Method and system for predicting the durability of coatings under actual exposure.

Abstract

To provide a method for quickly and accurately predicting a secular change in properties of a coating film of an object at an arbitrary point by predicting durability of actual exposure of the coating film by using a machine learning method.SOLUTION: A predetermined artificial intelligence model includes, as an input, (a) data of coating material information of a coating material composition and property information data before actual exposure of a coating film obtained by applying the coating material composition, and (b) data of at least one of an exposure time of the coating film, environmental information of exposure of the coating film, an installation angle and an installation orientation of a coated object on which the coating film is formed, and an evaluation result of an accelerated test of the coating film, and outputs a result of durability of the actual exposure of the coating film. In the artificial intelligence model, the durability of the coating film in actual exposure is calculated and predicted by inputting the data of (a) and the data of (b) by a computer.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] This invention relates to a method and system for predicting the actual exposure durability of a coating film obtained by applying a coating composition. [Background technology]

[0002] Various paint compositions are applied to the walls and roofs of buildings such as houses and office buildings to maintain the quality and appearance of the walls and roofs under conditions of exposure to wind, rain, and direct sunlight. Such paint compositions are required to be durable against wind, rain, and sunlight.

[0003] Therefore, various methods have been proposed to predict the degree of aging changes in the properties of coating films. For example, Patent Document 1 proposes using an artificial intelligence model that uses paint information as an explanatory variable and the evaluation of an item as an objective variable, employing machine learning techniques to predict accelerated test results as one of the evaluations of an item. It is said that such machine learning techniques enable rapid prediction. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2023-021558 [Overview of the project] [Problems that the invention aims to solve]

[0005] However, accelerated testing involves exposing test specimens with a coating to conditions where the intensity of factors such as ultraviolet light, heat, and water—which are factors that cause aging changes in coating properties—is artificially amplified, and then evaluating the changes in their state. This often results in significant discrepancies with actual exposure results, and a challenge has been that it may not be sufficient as an indicator of aging deterioration.

[0006] Therefore, the present invention aims to provide a method for quickly and accurately predicting the aging changes in the properties of a coating on an object at any given location by using machine learning techniques to predict the durability of the coating under actual exposure. [Means for solving the problem]

[0007] The gist of the present invention is as follows: (1) A method for predicting the actual exposure durability of a coating film obtained by applying a coating composition, using machine learning techniques, A method for predicting the durability of a coating film under actual exposure, comprising a predetermined artificial intelligence model that takes as input (a) data on paint information of the paint composition and data on the properties of the coating film obtained by applying the paint composition before actual exposure, and (b) data on at least one of the following: exposure time of the coating film, environmental information of the exposure of the coating film, installation angle, installation orientation of the coated object on which the coating film is formed, and evaluation results of an accelerated test of the coating film, and outputs the result of the durability of the coating film under actual exposure, wherein the computer inputs (a) data on paint information of the paint composition and data on the properties of the coating film obtained by applying the paint composition before actual exposure, and (b) data on at least one of the following: exposure time, environmental information of the exposure, installation angle, installation orientation of the coated object, and evaluation results of an accelerated test, thereby calculating and predicting the durability of the coating film under actual exposure, the method comprising a durability prediction step for actual exposure.

[0008] (2) The input to the artificial intelligence model includes (a) data on paint information of the paint composition and data on the properties of the coating film obtained by applying the paint composition before actual exposure, and (b) at least the exposure time of the coating film. The method for predicting the durability of a coating film under actual exposure according to (1), wherein the durability prediction step under actual exposure includes inputting (a) data on the coating information of the coating composition and data on the properties of the coating film obtained by applying the coating composition before actual exposure, and (b) at least the exposure time.

[0009] (3) The input to the artificial intelligence model includes (a) data on paint information of the paint composition and data on the properties of the coating film obtained by applying the paint composition before actual exposure, and (b) data on at least the exposure time of the coating film and environmental information of the exposure of the coating film. The method for predicting the durability of a coating film under actual exposure according to (1) or (2), wherein the durability prediction step under actual exposure includes inputting (a) data on the coating information of the coating composition and data on the properties of the coating film obtained by applying the coating composition before actual exposure, and (b) data on at least the exposure time and the environmental information of the exposure.

[0010] (4) A method for predicting the durability of a coating film under actual exposure according to any one of (1) to (3), wherein the coating information data includes at least information regarding the formulation of the coating composition.

[0011] (5) A method for predicting the durability of a coating film under actual exposure according to any one of (1) to (4) above, wherein the environmental information of the exposure of the coating film includes at least one of the following: latitude, solar radiation, ultraviolet radiation, temperature, precipitation, relative humidity, wind speed, amount of sea salt particles attached, and amount of sulfur oxides attached.

