Adhesive force estimation system, adhesive force estimation method, and computer

Abstract

To estimate the strength of adhesive bonding.SOLUTION: An adhesive force estimation system includes: an input interface for receiving an input of information on a mechanical characteristic value of a solidified adhesive and on the surface roughness of an adherend; a storage unit for storing a learned model learned using teacher data in which a combination of a test result of a mechanical characteristic value according to a test piece of a solidified adhesive, and information on the surface roughness of an adherend is used as input, and an adhesive strength is the output; and an adhesive strength prediction unit for estimating an adhesive strength by using the learned model, on the basis of the input to the input interface.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present invention relates to an adhesive strength estimation system, an adhesive strength estimation method, and a computer. [Background technology]

[0002] The increased use of lightweight structural materials with the aim of improving the energy efficiency of mobility is expected to lead to the widespread introduction of adhesive joining.

[0003] Strength reliability is an important factor for structural materials used in vehicles, etc., and it is necessary to accurately evaluate the strength of adhesive bonds in order to judge their soundness.

[0004] The strength of adhesive bonds can be determined by measuring the strength using tensile tests, etc. However, this method destroys the adhesive bonded components, so a non-destructive method is required to determine the strength.

[0005] Moreover, from the viewpoint of industrial application, it is necessary to estimate the strength of each individual adhesive joint, rather than estimating the average strength.

[0006] Non-Patent Document 1 attempts to estimate the adhesive strength using strength evaluation data of adhesive members by the finite element method. The method of numerical simulation using the finite element method can give the average strength of adhesive members non-destructively.

[0007] Patent Document 1 proposes a method for non-destructively estimating the fracture mode by accumulating data on adhesive bonding. This method estimates the fracture mode by providing the measured adhesive strength as an input value.

[0008] Non-Patent Document 2 claims that, based on the technology of Patent Document 1, adhesive strength can be estimated non-destructively by inputting the type of adhesive, the material of the adherend, the surface treatment conditions, etc. [Prior art documents] [Patent documents]

[0009] [Patent Document 1] Patent Publication No. 2022-37646 [Non-patent literature]

[0010] [Non-Patent Document 1] Z. Gu, Composites Part B 217 (2021) 108894, pp.2-3 [Non-Patent Document 2] Panasonic Holdings Co., Ltd., "Adhesive AI Diagnosis", [online], [Retrieved February 9, 2023], Internet, URL: https: / / holdings.panasonic / jp / corporate / pac / reliability / durability / adhesiveai.html Summary of the Invention [Problem to be solved by the invention]

[0011] The numerical simulation method found in Non-Patent Document 1 does not allow evaluation of the adhesive strength of individual adhesive bonds.

[0012] According to the estimated strength shown in Non-Patent Document 2, the actual measured value is about 4 MPa, but the estimated strength is sometimes about 12 MPa, which means that there is an error of three times the actual measured value. This corresponds to, for example, an adhesive bond predicted to withstand a load of 30 kg actually withstanding a load of only 10 kg. In order to further improve the reliability of adhesive bonds, it is necessary to improve the prediction accuracy of adhesive strength.

[0013] In order to solve the above problems, the present inventors have come up with a method for estimating the strength of adhesive bonds for individual adhesive members by using information on the adherend and physical properties of the adhesive in combination. [Means for solving the problem]

[0014] An example of an adhesive strength estimation system according to the present invention includes: an input interface that accepts input of mechanical property values ​​of the solidified adhesive and information on the surface roughness of the adherend; A memory unit for storing a trained model trained with training data in which a combination of test results of mechanical property values ​​of test pieces of solidified adhesive and information on the surface roughness of an adherend is input, and adhesive strength is output; an adhesive strength estimation unit that estimates an adhesive strength using the trained model based on an input to the input interface; Equipped with.

[0015] In one example of an adhesive strength estimation method according to the present invention, the adhesive strength estimation system described above estimates the adhesive strength.

[0016] An example of a computer according to the present invention stores data correlating a combination of test results of mechanical property values ​​of test pieces of solidified adhesive, information on the surface roughness of the adherend, and adhesive strength. Effect of the Invention

[0017] According to the present invention, it is possible to estimate the strength of an adhesive bond with high accuracy. [Brief description of the drawings]

[0018] [Figure 1] FIG. 1 is a diagram showing a basic configuration of an adhesive strength estimation system according to a first embodiment of the present invention. [Diagram 2] This is a procedure for estimating adhesive strength using the adhesive strength estimation system. [Diagram 3] This is a procedure that only estimates the adhesive strength. [Figure 4] FIG. 2 is a more specific configuration diagram of each element. [Diagram 5] 1 is an example of a configuration of teacher data. [Figure 6] The adhesive comes into contact with an adherend having a surface unevenness with a gentle gradient. [Figure 7] The contact of the adhesive with the adherend has a steep gradient in the surface irregularities. [Figure 8] FIG. 1 is a diagram of an adhesive bond. [Figure 9] FIG. 1 is a diagram showing measurement positions of line roughness. [Figure 10] The results show the estimated adhesive strength. [Figure 11] FIG. 4 is an explanatory diagram of thickness measurement points. [Figure 12] This is the estimation result of the adhesive strength using dataset 4. [Figure 13] This is the estimation result of the adhesive strength using dataset 1. [Figure 14] This is the relationship between the utilization rate of information related to RΔq and the estimation accuracy. [Figure 15] This is the data format of dataset 7. [Figure 16] This is the data format of dataset 8. [Figure 17] This is a data processing procedure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] Hereinafter, the embodiment of the present invention will be described in detail with reference to the drawings. With the advancement of technology in the field of information science, such as machine learning, there is an increasing movement to apply it to material development. The movement to apply machine learning and other techniques to material development is generally called Materials Informatics (MI). The present invention was born from research on MI.

[0020] Typically, in one embodiment of MI, machine learning is applied to create a mathematical model to estimate the physical properties of a material from the material's composition information, process information, etc. This method is known to be able to estimate the material properties, etc., and can also be used to estimate the strength of adhesive bonds.

