution after the temperature change occurs;
By inputting the feature data received by the feature amount receiving unit into a model that has learned the relationship between the feature data and position information representing the three-dimensional position of the internal defect, a defect position estimator configured to estimate a dimensional position;
Equipped with
The position information is information that divides each layer of the laminate into a predetermined number of meshes and indicates whether or not the internal defect exists in each of the meshes.
Defect estimation device. The defect estimating device according to claim 1,
The mesh is divided into orthogonal equal intervals,
Defect estimation device. The defect estimating device according to claim 1,
The mesh is divided finely into a region near a predetermined position and coarsely into a region different from the neighboring region.
Defect estimation device. The computer is
a feature value acceptance procedure for accepting input of feature data based on a difference between a temperature distribution before a temperature change occurs on the surface of a fiber-reinforced plastic laminate and a temperature distribution after the temperature change occurs;
By inputting the feature data received in the feature amount reception procedure into a model that has learned the relationship between the feature data and position information representing the three-dimensional position of the internal defect, a defect position estimation procedure for estimating a dimensional position;
Run
The position information is information that divides each layer of the laminate into a predetermined number of meshes and indicates whether or not the internal defect exists in each of the meshes.
Defect estimation method. to the computer,
a feature value acceptance procedure for accepting input of feature data based on a difference between a temperature distribution before a temperature change occurs on the surface of a fiber-reinforced plastic laminate and a temperature distribution after the temperature change occurs;
By inputting the feature data received in the feature amount reception procedure into a model that has learned the relationship between the feature data and position information representing the three-dimensional position of the internal defect, a defect position estimation procedure for estimating a dimensional position;
run the
The position information is information that divides each layer of the laminate into a predetermined number of meshes and indicates whether or not the internal defect exists in each of the meshes.
program. A defect estimation device that estimates internal defects of an estimation target stack using an estimation model generated from the learning target stack,
a support part that supports the laminate and applies stress to the laminate;
a temperature measurement unit that measures the temperature before and after applying stress to the laminate and the temperature difference;
an image generation unit that images the temperature difference as a temperature change image;
a feature calculation unit that calculates feature data from the temperature change distribution of the temperature change image;
The calculation results obtained by repeatedly performing the measurement of the temperature difference, the imaging of the temperature change image, and the calculation of the characteristic data, and the calculation results obtained by dividing each layer of the laminate into a predetermined number of meshes, a model learning unit that creates a model for estimating internal defects in the laminate from position information indicating whether or not internal defects exist;
a defect position estimating unit that estimates a three-dimensional position of an internal defect in the estimation target laminate;
Equipped with
supporting the estimation target laminate on the support part;
The temperature measuring unit measures the temperature before and after applying stress to the estimation target laminate and the temperature difference thereof as an estimation target laminate temperature difference,
The image generation unit images the estimation target laminate temperature difference as an estimation target laminate temperature change image,
The feature amount calculation unit calculates estimation target laminate feature data from the temperature change distribution of the estimation target laminate temperature change image,
The defect position estimating unit estimates a three-dimensional position of an internal defect in the estimation target laminate by inputting the estimation target laminate feature data into the estimation model.
Defect estimation device. The defect estimating device according to claim 6,
The feature data calculated by the feature amount calculation unit is a surface principal stress sum distribution of the laminate;
Defect estimation device. The defect estimating device according to claim 6,
The feature data calculated by the feature calculation unit is a surface temperature change distribution of the laminate;
Defect estimation device. The computer is
A procedure for applying stress to the support part that supported the laminate;
a step of measuring the temperature before and after applying stress to the laminate and the temperature difference;
a step of imaging the temperature difference as a temperature change image;
a step of calculating feature data from the temperature change distribution of the temperature change image;
The calculation results obtained by repeatedly performing the measurement of the temperature difference, the imaging of the temperature change image, and the calculation of the characteristic data, and the calculation results obtained by dividing each layer of the laminate into a predetermined number of meshes, a step of creating an estimation model of an internal defect in the laminate from position information indicating whether or not a defect exists;
a step of measuring the temperature before and after applying stress to the estimation target laminate and the temperature difference as the estimation target laminate temperature difference;
a step of imaging the temperature difference of the estimation target laminate as an estimation target laminate temperature change image;
a step of calculating estimation target laminate feature data from the temperature change distribution of the estimation target laminate temperature change image;
a step of estimating a three-dimensional position of an internal defect in the estimation target laminate by inputting the estimation target laminate feature data into the estimation model;
A defect estimation method that performs to the computer,
A procedure for applying stress to the support part that supported the laminate;
a step of measuring the temperature before and after applying stress to the laminate and the temperature difference;
a step of imaging the temperature difference as a temperature change image;
a step of calculating feature data from the temperature change distribution of the temperature change image;
The calculation results obtained by repeatedly performing the measurement of the temperature difference, the imaging of the temperature change image, and the calculation of the characteristic data, and the calculation results obtained by dividing each layer of the laminate into a predetermined number of meshes, a step of creating an estimation model of an internal defect in the laminate from position information indicating whether or not a defect exists;
a step of measuring the temperature before and after applying stress to the estimation target laminate and the temperature difference as the estimation target laminate temperature difference;
a step of imaging the temperature difference of the estimation target laminate as an estimation target laminate temperature change image;
a step of calculating estimation target laminate feature data from the temperature change distribution of the estimation target laminate temperature change image;
a step of estimating a three-dimensional position of an internal defect in the estimation target laminate by inputting the estimation target laminate feature data into the estimation model;
A program to run.

Images