[0012] (6) A system for predicting the actual exposure durability of a coating film obtained by applying a coating composition, using machine learning techniques, (a) Data on the paint information of the paint composition and data on the properties of the paint film before actual exposure obtained by applying the paint composition, and (b) At least any one of the data on the exposure time of the paint film, the environmental information of the exposure of the paint film, the installation angle of the coated object on which the paint film is formed, the installation orientation, and the evaluation results of the accelerated test of the paint film, are included as inputs, and in a predetermined artificial intelligence model that outputs the results of the durability of the actual exposure of the paint film, by a computer, (a) Data on the paint information of the paint composition and data on the properties of the paint film before actual exposure obtained by applying the paint composition, and (b) At least any one of the data on the exposure time, the environmental information of the exposure, the installation angle of the coated object, the installation orientation, and the evaluation results of the accelerated test, are input, and the durability of the actual exposure of the paint film is calculated and predicted. A prediction system for the durability of the actual exposure of a paint film, characterized by comprising a durability prediction unit for actual exposure.

Advantages of the Invention

[0013] According to the present invention, by using a machine learning method to predict the durability of the actual exposure of a paint film, it is possible to provide a method for quickly and accurately predicting the aging change of the properties of the paint film possessed by an object at any location.

Brief Description of the Drawings

[0014] [Figure 1] It is a flowchart of a method for predicting the durability of the actual exposure of a paint film according to an embodiment of the present invention.

Modes for Carrying Out the Invention

[0015] Hereinafter, embodiments of the present invention will be illustrated and described in detail with reference to the drawings.

[0016] <Method for Predicting the Durability of the Actual Exposure of a Paint Film> Figure 1 is a flowchart of a method for predicting the durability of a coating film against actual exposure according to one embodiment of the present invention. Hereinafter, with reference to Figure 1, an embodiment of a method for predicting the durability of a coating film against actual exposure obtained by applying a paint composition, using machine learning techniques, will be described as an example. Note that the method for predicting the durability of a coating film against actual exposure according to this embodiment can, for example, be performed using the coating film durability prediction system according to one embodiment of the present invention, which will be described later.

[0017] As shown in Figure 1, in this embodiment, first, relational data is prepared showing the relationship between (a) paint information data of the paint composition and property information data of the coating film obtained by applying the paint composition before actual exposure, and (b) data of at least one of the following: exposure time of the coating film, environmental information of the exposure of the coating film, installation angle and orientation of the coated object on which the coating film is formed, and evaluation results of the accelerated test of the coating film, and the durability results of the coating film after actual exposure (step S101).

[0018] The "paint information data of the paint composition" in (a) above refers to the paint information data of the paint composition in its initial state. Here, it is preferable that the "paint information data of the paint composition" in (a) above includes at least information regarding the formulation of the paint composition. Specifically, this information may be the names of the raw materials (also called compound names) of the paint composition and their proportions. In place of the raw material names, or in addition to the raw material names, data on product names (product numbers) may also be used. Furthermore, the color (L) of the resulting paint film may also be used. * value, a * value, b * The data can also include values ​​(spectral reflectance spectrum) and coating amount. Examples of shape include the shape of the colorant (spherical, flaky, fibrous, etc.), average primary particle diameter, average secondary particle diameter, average dispersed particle diameter, particle size distribution, aspect ratio, and thickness. Examples of chemical properties include molecular weight, molecular weight distribution, discoloration temperature, and reactivity.

[0019] Furthermore, the "pre-exposure property information data" in (a) above may be data on one or more of the color and gloss of the obtained coating film before actual exposure.

[0020] The color is, for example, L * a * b * L value in the color space * value, a * value, b * value (JIS Z_8781-4 (2013)), X-Y-Z color system, R-G-B color system, Yxy color system, Hunter L-a-b color system, L * -C * -h * color systems, and those based on color systems such as the Munsell color system can be used. The color can be measured using known color measurement methods. As an example, when using the CM-512m3 commercially available from Konica Minolta Co., Ltd. and setting the light-receiving part perpendicular to the coating film as 0°, the L * value, a * value, b * values can be measured. Alternatively, it can be measured using the X-Rite MA68II (manufactured by X-Rite). The measurement angle can be appropriately adjusted according to the purpose or the device to be used. Any other indicator can be used. Further, for example, reflection spectrum data, and any indicator such as an indicator using the reflection spectrum intensity every 5 nm from 380 nm to 780 nm as the color can also be used.

[0021] The gloss is not particularly limited, but gloss can be used as an indicator. The gloss can be measured using known gloss measurement methods. As an example, the 60° gloss of the coating film formed on the test plate can be measured in accordance with JIS K 5600-4-7 (specular gloss) using a specular gloss meter (gloss meter VG-7000 (manufactured by Nippon Denshoku Industries Co., Ltd.)).

