Coating weather resistance prediction method based on multi-task supervised neural network

Abstract

The invention discloses a coating weather resistance prediction method based on a multi-task supervised neural network. The method comprises the following steps: acquiring a mass ratio and a coating thickness of a three-primary-color pigment to an acrylic acid substrate of a target coating; constructing and training a first neural network for predicting the sunlight reflectivity of the coating, inputting the mass ratio, and outputting the sunlight reflectivity of the target coating; constructing and training a second neural network for predicting the weather resistance of the coating on the basis of a multi-task supervised neural network, inputting the sunlight reflectivity and the coating thickness of the target coating, and outputting the ultraviolet absorption amount and the surface temperature variation of the target coating as intermediate parameters, and outputting the weather resistance comprehensive index as a target coating weather resistance prediction result. According to the method, the intelligent neural network model is constructed, so that rapid prediction of the weather resistance of the coating is realized, the dependence on a traditional laboratory test is remarkably reduced, and the research and development cycle and the cost investment are reduced.

Description

Technical Field

[0001] This invention relates to the field of green building materials technology, and in particular to a method for predicting the weather resistance of coatings based on a multi-task supervised neural network. Background Technology

[0002] Architectural exterior wall coatings play a vital role in modern architecture. On one hand, they significantly enhance the aesthetics of buildings by offering a variety of colors, glosses, and textures. On the other hand, they effectively protect building exteriors from external factors such as ultraviolet radiation, rain, pollutants, and temperature and humidity fluctuations, making them an important component of green building. However, architectural exterior wall coatings are constantly exposed to complex environments, and their weather resistance is affected by a combination of factors, including ultraviolet radiation, temperature and humidity fluctuations, and air pollution. This directly determines the product's lifespan and performance.

[0003] Accurately predicting the weather resistance of coatings and optimizing their formulations has become a significant technical challenge in the field of coating research and development. To address this issue, the paper "Siru Q, Yuequan D, Xiaoxia L, et al. Prediction and influence of the mass proportion of trichromatic colourants and acrylic substrate on the optical and thermal performance of external wall coatings: An artificial neural network approach [J]. Solar Energy Materials and SolarCells, 2022, 236" proposes a coating performance prediction method based on artificial neural networks (ANNs). This study focuses on analyzing the influence of the proportions of trichromatic colourants and acrylic substrate on solar reflectance, validating the application potential of ANNs in optical performance prediction. However, the scope of this paper is limited to the single indicator of solar reflectance and fails to predict the comprehensive indicators of coating weather resistance. Summary of the Invention

[0004] This invention aims to propose an improved method based on the aforementioned paper. Using the proportions of the three primary color pigments, the proportion of the acrylic substrate, and the coating thickness as inputs, a multi-task supervised neural network is employed to achieve comprehensive prediction of the weather resistance of coatings. This method not only overcomes the limitations of existing research but also provides an intelligent solution for the comprehensive evaluation and optimization of coating performance.

[0005] To achieve the above-mentioned objectives, the technical solution provided by this invention includes: A method for predicting the weather resistance of coatings based on multi-task supervised neural networks includes the following steps: Obtain the mass ratio of the three primary color pigments to the acrylic substrate and the coating thickness of the target coating; A first neural network is constructed and trained to predict the solar reflectance of a coating, taking the mass ratio as input and outputting the solar reflectance of the target coating. A second neural network for predicting the weather resistance of coatings is constructed and trained based on a multi-task supervised neural network. The solar reflectance and coating thickness of the target coating are input, and the ultraviolet absorption and surface temperature change of the target coating are output as intermediate parameters. The comprehensive weather resistance index is output as the weather resistance prediction result of the target coating. The second neural network includes a shared hidden layer, an intermediate parameter output layer, an independent hidden layer, and a prediction output layer connected in sequence. The training methods for the second neural network include: The initialization of several shared hidden layers and intermediate parameter output layers is completed through transfer learning, and the training of the second neural network is completed through supervised learning.

[0006] Preferably, the method for initializing several shared hidden layers and intermediate parameter output layers through transfer learning includes: transferring the network parameters of the first neural network to the shared hidden layers by freezing the weights; and performing adaptive initialization on the weights of the intermediate parameter output layers to complete the transfer.

[0007] Preferably, the method for training the second neural network through supervised learning includes: Freeze the shared hidden layer and intermediate parameter output layer, and pre-train the second neural network using a publicly available material property dataset; The quality ratio data and coating thickness data used to train the first neural network are obtained as the first training set, and the second neural network is fine-tuned.