[0021] The concept of adhesive bond strength is generally understood in terms of the work of adhesion. The work of adhesion is the energy that bonds the adhesive and the adherend, and is equal to the work required to separate the adhesive and the adherend and create two new surfaces. This relationship is expressed by an equation called the Dupre equation. However, the work of adhesion is purely a thermodynamic concept, and is the sum of the work done until the adhesive interface is completely separated.

[0022] In order to discuss the strength of practical adhesive bonds, it is necessary to consider the microscopic situation when the adhesive interface breaks down. That is, in the actual tensile fracture phenomenon of adhesive bonds, if there is an area where the bonding strength of the adhesive interface is locally weakened due to the influence of surface unevenness, for example, micro-scale fracture first occurs in the area when stress is applied, and such cracks gradually grow and progress according to the stress, forming a peeling area, and the continued stress eventually leads to total fracture.

[0023] In the fracture process of such microscopic adhesive bonds, the propagation of peeling due to stress is the main process, and the energy required to propagate the peeling is much smaller than the work of adhesion, meaning that the work of adhesion is not appropriate for describing practical adhesive strength.

[0024] One method for estimating adhesive strength that takes into account the peeling process of an actual adhesive interface is numerical simulation using the finite element method. This method can be used to simulate crack growth and calculate the adhesive strength in a pseudo manner. However, the adhesive strength calculated using the finite element method is not sufficiently accurate, and it cannot be used to estimate the adhesive strength of individual components.

[0025] Therefore, the present inventors have devised a means for achieving high-precision estimation based on machine learning methods in order to estimate with high accuracy the adhesive strength of adhesively bonded individual components, taking into account the peeling process of the actual adhesive interface.

[0026] The adhesive strength estimation method is specifically described below. The adhesive strength estimation method is divided into two steps. One is to construct an estimation model, and the other is to estimate the adhesive strength using the estimation model.

[0027] Adhesion Strength Estimation System of the First Embodiment 1 is a diagram showing a basic configuration of an adhesive strength estimation system 109 in embodiment 1. The adhesive strength estimation system 109 estimates adhesive strength by the process described in this specification. In this specification, "estimation" can be replaced with "prediction".

[0028] The adhesive force estimation system 109 has a hardware configuration as a known computer, and includes, for example, a calculation means and a storage means. The calculation means includes, for example, a processor, and the storage means includes, for example, a storage medium such as a semiconductor memory device and a magnetic disk device. A part or all of the storage medium may be a non-transitory storage medium.

[0029] The storage means may store a program. The processor may execute the program, so that the computer functions as the adhesive strength estimation system 109.

[0030] The adhesive strength estimation system 109 is a device including a memory unit 104, an estimation model construction unit 105, an adhesive strength estimation unit 106, an input interface 103, and an output interface 107. These components are realized by the cooperation of a computing means and a memory means of a computer.

[0031] The input interface 103 is a device for inputting teacher data 101 used to train the estimation model and experimental data 102 for determining adhesive strength 108 of an adhesive bond whose adhesive strength is unknown. There is no limitation on the method of inputting the data, and data stored in a file may be read from a reading device, or the data may be input from an input device consisting of a monitor and a keyboard.

[0032] The storage unit 104 can store the teacher data 101, the experimental data 102, and the estimation model constructed in the estimation model construction unit 105.

[0033] The estimation model construction unit 105 constructs an estimation model of the adhesive strength by machine learning based on the teacher data 101.

[0034] An adhesive strength estimation unit 106 calculates an adhesive strength 108 corresponding to the experimental data 102, using the adhesive strength estimation model constructed by the estimation model construction unit 105. The adhesive strength 108 is output from an output interface 107. There is no limitation on the method of outputting the data, and the data may be directly written to a file on a storage medium, may be displayed on a monitor, or may be transmitted via a communication network.

[0035] The procedure for estimating adhesive strength using the adhesive strength estimation system 109 is shown in Fig. 2. First, the teacher data required to construct an estimation model is prepared (step S1). Next, the teacher data is input to the adhesive strength estimation system 109 (step S2), and an estimation model is constructed based on the teacher data (step S3). After that, experimental data is prepared to determine the adhesive strength of an adhesive bond whose adhesive strength is unknown (step S4). The experimental data is input (step S5), and the constructed estimation model is used to calculate and estimate the adhesive strength corresponding to the experimental data (step S6). Finally, the adhesive strength is output (step S7), and the process ends.

[0036] When estimating adhesive strength, it is not necessary to reconstruct the estimation model every time. In other words, after constructing the estimation model, the adhesive strength can be estimated any number of times using the model. The procedure in this case is shown in Fig. 3. Since the estimation model has already been constructed, it is sufficient to start from preparing experimental data (step S11). As shown in the figure, the experimental data is input (step S12), and then the adhesive strength corresponding to the experimental data is estimated by calculation using the estimation model stored in the memory unit (step S13), and the adhesive strength is output (step S14), completing the process.

[0037] A more specific configuration diagram for each element is shown in Figure 4. First, teacher data 401 is prepared in which measurement data 402 of the mechanical properties (mechanical property values) of the solidified adhesive and measurement data 403 of the surface line roughness of the adherend are used as explanatory variables, and measurement data 404 of the adhesive strength of the corresponding adhesive bond is used as the objective variable.

[0038] The estimation model construction device 405 has a hardware configuration as a known computer, and includes, for example, a calculation means and a storage means. The calculation means includes, for example, a processor, and the storage means includes, for example, a storage medium such as a semiconductor memory device or a magnetic disk device.

[0039] An input interface 406 of the estimation model construction device 405 inputs this teacher data 401 and sends it to an estimation model construction unit 407. The teacher data 401 is stored in a memory unit (not shown) of the estimation model construction device 405. At this point, the estimation model construction device 405 stores the teacher data 401, i.e., data in which a combination of test results of mechanical property values ​​using test pieces of the solidified adhesive and information on the surface roughness of the adherend are associated with adhesive strength. By using data with such a structure, machine learning, which will be described later, can be performed appropriately.

[0040] The estimation model construction unit 407 constructs an estimation model based on a machine learning method, and outputs a trained model 409 from an output interface 408.

[0041] Although not shown in Fig. 4, the trained model 409 is stored in the memory unit 104 (see Fig. 1) of the adhesive strength estimation system 109. That is, the memory unit 104 stores the trained model 409 trained with teacher data 401 that receives as input a combination of test results of mechanical property values ​​using test pieces of solidified adhesive and information on the surface roughness of an adherend, and outputs adhesive strength.