[0022] The "pre-exposure property information data" may include data on the smoothness of the coating film before actual exposure. Smoothness is preferably measured using wave scan values ​​as an indicator. The wave scan values ​​are preferably one or more of the following: du (wavelength 0.1 mm or less), Wa (wavelength 0.1 to 0.3 mm), Wb (wavelength 0.3 to 1.0 mm), Wc (wavelength 1.0 to 3.0 mm), Wd (wavelength 3.0 to 10.0 mm), We (wavelength 10.0 to 30.0 mm), Lw (wavelength 1.2 to 12 mm), and Sw (wavelength 0.3 to 1.2 mm). A smaller wave scan value indicates less surface irregularity at that wavelength, resulting in better appearance quality of the coating film. These indicators may also be calculated using digital analysis with 3D scanning. Flip-flop properties can be measured using a colorimeter or similar device. Coating abnormalities specifically include unevenness, blistering, cracking, sagging, pinholes, and frame patterns. These can also be measured using known methods. "Property information data before actual exposure" may also include other factors such as opacity, flip-flop properties, micro-luminosity, visual appearance, electrical resistance, contact angle, staining properties, solar reflectance, ultraviolet transmittance, and viscoelasticity.

[0023] Furthermore, the "property information data of the coating film before actual exposure" in (a) above may also include information on the thickness of the coating film. If the coating film consists of multiple layers, it may also include information on the formulation and thickness of each layer. Additionally, it may include information on the substrate to which the coating film is applied.

[0024] In (b) above, "exposure time of the coating" refers to information about the time (period) during which the coating is exposed to the environment. For example, it may be information about the start date (year, month, day) of the exposure and the date (year, month, day) at a point in time thereafter, or it may be the difference in time (period) calculated from these.

[0025] In (b) above, the “environmental information for the exposure of the coating” preferably includes at least one of the following: latitude, solar radiation, ultraviolet radiation, temperature, precipitation, relative humidity, wind speed, amount of sea salt particles attached, and amount of sulfur oxides attached.

[0026] "Latitude" can be expressed in any unit system, but one example uses "°". "Solar radiation" is expressed in SRAD (J / cm²). 2 (Periodic average or cumulative over time) etc. can be used. "Ultraviolet radiation amount" is UVA / UVB / UV (J / cm 2 )(period average or period cumulative), etc. can be used. "Temperature" can be in any unit system, but in one example "℃" (period average or period cumulative) can be used. "Precipitation" can be in any unit system, but in one example "mm" (period average or period cumulative) can be used. "Relative humidity" can be in any unit system, but in one example "%" (period average or period cumulative) can be used. "Wind speed" can be in any unit system, but in one example "m / s" (period average or period cumulative) can be used. "Sea salt particle adhesion amount (chloride)" can be in any unit system, but in one example "mgNaCl / (m 2 The unit "mgSO2 / (m³)" (period average or period cumulative) can be used. "Sulfur oxide deposit amount (SO2, etc.)" can be expressed in any unit system, but one example is "mgSO2 / (m³)". 2 You can use "day)" (average or cumulative over a period).

[0027] Regarding the "installation angle and installation orientation of the coated object on which the coating film is formed" in (b) above, the "installation angle" may be an angle based on any direction, and as an example, the unit can be "°". The "installation orientation" may be an orientation based on any direction, and as an example, the unit can be "°". The "installation angle and installation orientation of the coated object on which the coating film is formed" may be the installation angle and / or installation orientation of the coated object on which the coating film is formed.

[0028] In (b) above, the "evaluation results of the accelerated coating test" can be expressed using SWOM, QUV-A, QUV-B, SUV, Xe, SST, CCT, SDT, hot water resistance test, etc. as evaluation (test) methods for the accelerated test. Here, "SWOM" is the sunshine carbon arc lamp weathering test, "QUV" is a test using the ultraviolet fluorescent lamp weather tester QUV (ASTM G154, JIS K 7350-3), "SUV" is an abbreviation for Super UV and means a weathering test using a metal halide lamp, and "Xe" is a test using a xenon arc. Furthermore, "SST" is an abbreviation for salt spray test, meaning a durability test in which sprayed salt water is constantly adhering to the test piece; "CCT" is an abbreviation for salt combined cycle test, meaning a durability test that repeats cycle tests such as salt spray, drying, and wetting; "SDT" is an abbreviation for hot salt water test, meaning a durability test in which the test piece is immersed in heated hot salt water; and "hot water resistance test" means a durability test in which the test piece is immersed in hot water maintained at a constant temperature. The "evaluation results of the accelerated coating test" can be deterioration information at any test time (one or more fluctuation values ​​from among color difference, gloss, rust width, blister width, chalking level, peeling area, and film thickness).

[0029] Furthermore, the "durability of the coating under actual exposure" can be evaluated using one or more of the following: color difference, gloss, rust width, blister width, chalking level, peeling area, and film thickness.

[0030] Furthermore, it is preferable that the aforementioned related data be updated constantly, in a timely manner, or periodically in order to enable more accurate predictions.

[0031] As shown in Figure 1, in this embodiment, the following are then input into the computer (step S102): (a) data on paint information of the paint composition and data on the properties of the coating film obtained by applying the paint composition before actual exposure, and (b) relational data showing the relationship between at least one of the following data: exposure time of the coating film, environmental information of the exposure of the coating film, installation angle and orientation of the coated object on which the coating film is formed, and evaluation results of the accelerated test of the coating film, and the durability results of the coating film after actual exposure.