[0008] Preferably, the fine-tuning method includes: The mass ratio data and coating thickness data used to train the first neural network are used as the experimental basis. Standard weather resistance tests are conducted to obtain weather resistance evaluation indicators. The mass ratio data, coating thickness data and corresponding weather resistance evaluation indicators are used as the first training set to supervise the training of the second neural network.

[0009] Preferably, the method for obtaining the first training set further includes: expanding the first training set by random perturbation.

[0010] The present invention also provides an apparatus for implementing the above-described method.

[0011] The present invention also provides a storage medium comprising a stored program, wherein the program executes the method described above when it is run.

[0012] The present invention also provides a processor for running a program, wherein the program executes the above-described method when it runs. Beneficial effects

[0013] 1. Improve prediction efficiency and reduce reliance on experiments: This invention achieves rapid prediction of coating weather resistance by constructing an intelligent neural network model, which significantly reduces reliance on traditional laboratory testing and reduces R&D cycle and cost investment.

[0014] 2. Comprehensive coverage of weather resistance indicators, enhancing the comprehensiveness of performance evaluation: This invention not only predicts the solar reflectivity of the coating, but also outputs intermediate parameters such as ultraviolet absorption and surface temperature change, ultimately achieving a comprehensive evaluation of weather resistance indicators, overcoming the limitations of the single target of existing technologies.

[0015] 3. Introducing transfer learning to improve the prediction ability of small datasets: By transferring the parameters of the first neural network to the shared hidden layer and intermediate parameter output layer of the second neural network, this invention effectively utilizes existing data resources and significantly improves the generalization performance of the model under small dataset conditions.

[0016] 4. Optimize model architecture to enhance prediction accuracy and applicability: This invention is based on a multi-task supervised neural network, combining the design of shared hidden layers and independent hidden layers, which can handle multiple target tasks simultaneously and improve the prediction accuracy of the model; and expand the training set through data augmentation technology to adapt to the prediction needs of different coating formulations and environmental conditions. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the process for predicting the weather resistance of coatings based on a multi-task supervised neural network, provided in a preferred embodiment of the present invention. Figure 2 This is a schematic diagram of data transmission for a coating weather resistance prediction method based on a multi-task supervised neural network provided in a preferred embodiment of the present invention. Figure 3 This is a schematic diagram of the second neural network structure and data transmission of a coating weather resistance prediction method based on a multi-task supervised neural network provided in a preferred embodiment of the present invention; Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described below with reference to the accompanying drawings. In the description of this invention, it should be understood that the terms "upper," "lower," "front," "rear," "left," "right," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention.

[0019] like Figure 1 and Figure 2 As shown, this embodiment provides a method for predicting the weather resistance of coatings based on a multi-task supervised neural network, including the following steps: S1. Obtain the mass ratio of the three primary color pigments to the acrylic substrate and the coating thickness of the target coating.

[0020] It should be noted that the three primary color pigments refer to the three basic pigments used to mix colors through the subtractive color mixing principle, specifically including cyan, magenta, and yellow. The acrylic base refers to a resin matrix formed by polymerizing acrylic acid or its derivatives, serving as a carrier for pigments, additives, and solvents in coatings. The three primary color pigments and the acrylic base are core components of coating formulations, playing a crucial role in predicting the optical, thermal, and weather resistance properties of the coatings.

[0021] In their earlier paper, "Siru Q, Yuequan D, Xiaoxia L, et al. Prediction and influence of the mass proportion of trichromatic colourants and acrylic substrate on the optical and thermal performance of external wallcoatings: An artificial neural network approach [J]. Solar Energy Materials and Solar Cells, 2022, 236," the inventors conducted an in-depth study on the influence of the proportion of trichromatic colourants on the solar reflectivity of coatings, and performed detailed analysis through experiments and an artificial neural network (ANN) model. The results show that the proportion of trichromatic colourants is a key factor determining the solar reflectivity of coatings. By absorbing and reflecting light of different wavelengths, adjusting the proportion of trichromatic colourants significantly affects the optical properties of the coating, thereby optimizing its solar reflectivity.

[0022] Furthermore, the acrylic base plays a crucial regulatory role in coating formulations. Its transparency, refractive index, and pigment dispersion not only enhance the optical performance of the three primary color pigments but also further improve the coating's ability to reflect sunlight. The synergistic effect of the acrylic base and the three primary color pigments has been proven to be significant in improving coating performance.