[0042] In order to estimate the adhesive strength of an unknown adhesive bond, an adhesive strength 417 is obtained as an estimated value by the adhesive strength estimation system 109 using experimental data 410 consisting of mechanical property data 411 (mechanical property values) obtained by solidifying the same type of adhesive as that used in the adhesive bond and measuring its mechanical property values, and measurement data 412 of the surface line roughness of the adherend.

[0043] Specifically, experimental data 410 is input from the input interface 103, and the adhesive strength estimation unit 106 processes the data using a trained model 409 constructed based on the machine learning processing to estimate the adhesive strength, and outputs the adhesive strength 417 from the output interface 107. Here, the input interface 103 accepts input of mechanical property values ​​of the solidified adhesive and information on the surface roughness of the adherend, and the adhesive strength estimation unit 106 estimates the adhesive strength using the trained model 409 based on the input to the input interface 103.

[0044] In the example of FIG. 4, line roughness is used as the surface roughness, but area roughness may be used instead of or in addition to line roughness.

[0045] There are various methods for solidifying adhesives to measure their mechanical properties. For example, it is preferable to use a silicone or polyethylene mold to obtain JIS K 7139 Type A1 test piece dimensions, pour in an adhesive mixture of the base agent and hardener, and use a desiccator or the like to mold the adhesive into a dumbbell test piece according to the specified hardening temperature and hardening time.

[0046] It is preferable that the method of solidifying the adhesive be consistent across the teacher data and the experimental data. If data solidified using different methods is mixed, the estimation accuracy will decrease, which is not preferable. The same applies to all data below, and it is preferable that the test method be consistent across the teacher data and the experimental data.

[0047] There are various methods for measuring the mechanical properties of solidified adhesives. For example, it is recommended to measure the mechanical properties according to JIS K-7161. From the stress-strain diagram measured in this way, mechanical properties such as tensile stress, tensile stress at yield, tensile strength, tensile modulus, and breaking elongation can be obtained.

[0048] It is preferable to prepare a plurality of samples of the solidified adhesive and measure the mechanical properties of each sample. The number of repeated preparations is preferably 5 or more. It is preferable to remove the maximum and minimum values ​​of the obtained mechanical properties and adopt the average value of the mechanical properties of the remaining samples as the mechanical property value of the adhesive.

[0049] There are various methods for measuring the surface roughness of an adherend. For example, it can be measured using a one-shot 3D shape measuring instrument. Specifically, the surface of the adherend is degreased by wiping it with hexane or acetone, and then the surface roughness is measured. The following formulas can be used to express the surface roughness: maximum peak height, maximum valley depth, maximum height, average height, arithmetic mean roughness, maximum cross-sectional height, root-mean-square height, skewness, kurtosis, element mean length, root-mean-square slope, ten-point average roughness, etc. The adhesive roughness is measured on both surfaces of the two adherends that form an adhesive bond.

[0050] The line for measuring line roughness is preferably a straight line in the adhesive area on the adherend surface. The end points of the line should preferably be the boundaries of the adhesive area. The line should also preferably pass through the center of the adhesive area, and if the adhesive strength is considered to be tensile shear strength, the line should preferably be parallel to the shear direction. However, these are just examples, and it is not necessary to meet these conditions.

[0051] There are various methods for preparing adhesive joint samples to obtain strength data for adhesive joints. Types of joints for joining two adherends include lap joints, butt joints, double butt joints, slit lap joints, slit joints, indented joints, stepped lap joints, etc. Any type of joint may be selected, but it is preferable to use strength values ​​for one type of joint in the same data. It is not preferable to mix strength values ​​for different types of joints in the same data, as this reduces the estimation accuracy.

[0052] In this embodiment, a regression method based on supervised machine learning is used to construct an estimation model. That is, a means capable of constructing a regression equation for calculating a response variable using explanatory variables is used. There are various methods for regression algorithms, such as multiple linear regression, support vector regression, random forest regression, regression by neural network, and regression based on gradient boosting, and any method may be used.

[0053] When constructing an estimation model, it is preferable to take care that the model does not overlearn the training data. One such method is the cross-validation method. In the cross-validation method, K is set to be 2 or more and less than the number of records in the training data, the training data is divided into K pieces, one of the divided pieces is used as the validation data, and the remaining K-1 pieces are used as the training data to perform training. Then, K models with different estimation accuracies are constructed, and the average estimation accuracies of these are recorded. Next, the hyperparameters are changed, and K models are constructed for the K divided data in the same way, and the average estimation accuracy is calculated. If this value is higher than the average estimation accuracy of the previous time, this model is adopted as a candidate for the generalization model. By repeating this process, an optimized generalization model is obtained.

[0054] Hyperparameters are parameters that are determined in advance before training a model based on machine learning. In the case of random forest regression, the hyperparameters are the number of decision trees, which are weak learners, and the number of branches. There are various methods for searching for hyperparameters, such as the grid search method and the search method using Bayesian estimation, and any method may be used.

[0055] The constructed estimation model is used to estimate the adhesive strength. That is, experimental data is input to the estimation model to calculate an estimated value of the adhesive strength.

[0056] According to this embodiment, since the adhesive strength is estimated from the explanatory variables included in the experimental data, it is not necessary to measure the adhesive strength of the adhesive joints included in the experimental data for the purpose of estimation. If there is an experimental value of the adhesive strength for the adhesive joints included in the experimental data, the estimation accuracy can be calculated by comparing it with the estimated value of the adhesive strength.

[0057] The estimation accuracy is calculated by the difference between the adhesive strength calculated by applying the constructed estimation model to the experimental data and the experimental value of the adhesive strength of the adhesive joint included in the experimental data. When the difference is evaluated by the correlation coefficient, it can be evaluated dimensionlessly, and when the difference is evaluated by the root mean square error, it can be evaluated in units of MPa.

[0058] An example of the configuration of training data is shown in Figure 5. In this example, the training data is configured in a matrix format, but any format is acceptable as long as the objective variable is linked to the explanatory variable. For ease of understanding, the first column lists the sample numbers of the adhesively bonded samples, but the sample numbers are not used in model construction.