[0032] Next, as shown in Figure 1, a predetermined artificial intelligence model is created using machine learning by a computer, taking as input (a) data on paint information of the paint composition and data on the properties of the coating obtained by applying the paint composition before actual exposure, and (b) data on at least one of the following: exposure time of the coating, environmental information of the exposure of the coating, installation angle and orientation of the coated object on which the coating was formed, and evaluation results of the accelerated test of the coating, and outputting the result of the durability of the coating under actual exposure (step S103). In this embodiment, an artificial intelligence model is created in this step, but it is not necessarily required to create an artificial intelligence model, and the created artificial intelligence model can also be obtained from an external source. For example, the artificial intelligence model can be obtained by the computer's communication unit, or it can be obtained by a human through transfer or other means.

[0033] The machine learning algorithm can be any known algorithm, such as decision trees, linear regression, partial least squares regression, Lasso regression, ridge regression, polynomial regression, Gaussian process regression, support vector machines, random forests, gradient boosting, K-nearest neighbors, neural networks, Bayesian estimation, or prediction algorithms using ensemble learning of these. Machine learning is particularly preferably performed using a prediction algorithm based on a neural network. The neural network may be a convolutional neural network or a deep learning network with 3 to several hundred layers while performing dropout. Furthermore, machine learning frameworks such as TensorFlow, Keras, Caffe, PyTorch, Chainer, and Scikit-learn may be used to create the artificial intelligence.

[0034] Here, machine learning can be supervised learning (including reinforcement learning). In this embodiment, step S103 is performed by a computer using the relational data input in step S102 as training data (supervised data in this example) to perform machine learning (supervised learning).

[0035] Here, the algorithm is preferably determined by the following method. That is, multiple artificial intelligence models are created using various machine learning algorithms, and the accuracy of each artificial intelligence model is confirmed by comparing the predicted output with the answer using data for which the answer is known in advance. Then, the optimal hyperparameters are determined for the algorithm with the highest accuracy among the multiple algorithms, for example, by creating a model using cross-validation or by using Bayesian optimization. By retraining all data with the determined hyperparameters, an artificial intelligence model can be created that takes as input (a) data on paint information of the paint composition and data on the properties of the paint film obtained by applying the paint composition before actual exposure, and (b) data on at least one of the following: exposure time of the paint film, environmental information of the exposure of the paint film, installation angle, installation orientation of the coated object on which the paint film is formed, and evaluation results of the accelerated test of the paint film, and outputs the result of the durability of the paint film under actual exposure.

[0036] In this embodiment, preprocessing is performed on the prepared relational data. Here, processing is performed on the prepared relational data, such as normalizing / standardizing the data, applying predetermined operations such as taking logarithms or multiplying by powers (exponentiation processing), generating new data, or performing data filtering to remove inappropriate data. For example, data with large differences between measured and predicted values ​​of paint properties can be examined, and data suspected of being input errors or description errors can be deleted. As described above, various machine learning algorithms can be used in this embodiment, but the preprocessing can be applied to multiple or all algorithms in common (for example, removing outlier data that would be judged as inappropriate by any machine learning algorithm), or it can be applied exclusively to the machine learning algorithm actually used (for example, removing data that is likely to cause errors in a particular algorithm).

[0037] In the pretreatment stage, it is preferable to add zero-point relationship data (zero-point addition) that shows the relationship between the durability of the coating under actual exposure and the pre-exposure durability of the coating, where the exposure time is 0, and the change in the actual exposure time remains the same (zero change). This can further improve the accuracy of the prediction. Here, instead of zero, a value close to zero may also be used.

[0038] As a preprocessing step, inferred data can be added between data points. For example, a approximation line or curve can be derived from the data, and points on that approximation line or curve can be used as inferred data. This can increase the amount of data and further improve the accuracy of predictions. Alternatively, past data points can be connected with linear or curved lines, and any number of data points can be amplified along those lines.

[0039] Such preprocessing and normalization enable more accurate predictions of real-world exposure endurance results in machine learning, as described later. However, preprocessing is not mandatory in this disclosure and can be omitted. A predetermined artificial intelligence model can be created using the preprocessed data in the same manner as described above.

[0040] Next, in this embodiment, the computer calculates and predicts the actual durability of the coating film under actual exposure by inputting (a) data on paint information of the paint composition and data on the properties of the coating film obtained by applying the paint composition before actual exposure, and (b) data on at least one of the following: exposure time, exposure environment information, installation angle of the coating object, installation orientation, and evaluation results of the accelerated test (Step S104 (Actual Exposure Durability Prediction Step)).

[0041] The predicted value of the durability result of the coating under actual exposure may be a single value or may consist of multiple candidate groups. If the calculated predicted value consists of multiple candidate groups, it is preferable to further include a step of appropriately selecting a single value from among them using predetermined criteria.

[0042] According to the method for predicting the durability of a coating film under actual exposure in this embodiment, by inputting data using an artificial intelligence model, a predicted value for the durability of the coating film under actual exposure can be obtained, thereby achieving efficiency. Furthermore, since the relationship between the paint information data and the property information data of the coating film before actual exposure, and the exposure conditions and accelerated test results in (b) and the durability results of "actual exposure" is learned, the problem of discrepancies between accelerated tests and actual exposure, which is the aforementioned issue, can be avoided. As a result, as shown in the embodiments described later, it becomes possible to predict with high accuracy the durability results of the coating film under actual exposure. This makes it possible to predict in advance whether or not there is a risk of coating film deterioration, or the degree of that risk, and to take measures to avoid that risk, for example, by adjusting the formulation. As described above, according to the method for predicting the durability of a coating under actual exposure in this embodiment, by using machine learning techniques to predict the durability of the coating under actual exposure, it is possible to quickly and accurately predict the changes in the properties of the coating on an object at any given location over time.

[0043] Furthermore, according to the method for predicting the durability of a coating film under actual exposure in this embodiment, it is possible to predict the durability of a coating film produced using a paint composition with a formulation that has not been subjected to exposure testing.

[0044] The input to the artificial intelligence model includes (a) data on paint information of the paint composition and data on the properties of the coating film obtained by applying the paint composition before actual exposure, and (b) at least the exposure time of the coating film. In the actual exposure durability prediction step (step S104), it is preferable to input (a) data on paint information of the paint composition and data on the properties of the coating film obtained by applying the paint composition before actual exposure, and (b) at least the exposure time. This is because including at least the exposure time of the coating film in the input makes it possible to predict the deterioration behavior of the coating film over time and to obtain the prediction error.

[0045] Furthermore, in (b) above, it is also preferable to include the installation angle and / or orientation of the coated object on which the coating film is formed. This is because it becomes possible to predict the deterioration behavior of the coating film at various installation angles and / or orientations, as well as the error in such predictions.

[0046] The input to the artificial intelligence model includes (a) data on paint information of the paint composition and data on the properties of the coating film obtained by applying the paint composition before actual exposure, and (b) data on at least the exposure time of the coating film and environmental information of the exposure of the coating film. In the actual exposure durability prediction step (step S104), it is preferable to input (a) data on paint information of the paint composition and data on the properties of the coating film obtained by applying the paint composition before actual exposure, and (b) data on at least the exposure time and environmental information of the exposure. This is because, by including the exposure time of the coating film and environmental information of the exposure of the coating film in (b), it becomes possible to predict the time-series deterioration behavior of the coating film that takes environmental information into account, and to obtain the prediction error thereof.

[0047] <System for predicting the durability of coatings under actual exposure> A system for predicting the durability of a coating film under actual exposure according to one embodiment of the present invention is a system for predicting the durability of a coating film obtained by applying a paint composition under actual exposure using machine learning techniques. The system of this embodiment includes an actual exposure durability prediction unit that takes as input (a) paint information data of the paint composition and property information data of the coating film obtained by applying the paint composition before actual exposure, and (b) data of at least one of the following: exposure time of the coating film, environmental information of the exposure of the coating film, installation angle, installation orientation of the coated object on which the coating film is formed, and evaluation results of an accelerated test of the coating film, and outputs the result of the durability of the coating film under actual exposure, and uses a predetermined artificial intelligence model to calculate and predict the durability of the coating film under actual exposure by inputting (a) paint information data of the paint composition and property information data of the coating film obtained by applying the paint composition before actual exposure, and (b) data of at least one of the following: exposure time, environmental information of the exposure, installation angle, installation orientation of the coated object, and evaluation results of an accelerated test. The durability prediction unit can be any known processor. The above (a) and (b) concerning the system are the same as those described for (a) and (b) in the embodiment of the previous method, so a further explanation will be omitted.

[0048] The paint film durability prediction system of this embodiment also achieves efficiency by obtaining a predicted value for the paint film's durability under actual exposure by inputting data using an artificial intelligence model. Furthermore, because the relationship between the paint information data and the paint film's properties data before actual exposure, and the exposure conditions and accelerated test results in (b) and the durability results under "actual exposure" is learned, the problem of discrepancies between the accelerated test and actual exposure can be avoided. As a result, as shown in the embodiments described later, it becomes possible to make highly accurate predictions regarding the durability results under actual exposure of the paint film. This makes it possible to predict in advance whether or not there is a risk of paint film deterioration, or the degree of that risk, and to take measures to avoid that risk, for example, by adjusting the formulation. As described above, the coating film durability prediction system of this embodiment can predict the aging changes in the properties of a coating film on an object at any given location quickly and accurately by using machine learning techniques to predict the durability of the coating film under actual exposure. The following describes embodiments of the present invention, but the present invention is not limited to the following embodiments. [Examples]