[0059] The second column contains the adhesive strength. For example, the adhesive strength can be expressed as tensile shear strength. This is the column in which the measured adhesive strength values ​​corresponding to the sample numbers are entered, and this is the response variable.

[0060] All columns from the third onwards are used as explanatory variables. Here, breaking strength, breaking elongation, elastic modulus, maximum peak height, root mean square slope, and ten-point average roughness are entered as examples. Breaking strength may be expressed in tensile stress or tensile yield stress, breaking elongation may be expressed in tensile elongation, and elastic modulus may be expressed in tensile modulus. These are values ​​corresponding to the sample number specified in the corresponding row, and are the measured values ​​of the mechanical properties (breaking strength, breaking elongation, elastic modulus) of the adhesive used for that sample when solidified, and the linear roughness (maximum peak height, root mean square slope, ten-point average roughness) of the adherend surface used for that sample.

[0061] As explanatory variables, in addition to the mechanical properties of the solidified adhesive and the measured values ​​of the line roughness of the adherend surface, any explanatory variables can be used. For example, since it is known that the thickness of the adhesive layer affects the adhesive strength, the thickness of the adhesive layer can be added as an explanatory variable.

[0062] In the example of FIG. 5, each training data (one record) includes data on a single adherend as an explanatory variable, i.e., the adhesive strength is expressed as the strength when two adherends of the same kind are bonded together. However, each training data may include data on two adherends as explanatory variables. For example, the breaking strength may include the breaking strength of a first adherend and the breaking strength of a second adherend. In such a configuration, the adhesive strength is expressed as the strength when two adherends of different kinds are bonded together.

[0063] Although not shown in FIG. 5, the training data includes test results of the mechanical properties of the adhesive.

[0064] The inventors have investigated the accuracy of estimating adhesive strength using the explanatory variables described above. As a result, they found that explanatory variables related to the mechanical property values ​​(e.g., mechanical strength) of the adhesive are effective in expressing differences in adhesive strength due to adhesives. However, with a method of estimating adhesive strength using only the mechanical property values ​​of the adhesive, it is theoretically impossible to determine the strength of each individual adhesive member, even if it is possible to determine the average strength of the adhesive members.

[0065] It is known that there is variation in the adhesive strength of individual adhesive materials. Even if the same worker performs the adhesive work according to a predetermined work procedure, the adhesive strength will not be completely equal. For example, when polishing the surface of an adherend, particles scraped off from the adherend surface may get between the adherend and the polishing body, and even if the polishing state appears to be the same to the naked eye, there are differences on the order of micrometers, such as the surface not being the same.

[0066] As a result of investigating the adhesive strength, the inventors found that, although both the surface roughness and the line roughness of the surface can affect the adhesive strength of an individual adhesive member, the factor that has the greatest influence is the line roughness, not the surface roughness. Among the line roughness factors, they found that the influence of factors related to the inclination of the slope of the surface unevenness is large.

[0067] To investigate why the inclination of the slope has a large effect on adhesive strength, the inventors considered the inclination of the slope of the surface unevenness of the adherend and the state of contact with the adhesive. Figure 6 shows the state of contact of an adhesive with an adherend having a gently sloping surface unevenness. Adherend 602 has an uneven surface, and the inclination of the unevenness is gentle. Adhesive 601 is applied to the adherend surface and comes into contact in a liquid state. Because the inclination of the unevenness is gentle, the adhesive reaches near the bottom of the valleys, and the contact area between the adhesive and the adherend is large. For this reason, it is believed that adhesive strength is greater when the inclination of the unevenness is gentle.

[0068] On the other hand, the contact state between the adherend and the adhesive when the adherend has a steep incline in the surface unevenness is shown in Figure 7. It shows the state in which adhesive 701 has been applied to the surface of adherend 702. Because the incline of the adherend's surface unevenness is steep, it is difficult for the adhesive to reach the bottom of the valleys, resulting in large gaps being formed between the adherend and the adhesive, which in turn reduces the adhesive area. It is believed that this is how the adhesive strength decreases.

[0069] A representative example of such factors is the root-mean-square slope. If the surface roughness information includes the root-mean-square slope, the adhesive strength can be more appropriately estimated. Similarly, factors whose absolute value of correlation with the root-mean-square slope is 0.6 or more are also effective. An example of such a factor is the ten-point mean roughness. If the surface roughness information includes the ten-point mean roughness, the adhesive strength can be more appropriately estimated.

[0070] Ten-point average roughness is a quantity that represents the sum of the average of the five largest peaks on a profile curve and the average of the five largest valleys on a reference length. Therefore, if the ten-point average roughness is large, it means that the valleys are deep when viewed from the highest peaks, which means that it is difficult for adhesive to reach the bottom of the valleys, and it can be strongly correlated with the root-mean-square slope.

[0071] According to the inventors' study, it was found that this effect is particularly pronounced when the adherend is carbon fiber reinforced plastic (CFRP). CFRP is a composite material that is lightweight yet has significantly improved resin strength, achieved by weaving together bundles of carbon fibers and shaping the components into a sheet-like shape, which is then impregnated with resin. In the manufacture of CFRP, liquid resin is impregnated into the carbon fibers, and the resin is then solidified. When the inventors investigated the surface irregularities of CFRP, they found that irregularities are formed on the surface of CFRP that reflect the weaving state of the carbon fibers inside. It is believed that such irregularities affect the adhesive strength of the adhesive member.

[0072] In this way, the adhesive strength estimation system 109 according to this embodiment can estimate the strength of the adhesive bond with high accuracy by using both the information on the adherend and the physical properties of the adhesive.

[0073] The present invention will be described in more detail below with reference to various examples, although the present invention is not limited to the configurations and / or structures described in these examples.

[0074] [Example 1] In this example, the adhesive strength between an aluminum plate and a CFRP plate using an epoxy adhesive A is estimated.

[0075] As shown in Figure 8, a single lap joint structure was fabricated by bonding an aluminum alloy plate 801 and a CFRP plate 802 with adhesive 803. The fabrication method was in accordance with JIS K 6850, and aluminum alloy and CFRP plates measuring 100 mm x 25 mm x 2 mm were bonded with adhesive. A standard test piece (type: JISK6850AJAA01) was used as the CFRP, and A5083P and ADC-12 were used as the aluminum alloy.