[0049] (Method for preparing paint composition) Method for preparing polyester resin (A) In a reaction vessel equipped with a thermometer, condenser, and stirrer, 155 parts by mass of neopentyl glycol, 222 parts by mass of 1,6-hexanediol, 50 parts by mass of trimethylolpropane, 441 parts by mass of isophthalic acid, 132 parts by mass of adipic acid, and 26 parts by mass of xylene were mixed. The mixture was gradually heated to 230°C under a nitrogen atmosphere, and the esterification reaction was carried out while distilling off the generated water until the amount of dewatered water reached 126 parts by mass and the reactants reached a predetermined viscosity. After that, the temperature of the reaction vessel was lowered to 50°C, and 90 parts by mass of xylene, 379 parts by mass of T-SOL 150, and 87 parts by mass of propylene glycol monomethyl ether acetate were mixed to prepare a coating resin (A) (solid content concentration: 60% by mass, number average molecular weight: 3,000, hydroxyl value: 55 mgKOH / g).

[0050] Method for preparing coloring agents Method for preparing yellow coloring agent 1 59 parts by mass of polyester resin (A), 16 parts by mass of crosslinking agent 1, 6 parts by mass of organic solvent 1, and 1 part by mass of organic solvent 2 were uniformly mixed using a disperser, and dispersed using a sand mill (dispersion medium: glass beads) until the maximum particle size of the pigment coarse particles was 10 μm or less to prepare yellow colorant 1.

[0051] The colorant was prepared in the same manner as the yellow colorant Y1, except that the type and / or amount of each component was changed as shown in Table 1.

[0052] [Table 1] TIFF0007856320000002.tif93170

[0053] Method for preparing paint composition A 173 parts by mass of white colorant, 11 parts by mass of blue colorant, 111 parts by mass of black colorant, 14 parts by mass of yellow colorant, 18 parts by mass of aggregate, and 3 parts by mass of surface modifier were added and uniformly mixed using a disperser to prepare paint composition A.

[0054] The details of the materials used in the preparation of paint composition A are as follows: • Acrylic resin (A): Paraloid B44 (manufactured by Rohm & Haas), solids content: 100% by mass • Fluororesin (A): KYNAR500 (manufactured by ARKEMA), solids content: 100% by mass Crosslinking agent 1: Super Beccamine L-155-70 (manufactured by DIC Corporation), butylated melamine resin; solid content concentration: 70% by mass Aggregate 1: NipSeal E-200A (manufactured by Tosoh Silica Co., Ltd.), silica microparticles; average particle size: 3 μm ·Aggregate 2: GASIL HP395 (manufactured by INEOS SILICAS), synthetic silica; average particle size: 15 μm Aggregate 3: Silysia 435 (manufactured by Fuji Silysia Chemical Co., Ltd.), silicon dioxide; average particle size: 4 μm • Surface modifier: Disparon OX-70 (manufactured by Kusumoto Chemical Co., Ltd.), acrylic-based surface modifier; solid content concentration: 30% by mass • White pigment 1: Typake CR-97 (manufactured by Ishihara Sangyo Co., Ltd.), titanium dioxide • White pigment 2: TI-PURE R-706 (manufactured by DuPont), titanium dioxide • Black pigment 1: Black 6350 (manufactured by Asahi Chemical Industries Co., Ltd.), chromium iron oxide • Black pigment 2: Mitsubishi Carbon Black MA-100 (manufactured by Mitsubishi Chemical Corporation), Carbon Black • Yellow pigment 1: TAROX Synthetic Iron Oxide HY-100 (manufactured by Titanium Industries Co., Ltd.), Yellow Iron Oxide • Yellow pigment 2: TAROX synthetic iron oxide LL-XLO (manufactured by Titanium Industries Co., Ltd.), yellow iron oxide • Red pigment 1: TODA COLOR KN-V (manufactured by Toda Pigment Co., Ltd.), iron oxide • Red pigment 2: TODA COLOR 140ED (manufactured by Toda Kogyo Co., Ltd.), iron oxide • Organic solvent 1: T-SOL 150 (manufactured by JXTG Energy Corporation), aromatic solvent • Organic solvent 2: Cyclohexanone (manufactured by Shoei Chemical Industry Co., Ltd.), ketone-based solvent • Organic solvent 3: Isophorone (ARKEMA) ketone-based solvent

[0055] Paint compositions B to F were prepared in the same manner as paint composition A, except that the types and / or amounts of each component were changed as shown in Table 2.

[0056] [Table 2] TIFF0007856320000004.tif150170

[0057] (Method for preparing a coating film of paint composition A) After alkaline degreasing a 0.4 mm thick hot-dip galvanized steel sheet, a chemical conversion treatment was performed by applying the phosphate treatment agent Surfcoat EC2310 (manufactured by Nippon Paint Surf Chemicals Co., Ltd.) to the front and back surfaces of the steel sheet, and then it was dried. As an undercoat, Flexcoat 612WW Youprimer (manufactured by Nippon Paint Industrial Coatings Co., Ltd., a polyester resin-based primer) was applied to the surface of the resulting steel sheet using a bar coater to a dry film thickness of 5 μm, and then baked for 25 seconds under conditions that resulted in a maximum material temperature of 215°C to form an undercoat film. Subsequently, the paint composition A obtained above was applied using a bar coater to a dry film thickness of 15 μm, and then baked for 25 seconds under conditions that resulted in a maximum material temperature of 220°C to form a topcoat film, thereby obtaining a paint film of the paint composition.

[0058] (Method for preparing coating films of paint compositions B, C, and D) Coating films of coating compositions B, C, and D were obtained in the same manner as in the case of coating composition A.

[0059] (Method for preparing a coating film of paint composition E) After alkaline degreasing a 0.4 mm thick hot-dip galvanized steel sheet, a chemical conversion treatment was performed by applying the phosphate treatment agent Surfcoat EC2310 (manufactured by Nippon Paint Surf Chemicals Co., Ltd.) to the front and back surfaces of the steel sheet, and then it was dried. Next, as a primer, Fine Tough G Primer (epoxy resin primer: manufactured by Nippon Paint Industrial Coatings Co., Ltd.) was applied using a bar coater to a dry film thickness of 5 μm, and then baked for 60 seconds under conditions that resulted in a maximum material temperature of 210°C to form the primer film. Subsequently, the paint composition E obtained above was applied using a bar coater to a dry film thickness of 18 μm, and then baked for 60 seconds under conditions that resulted in a maximum material temperature of 250°C to form the surface topcoat film, obtaining the paint film of paint composition E.

[0060] (Method for preparing a coating film of paint composition F) A coating film of coating composition F was obtained in the same manner as in the case of coating composition E.

[0061] (Method for measuring color difference) For the various coating films obtained above, the color was measured using a spectrophotometer colorometer SM-T45 (manufactured by Suga Test Instruments Co., Ltd.) (L before the test). * a * , b * The values ​​and the color difference (ΔE) before and after the test were measured.

[0062] (Method for measuring gloss level) For the various coating films obtained as described above, the 60° gloss was measured using a gloss meter VG7000 (manufactured by Nippon Denshoku Industries Co., Ltd.) in accordance with JIS K 5600-4-7 (specular gloss), and the gloss retention rate before and after the test was calculated using the following formula. Gloss retention rate (%) = 60° gloss after testing / 60° gloss before testing × 100

[0063] (Accelerated weathering test: SWOM) The various test panels obtained as described above were subjected to accelerated weathering tests for a maximum of 5,000 hours using a Sunshine Weather Meter S80 (manufactured by Suga Test Instruments Co., Ltd.), which is a sunshine carbon arc lamp type accelerated weathering tester specified in JIS B 7753. The operating conditions were as follows: Irradiance: 255W / m 2 Black panel temperature: 63℃ Water injection time: 18 minutes out of 120 minutes

[0064] (Exposure test) The test panels were erected for 10 years at various latitudes, including Miyako Island and Iioka, facing south (equator plane) at an angle of 20° to the horizontal plane, in accordance with JIS K 5600-7-6. Exposure location 1: Nippon Paint Miyakojima Weathering Center: 3742 Karimata, Hirara, Miyakojima City, Okinawa Prefecture (24°51′N) Exposure location 2: Nippon Paint Iioka Weathering Center: 1351-3 Nagai, Asahi City, Chiba Prefecture (35°69′N)

[0065] 854 paint types (pre-exposure property data; film thickness: 6-28 μm, L) prepared by the same method as the example of the pre-adjustment paint composition described above. * Value: 7~91, a * Values: -30 to 48, b * For values ​​(-49 to 41, 60° glossiness: 1 to 93), exposure tests were conducted at six locations, including the pre-exposure site, and a total of 20,381 data points were prepared for exposure times ranging from 0 to 240 months.

[0066] Examples 1-4, Comparative Example 1 Table 3 shows the combinations of variables used as explanatory variables from (a) paint information and pre-exposure property information data, and (b) exposure time, exposure environment information, installation angle of the coated object, installation orientation, and evaluation results of the accelerated weathering test (SWOM). Paint information includes the formulation of the paint composition, and pre-exposure property information includes film thickness and initial L * Initial a * , initial b *Initial gloss was measured, and exposure environment information included latitude, UV radiation level, and precipitation. The dependent variables were the color difference (ΔE) and gloss after the exposure test. Gaussian regression was used for training. The program obtained through the aforementioned training was used to input the input information shown in Table 3, and predictions were made for untrained data. The prediction results were then compared with the exposure results for the untrained data. For accuracy verification, the Leave-Group-out method was used to ensure that untrained data was not included in the training data. The root mean square error (RMSE) was calculated to compare accuracy (the degree of difference between the predicted value and the exposure result). The results are shown in Table 3. Here, the degree of difference between the predicted and actual values ​​was evaluated using the Root Mean Squared Error (RMSE). The Root Mean Squared Error is an indicator that measures how much the predicted and actual values ​​deviate from each other; a value closer to zero indicates a smaller difference. In Table 3, "○" indicates that the variable was used as an explanatory or dependent variable, and "×" indicates that it was not used as an explanatory or dependent variable.