[0076] For adhesion, first, the adherend surface was degreased (wiped with hexane or acetone) to clean it, and then roughened with sandpaper to control the breaking strength. Sandpaper used was #80, #400, #1000, and #2000, and five types of samples were prepared, including an untreated sample. In addition, some samples were UV-treated (for 5 minutes) to improve the hydroxyl group bonding property of the Al surface.

[0077] The adhesive used was a two-part epoxy adhesive A, which is strong enough to be applied to the bonded surfaces after mixing thoroughly, and a Teflon (registered trademark) spacer was used to bond the adhesive to a thickness of 0.1 mm. After bonding, the adhesive was left to harden for 24 hours at room temperature in a desiccator.

[0078] 74 single lap joint specimens were prepared as samples for use in constructing the strength estimation model.

[0079] The surface unevenness was measured before bonding the adherend. The surface roughness was measured by measuring the areal roughness and line roughness using a one-shot 3D shape measuring instrument (Keyence VR3000).

[0080] As indexes of surface roughness, seven types of surface roughness values ​​were measured for an adhesive area of ​​12.5 mm x 25 mm: root mean square height, maximum peak height, maximum valley depth, arithmetic mean height, maximum height, skewness, and kurtosis. There are seven types of measurements for the surface roughness of one adherend, and since there are two adherends, the number of measurements for the surface roughness of one pair of adherends is 7 x 2 = 14.

[0081] The positions where the line roughness was measured are shown in Figure 9. The adherend 909 is divided into an adhesive region 907 and a non-adhesive region 908. The line roughness was measured in the adhesive region 907, and measurements were taken at a total of three points for line segment A 901, line segment B 902, and line segment C 903, each 10 mm long, along the longitudinal direction of the plate, and at a total of three points for line segment D 904, line segment E 905, and line segment F 906, each 22 mm long, perpendicular to the longitudinal direction of the plate. In this way, a total of six line segments were measured.

[0082] Line segments B902 and E905 were set so that they passed through the center of the adhesive area. The other lines were set at positions far enough away from the center line segment, since if they were close to the center line segment, the values ​​would be nearly the same. However, if they were close to the edge of the adhesive area, they would be affected by deformation caused by cutting when preparing the adherend. Therefore, line segments A901, C903, D904, and F906, which do not pass through the center of the adhesive area, were measured at positions 2.25 mm away from the edge of the adhesive area.

[0083] Twelve types of line roughness indices were measured: maximum peak height, maximum valley depth, maximum height, average height, arithmetic mean roughness, maximum cross-sectional height, root mean square height, skewness, kurtosis, average element length, root mean square slope, and ten-point average roughness. Since there are 12 types of line roughness values ​​for a total of six line segments, the number of measured values ​​for the line roughness of one adherend is 6 x 12 = 72. Since there are two adherends, the number of measured values ​​for the line roughness of one pair of adherends is 72 x 2 = 144.

[0084] To estimate the adhesive strength of adhesive joints, a strength estimation model was constructed using 158 explanatory variables related to surface roughness (14 types of surface roughness and 144 types of line roughness). The model was constructed using the Random Forest method, a type of regression method in machine learning. The Random Forest method is a method for constructing a regression model using a large number of decision trees. In this study, this method was used because the sandpaper numbers used to roughen the sample surface were discrete and there were many factors that were determined by branching depending on whether or not treatment was performed, such as UV treatment. In addition, this method has the advantages of being able to model nonlinear behavior and being less prone to overlearning, making it suitable for estimating phenomena in which a large number of factors are intertwined and complex behavior is expected, as in this example.

[0085] Of the 74 records in the created data, 66 were used as training data and the remaining 8 were used as experimental data, and an estimation model for adhesive strength was then constructed. Figure 10 shows the estimation results for adhesive strength using the estimation model. The horizontal axis is the experimental adhesive strength value, and the vertical axis is the estimated adhesive strength value. If the estimated value matched the experimental value, the plot would be located on the dotted line. In this estimation result, the plot was close to the dashed line, and the estimated 8 points generally gave strength similar to the experimental value. The maximum difference between the experimental value and the theoretical value was 2 MPa or less. The correlation coefficient between the experimental value and the estimated value was R = 0.88, which means that the estimation was performed with high accuracy.

[0086] Table 1 shows the top 10 variables with respect to the importance of the variables used in the estimation model using the out-of-bag estimation method.

[0087] [Table 1]

[0088] A higher variable importance indicates a greater influence on the objective variable. The variable with the greatest influence on the adhesive strength was the root-mean-square slope of line segment B902 on the CFRP surface. Similarly, the variable importance of the root-mean-square slopes of line segments A901, D904, and E905 was also high, so it can be said that the root-mean-square slope determines most of the adhesive strength. Among these, the importance of line segment B902 accounts for more than half, and it can be seen that line segment B902, which is parallel to the tensile direction and passes through the center of the adhesive region, is the most important.

[0089] As an index of line roughness, not only the root mean square slope but also the average 10-point roughness had importance, although its importance was much lower than that of the root mean square slope.

[0090] [Example 2] In this example, the adhesive strength between an aluminum plate and a CFRP plate is estimated when various types of adhesives are used.

[0091] In this example, 12 types of adhesives were selected. All adhesives are two-part types, and can be hardened and bonded by mixing the main agent with a hardener in a specified ratio. In general, epoxy adhesives have high tensile shear strength but are brittle, urethane adhesives have low tensile shear strength but good elongation, and acrylic adhesives have intermediate properties between epoxy and urethane adhesives. For this reason, acrylic, epoxy, and urethane adhesives were used without bias.

[0092] The types of adhesives used are shown in Table 2.

[0093] [Table 2]

[0094] Seven types of epoxy adhesives, A to G, were used. Two types of acrylic adhesives, H and I, were used. Three types of urethane adhesives, J, K, and L, were used.

[0095] The mechanical properties of the solidified adhesive obtained for each adhesive A to L were the elastic modulus, breaking strength, and breaking elongation. In this way, if the mechanical property values ​​of the solidified adhesive include at least one of the elastic modulus, breaking strength, and breaking elongation, the dependency of the adhesive strength on the adhesive properties can be reflected in the estimation model by using the property values ​​of many types of adhesives that change over a wide range, making it possible to estimate the adhesive strength with higher accuracy. In addition, the adhesive viscosity was used to describe the affinity of the adhesive with the adherend surface when the adhesive is applied. A total of four parameters were obtained.

[0096] The elastic modulus, breaking strength, and breaking elongation were obtained by preparing dumbbell test pieces of the adhesive and conducting tensile tests. The dumbbell test pieces were molded in a silicone or polyethylene mold to have the dimensions of JIS K 7139 Type A1 test pieces, and the adhesive mixed with the base agent and hardener was poured into the mold. Five samples were prepared for each type of adhesive (N=5), and the mechanical properties of each sample were measured and the average value was calculated and used. The dumbbell tensile test was evaluated using a tensile tester (Autograph AG-100kNX, Shimadzu Corporation). The tensile speed was 1 mm / min, and the test force at break was measured as the breaking strength, and the elongation at break was also measured. The slope of the plot in the stress-strain curve from 0.05% to 0.25% strain range was calculated as the elastic modulus.

[0097] Regarding viscosity, since hardening starts immediately when the base resin and hardener are mixed, it is difficult to experimentally measure the viscosity of the two-liquid mixture. Therefore, the viscosity of the mixture was estimated using the Kendahl equation, which shows that the viscosity of the mixture is additivity between the viscosity of the base resin and the viscosity of the hardener.

[0098] Adhesive bonded samples were prepared in the same manner as in Example 1. That is, a single lap joint structure was prepared as shown in Fig. 8. However, adhesive bonded samples were prepared for each of 12 types of adhesives, instead of one type of adhesive.

[0099] For adhesion, first, the adherend surface was degreased (wiped with hexane or acetone) to clean it, and then roughened with sandpaper to control the breaking strength. The sandpaper used was #2000, and two types of samples were prepared, including an untreated one.

[0100] The surface unevenness was measured before the adherend was bonded. Surface roughness and line roughness were measured using a one-shot 3D shape measuring instrument (Keyence VR3000). Seven types of surface roughness values, including root mean square height, maximum peak height, maximum valley depth, arithmetic mean height, maximum height, skewness, and kurtosis, were measured for an adhesion area of ​​12.5 mm x 25 mm as indexes of surface roughness. Linear roughness was measured as shown in FIG. 9 in the same manner as in Example 1. Twelve types of indexes of line roughness were measured, including maximum peak height, maximum valley depth, maximum height, average height, arithmetic mean roughness, maximum cross-sectional height, root mean square height, skewness, kurtosis, average element length, root mean square slope, and ten-point average roughness.

[0101] The adhesive was applied to the bonding surface after thoroughly mixing the base agent and hardener, and then bonded using a Teflon spacer so that the adhesive layer was 0.1 mm thick. For epoxy adhesives B and C, additional samples were prepared so that the adhesive layer was 1.0 mm thick in order to examine the effect of the adhesive layer thickness. After bonding, the samples were left to harden at room temperature for 24 hours in a desiccator.

[0102] The number of adhesive bonding samples produced was 73. The breakdown is as follows: 11 x 3 x 2 = 66 samples (N = 3 for each of the 11 types of adhesives excluding epoxy adhesive A, with and without surface polishing treatment, two samples each), one sample produced with N = 1 using epoxy adhesive A without roughening treatment for comparison, and 2 x 3 = 6 samples with increased adhesive layer thickness for epoxy adhesive B and epoxy adhesive C, for a total of 66 + 1 + 6 = 73 samples.

[0103] The adhesive layer thickness and the inclination of the adhesive layer were adopted as explanatory variables. To take into account the variation in the adhesive layer thickness in the single lap joint structure, the sample thickness was measured with a micrometer at three points each for the aluminum plate and the CFRP plate, and at six points in the adhesive region. The adhesive layer thickness was calculated by subtracting the average thickness of the aluminum plate and the average thickness of the CFRP plate from the average thickness of the adhesive region.

[0104] The thickness measurement points will be explained with reference to Fig. 11. Fig. 11(a) shows a side view and Fig. 11(b) shows a top view of a sample in which an aluminum plate 1101 and a CFRP plate 1102 are bonded by an adhesive layer 1103. In Fig. 11(b), three measurement points, measurement point A 1107, measurement point B 1108, and measurement point C 1109, are taken on a line perpendicular to the long side of the aluminum plate 1101. The average value of these three points is calculated to obtain the average thickness of the aluminum plate 1101.

[0105] Similarly, the average thickness of the CFRP plate is obtained by measuring the thickness at three measurement points, measurement point J1116, measurement point K1117, and measurement point L1118, and calculating the average value.

[0106] The average thickness of the adhesive region is calculated by measuring the thickness at six measurement points, namely, measurement point D 1110, measurement point E 1111, measurement point F 1112, measurement point G 1113, measurement point H 1114, and measurement point I 1115. The adhesive thickness is obtained by subtracting the average thickness of the aluminum plate 1101 and the average thickness of the CFRP plate 1102 from the average thickness of the adhesive region.

[0107] The explanatory variables used to estimate the adhesive strength of the adhesive bonding member were 158 explanatory variables related to surface roughness (14 types of surface roughness and 144 types of line roughness), three types of mechanical properties of the adhesive, adhesive viscosity, and adhesive layer thickness, for a total of 163 types. The number of records used was 147 records in total, including 74 records prepared in Example 1 and 73 records prepared in this example. The model was constructed using the random forest method, which is a type of regression method in machine learning. A strength model was constructed and estimated using 146 of the 147 records as training data and the remaining 1 record as experimental data, and this was repeated for all data to calculate the average estimation accuracy.