[0067] Comparative Example 2 Based on the results of accelerated weathering tests (SWOM), the RMSE was calculated assuming a test duration of 250 hours was equivalent to one year of actual exposure, and the prediction accuracy was compared. The evaluation results are shown in Table 3 below.

[0068] [Table 3]

[0069] Example 1 is an example in which all of the above explanatory variables are used. Example 2 is an example in which the accelerated weathering evaluation results are not included as explanatory variables. Example 3 is an example in which the exposure environmental information is not included as explanatory variables. Example 4 is an example in which only the exposure time is included as an explanatory variable. In all examples, the RMSE values ​​for color difference and gloss are small, and it can be seen that the aging changes in the properties of the coating film can be predicted with high accuracy. On the other hand, Comparative Example 1 was an example that did not include all of the explanatory variables (b), and the RMSE values ​​for color difference and gloss were large, so the accuracy of predicting the aging changes in the properties of the coating film was not sufficiently satisfactory. Furthermore, Comparative Example 2 is an example of predicting exposure test results from accelerated weathering (SWOM) evaluation results, and the RMSE values ​​for color difference and gloss were large, so the accuracy of predicting the aging changes in the properties of the coating film was not entirely satisfactory. [Industrial applicability]

[0070] According to the present invention, by using machine learning techniques to predict the durability of a coating under actual exposure, it is possible to provide a method for quickly and accurately predicting the aging changes in the properties of a coating on an object at any given location. For example, it can also be applied to predicting the degradation of plastics.

Claims

1. A method for predicting the actual exposure durability of a coating film obtained by applying a coating composition, using machine learning techniques, A method for predicting the durability of a coating film under actual exposure, comprising a predetermined artificial intelligence model that takes as input (a) data on paint information of the paint composition and data on the properties of the coating film obtained by applying the paint composition before actual exposure, and (b) data on at least one of the following: exposure time of the coating film, environmental information of the exposure of the coating film, installation angle, installation orientation of the coated object on which the coating film is formed, and evaluation results of an accelerated test of the coating film, and outputs the result of the durability of the coating film under actual exposure, wherein the computer inputs (a) data on paint information of the paint composition and data on the properties of the coating film obtained by applying the paint composition before actual exposure, and (b) data on at least one of the following: exposure time, environmental information of the exposure, installation angle, installation orientation of the coated object, and evaluation results of an accelerated test, thereby calculating and predicting the durability of the coating film under actual exposure, the method comprising a durability prediction step for actual exposure.

2. The input to the artificial intelligence model includes (a) data on paint information of the paint composition and data on the properties of the coating film obtained by applying the paint composition before actual exposure, and (b) at least the exposure time of the coating film. The method for predicting the durability of a coating film under actual exposure according to claim 1, wherein in the step of predicting durability under actual exposure, (a) data on the coating information of the coating composition and data on the properties of the coating film obtained by applying the coating composition before actual exposure, and (b) at least the exposure time are input.

3. The input to the artificial intelligence model includes (a) data on paint information of the paint composition and data on the properties of the coating film obtained by applying the paint composition before actual exposure, and (b) data on at least the exposure time of the coating film and environmental information of the exposure of the coating film. The method for predicting the durability of a coating film under actual exposure according to claim 1 or 2, wherein in the step of predicting the durability under actual exposure, (a) data on the coating information of the coating composition and data on the properties of the coating film obtained by applying the coating composition before actual exposure, and (b) data on at least the exposure time and the environmental information of the exposure.

4. The method for predicting the durability of a coating film under actual exposure according to claim 1 or 2, wherein the data of the coating information includes at least information regarding the formulation of the coating composition.

5. The method for predicting the durability of a coating film under actual exposure according to claim 1 or 2, wherein the environmental information of the exposure of the coating film includes at least one of the following: latitude, solar radiation, ultraviolet radiation, temperature, precipitation, relative humidity, wind speed, amount of sea salt particles attached, and amount of sulfur oxides attached.

6. A system for predicting the actual exposure durability of a coating film obtained by applying a paint composition, using machine learning techniques, A system for predicting the durability of a coating film under actual exposure, comprising a predetermined artificial intelligence model that takes as input (a) data on paint information of the paint composition and data on the properties of the coating film obtained by applying the paint composition before actual exposure, and (b) data on at least one of the following: exposure time of the coating film, environmental information of the exposure of the coating film, installation angle, installation orientation of the coated object on which the coating film is formed, and evaluation results of an accelerated test of the coating film, and outputs the result of the durability of the coating film under actual exposure, wherein the computer calculates and predicts the durability of the coating film under actual exposure by inputting (a) data on paint information of the paint composition and data on the properties of the coating film obtained by applying the paint composition before actual exposure, and (b) data on at least one of the following: exposure time, environmental information of the exposure, installation angle, installation orientation of the coated object, and evaluation results of an accelerated test, and further comprising a durability prediction unit for actual exposure of a coating film.

Images