[0108] In order to consider the extent to which the root mean square slope RΔq of the line segment B902 on the CFRP surface, among the line roughness indices, and related information have an impact on the adhesive strength, a data set was created in which line roughness indices with a high correlation coefficient with RΔq were removed from all data. Specifically, the correlation coefficient between RΔq and each line roughness indices was calculated, and a data set was created in which indices whose value exceeds φ were removed. This φ is called the usage rate of RΔq-related information. Here, four types of data sets were created with φ=1.0, 0.6, 0.4, and 0.3, and are called Data Set 1, Data Set 2, Data Set 3, and Data Set 4, respectively. When φ=1.0, it is a complete data set with nothing removed, and when φ=0.3, it is the data set with the lowest usage rate of information related to RΔq.

[0109] Figure 12 shows the results of constructing an estimation model and estimating adhesive strength using Dataset 4, and evaluating its accuracy. The horizontal axis represents the experimental adhesive strength value, and the vertical axis represents the estimated adhesive strength value. If the adhesive strength could be accurately estimated, the plot of the estimated value would be on the dashed line. Here, the plot points vary due to error, and the correlation coefficient between the experimental and estimated adhesive strength values ​​was R = 0.66. This shows that the estimation accuracy is low when the usage rate of related information for RΔq is only 0.3.

[0110] An estimation model was constructed and its accuracy was evaluated using Dataset 1. The results of adhesion strength estimation are shown in Figure 13. The horizontal axis represents the experimental value of adhesion strength, and the vertical axis represents the estimated value of adhesion strength. In this case, the correlation coefficient between the experimental and estimated values ​​of adhesion strength was R = 0.80. It can be seen that the usage rate of related information for RΔq became 1.0, which improved the estimation accuracy compared to when Dataset 4 was used.

[0111] When adhesive strength estimation was performed using dataset 2 in the same manner as in the verification of estimation accuracy for dataset 1 and dataset 4, the correlation coefficient between the experimental and estimated adhesive strength values ​​was R = 0.79. Similarly, when adhesive strength estimation was performed using dataset 3, the correlation coefficient between the experimental and estimated adhesive strength values ​​was R = 0.76. Figure 14 shows a graph of the relationship between the usage rate of information related to RΔq and estimation accuracy. It can be seen that when the usage rate of information related to RΔq exceeds 0.4, accuracy improves rapidly.

[0112] From this, the surface roughness information is a line roughness having a correlation coefficient with the root mean square slope of 0.4 or more, and - Surface roughness with a correlation coefficient of 0.4 or more with the root mean square slope It can be said that a more accurate estimation is possible by configuring the estimation method so as to include at least one of the indicators.

[0113] [Example 3] In this example, the adhesive strength between an aluminum plate and a CFRP plate was estimated when various types of adhesives were used, as in Example 2, but the estimation accuracy was improved by dividing the data set. In the adhesive strength estimation method according to this example, the adhesive strength estimation system estimates the adhesive strength by using an adhesive strength estimator for one type of adhesive based on surface roughness information of the adherend as part of an estimator for the adhesive strength of another type of adhesive.

[0114] In this example, the data set used in Example 2 was utilized. However, in this example, the data set was divided to improve the accuracy of estimating the adhesive strength.

[0115] An example of an improved method for estimating adhesive strength will be described below. The overall picture of the method for improving prediction accuracy implemented in this embodiment consists of three parts. One is a part for constructing an estimation model of adhesive strength based on adhesive A. The next part is a part for constructing a model for estimating adhesive strength when an arbitrary adhesive is used, using the estimation model of adhesive strength based on adhesive A. The last part is a part for predicting the unknown adhesive strength of an adhesive bond, using the estimation model of adhesive strength based on adhesive A and the model for estimating adhesive strength when an arbitrary adhesive is used.

[0116] The datasets used in the above method will be described below. The number of records in the dataset used in Example 2 is 147, which was separated into 74 records prepared in Example 1 and 73 records prepared in Example 2. The separated datasets are called dataset 5 and dataset 6, respectively.

[0117] Two new datasets were created from Dataset 5. The first dataset was created using information on adhesive strength, surface roughness, and line roughness from Dataset 5. This dataset is called Dataset 7. The data format of Dataset 7 was designed as shown in Figure 15. The other dataset was created using information on adhesive strength, adhesive breaking strength, breaking elongation, elastic modulus, adhesive viscosity, and adhesive layer thickness from Dataset 5. This dataset is called Dataset 8. The data format of Dataset 8 was designed as shown in Figure 16. Dataset 9 and Dataset 10 were created from Dataset 6 in a similar manner. The data format of Dataset 9 is as shown in Figure 15, and the data format of Dataset 10 was designed as shown in Figure 16.

[0118] The part of constructing an estimation model of adhesive strength based on adhesive A will be described below. The data processing procedure is shown in FIG. 17. First, a trained model of adhesive strength is constructed using data set 7 1701. As described above, data set 7 consists of surface line roughness measurement value 1702 and adhesive strength measurement value 1703. Data set 7 is input to an estimation model construction device 1704. Since data set 7 is all based on adhesive A, the estimation model constructed here is a model that predicts the adhesive strength when adhesive A is used for bonding. This is called an A-based adhesive strength estimation model. Data set 7 is passed to an A-based adhesive strength estimation model construction unit 1706 via an input interface 1705, and a trained model of adhesive strength is constructed. The constructed model is output via an output interface 1707, becoming a data set 7 trained model 1708.

[0119] Hereinafter, a part of constructing a model for estimating the adhesive strength when an arbitrary adhesive is used, using an estimation model of adhesive strength based on adhesive A, will be described. Here, a prediction model for other types of adhesives is constructed using dataset 9 and dataset 10. Here, records related to the adhesive to be predicted are extracted from dataset 9 and dataset 10, and are called dataset 9 (experimental data) 1724 and dataset 10 (experimental data) 1731, respectively. Dataset 10 (experimental data) 1731 includes mechanical property measurements 1732 and adhesive strength measurements 1733. The remaining data are called dataset 9 (teacher data) 1709 and dataset 10 (teacher data) 1716, respectively, and these are used to construct an adhesive strength trained model 1723. Dataset 10 (teacher data) 1716 includes mechanical property measurements 1717 and adhesive strength measurements 1718.

[0120] Data set 9 (teaching data) 1709 consists of measured values ​​1710 of surface line roughness, which are processed by an A-standard bond strength estimation device 1711. Specifically, data set 9 (teaching data) 1709 is input using an input interface 1712 and sent to an A-standard bond strength estimation unit 1713. The A-standard bond strength estimation unit 1713 is an estimation unit created using a data set 7 trained model 1708. This calculates the A-standard bond strength 1715 and outputs it from an output interface 1714.

[0121] The reference adhesive strength A 1715 is sent to an estimation model construction device 1719 and processed in an adhesive strength estimation model construction unit 1721 via an input interface 1720. Similarly, the data set 10 (teacher data) 1716 is also sent to the adhesive strength estimation model construction unit 1721 via the input interface 1720. In this manner, an adhesive strength learned model 1723 is constructed and output from an output interface 1722.

[0122] Hereinafter, a part for predicting an unknown adhesive strength of an adhesive bond using an estimation model of adhesive strength based on adhesive A and a model for estimating adhesive strength when an arbitrary adhesive is used will be described. Data set 9 (experimental data) 1724 is input to an A-standard adhesive strength estimation device 1726. Specifically, the data is processed in an A-standard adhesive strength estimation unit 1728 through an input interface 1727. The A-standard adhesive strength estimation unit 1728 is an estimation unit created using the Data set 7 trained model 1708, and has the same function as the A-standard adhesive strength estimation unit 1713. Here, an A-standard adhesive strength 1730 is calculated and output from an output interface 1729.

[0123] The output A-standard adhesive strength 1730 is sent to an adhesive strength estimation device 1734 and processed in an adhesive strength estimation unit 1736 via an input interface 1735. Similarly, the data set 10 (experimental data) 1731 is also sent to the adhesive strength estimation unit 1736 via the input interface 1735. The adhesive strength estimation unit 1736 is an estimation unit created using the adhesive strength learned model 1723. In this manner, an adhesive strength 1738 is calculated and output from an output interface 1737.

[0124] The prediction accuracy of the adhesive strength of the adhesive bond thus obtained was R = 0.82. This value exceeds the prediction accuracy R = 0.80 when φ = 1.0 in Example 2, and therefore it is recognized that the prediction method in this example is effective. [Explanation of symbols]

[0125] 101 Teacher Data 102 Experimental Data 103 Input Interface 104 Storage section 105 Estimation Model Construction Department 106 Adhesive strength estimation section 107 Output Interface 108 Adhesive strength of adhesive joints 109 Adhesion Strength Estimation System 401 Teacher Data 402 Mechanical property measurement data of solidified adhesive 403 Measurement data of surface roughness of adherend 404 Measurement data of adhesive strength of adhesive joints 405 Estimation Model Building Device 406 Input Interface 407 Estimation Model Construction Department 408 Output Interface 409 trained models 410 Experimental Data 411 Mechanical property data 412 Measurement data of surface roughness of adherend 417 Adhesive strength 601 Adhesive 602 Adherent 701 Adhesive 702 Adherent 801 Aluminum alloy plate 802 CFRP board 803 Adhesive 901 Line A 902 Line segment B 903 Line C 904 Line D 905 Line E 906 Line segment F 907 Adhesive area 908 Non-adhesive area 909 Adherent 1101 Aluminum plate 1102 CFRP board 1103 Adhesive layer 1107 Measurement point A 1108 Measurement point B 1109 Measurement point C 1110 Measurement point D 1111 Measurement point E 1112 Measurement point F 1113 Measurement point G 1114 Measurement point H 1115 Measuring point I 1116 Measurement point J 1117 Measurement point K 1118 Measurement point L 1701 Dataset 7 1702 Surface roughness measurement value 1703 Adhesive strength measurement value 1704 Estimation model construction device 1705 Input Interface 1706 A Standard Adhesive Strength Estimation Model Construction Section 1707 Output Interface 1708 Dataset 7 Trained Model 1709 Dataset 9 (Teaching Data) 1710 Surface roughness measurement 1711 A standard bond strength estimation device 1712 Input Interface 1713 A Standard Adhesive Strength Estimation Part 1714 Output Interface 1715 A standard adhesive strength 1716 Dataset 10 (training data) 1717 Mechanical property measurements 1718 Adhesive strength measurement 1719 Estimation model construction device 1720 Input Interface 1721 Bond strength estimation model construction section 1722 Output Interface 1723 Adhesive strength trained model 1724 Dataset 9 (Experimental Data) 1725 Surface roughness measurement 1726 A-standard bond strength estimation device 1727 Input Interface 1728 A Standard Adhesive Strength Estimation Part 1729 Output Interface 1730 A standard adhesive strength 1731 Dataset 10 (Experimental Data) 1732 Mechanical property measurements 1733 Adhesive strength measurement 1734 Bond strength estimation device 1735 Input Interface 1736 Adhesive strength estimation part 1737 Output Interface 1738 Adhesive strength

Claims

1. an input interface that accepts input of mechanical property values ​​of the solidified adhesive and information on the surface roughness of the adherend; A memory unit for storing a trained model trained with training data in which a combination of test results of mechanical property values ​​of test pieces of solidified adhesive and information on the surface roughness of an adherend is input, and adhesive strength is output; an adhesive strength estimation unit that estimates an adhesive strength using the trained model based on an input to the input interface; An adhesive force estimation system comprising:

2. In claim 1, The surface roughness information includes a root mean square slope.

3. In claim 1, The surface roughness information includes a ten-point average roughness.

4. In claim 1, The surface roughness information is A line roughness having a correlation coefficient with the root mean square slope of 0.4 or more; and Surface roughness with a correlation coefficient of 0.4 or more with the root mean square slope An adhesion strength estimation system including at least one of the indicators above.

5. In claim 1, An adhesive strength estimation system, wherein the mechanical property values ​​of the solidified adhesive include at least one of elastic modulus, breaking strength, and breaking elongation.

6. An adhesive strength estimation method, comprising: the adhesive strength estimation system according to claim 1 estimating an adhesive strength.

7. In claim 6, The adhesive strength estimation system is an adhesive strength estimation method in which an adhesive strength estimator for one type of adhesive based on surface roughness information of the adherend is used as part of an estimator for the adhesive strength of another type of adhesive.

8. A combination of test results of mechanical properties of a test piece of the solidified adhesive and information on the surface roughness of the adherend; Adhesive strength and A computer that stores data associated with the

Images