[0023] By optimizing the ratio of pigments to substrate, this study successfully explored an effective method to improve coating formulations, thereby significantly enhancing solar reflectivity. This research provides a solid theoretical basis for the optical performance optimization design of exterior wall coatings and lays a scientific foundation for further research on weather resistance prediction and coating performance improvement. The introduction of artificial neural networks in this study not only improved the efficiency of formulation optimization but also provided important insights for the development of intelligent coating performance prediction technologies.

[0024] Based on this, the present invention considers the close relationship between coating thickness and the optical properties, thermal properties, and weather resistance of the coating. Furthermore, it can form a complete coating formulation description with the ratio of the three primary color pigments and the acrylic substrate, providing multi-dimensional data support and possessing significant technical value and practical significance. The selection of coating thickness is explained in detail below from the three aspects directly affecting optical properties, thermal properties, and weather resistance: 1. The effect of coating thickness on optical performance: If the coating is too thin, light may pass through the coating without being effectively reflected, resulting in reduced reflectivity. If the coating is too thick, light scattering may be weakened, while increasing material costs. Obviously, reflectivity is directly related to the absorption of different wavelengths of light; higher reflectivity leads to lower absorption, and vice versa.

[0025] 2. The impact of coating thickness on thermal properties: Thick coatings lengthen the heat conduction path, slowing heat transfer and reducing the surface temperature of the exterior wall. Thin coatings allow heat to penetrate more easily, resulting in higher surface temperatures. Furthermore, thicker coatings can absorb more ultraviolet radiation, thus slowing down the aging of the substrate. However, excessively thick coatings may increase heat accumulation, which is detrimental to the paint's heat dissipation performance.

[0026] 3. The Impact of Coating Thickness on Weather Resistance: Besides the indirect impacts on weather resistance caused by the effects on optical and thermal properties mentioned above, coating thickness is a key factor determining the weather resistance of a coating. Thinner coatings may crack or wear under long-term exposure, reducing weather resistance. Thicker coatings offer better protection against ultraviolet radiation, humidity, and chemical corrosion. Furthermore, excessively thick coatings may cause cracking or peeling due to stress concentration, affecting the overall weather resistance index.

[0027] In summary, in addition to considering the influence of the proportions of the three primary color pigments and the acrylic substrate on the optical and thermal properties of the coating, this invention also considers the feature input of increasing the coating thickness to further supplement the spatial distribution characteristics of the coating in practical applications, which is a necessary parameter for predicting coating performance.

[0028] S2. Construct and train a first neural network for predicting the solar reflectance of the coating, taking the mass ratio as input and outputting the solar reflectance of the target coating. The first neural network can directly adopt the model in the aforementioned paper, and its construction and training methods will not be described in detail here.

[0029] S3. Construct and train a second neural network based on a multi-task supervised neural network to predict the weather resistance of the coating. Input the solar reflectance and coating thickness of the target coating, output the ultraviolet absorption and surface temperature change of the target coating as intermediate parameters, and output the comprehensive weather resistance index as the weather resistance prediction result of the target coating.

[0030] It should be understood that the comprehensive weather resistance index is a quantitative indicator used to evaluate the overall performance stability and anti-aging ability of coatings under long-term environmental exposure. It comprehensively considers the optical, physical, and chemical properties of the coating, providing a unified evaluation standard through a weighted or unweighted combination of various key performance parameters. Those skilled in the art should know that the weather resistance data can be obtained by testing the target coating using weather resistance test methods specified in national standards (such as GB / T 14522) or industry standards. Such weather resistance data is authoritative and reliable. Since this invention uses a neural network for black-box prediction, it cannot provide conclusions as authoritative as national standard test results. Therefore, this invention considers embedding experimental data into the components and training method of the second neural network, using the weather resistance evaluation index obtained through standard weather resistance tests as the training set to train the second neural network, thereby enhancing its robustness within the standard range.

[0031] Specifically, such as Figure 3As shown, the second neural network includes a shared hidden layer, an intermediate parameter output layer, an independent hidden layer, and a prediction output layer connected in sequence. The shared hidden layer extracts basic features from the input data and provides general feature representations for subsequent tasks (intermediate parameter prediction and final weather resistance index prediction), reducing redundant computation and improving model efficiency. The input data consists of the solar reflectance and coating thickness of the target coating, output by the first neural network. The intermediate parameter output layer predicts the UV absorption and surface temperature changes of the coating; these parameters are crucial intermediate results connecting the input features and the weather resistance index. UV absorption and surface temperature changes are fundamental to the weather resistance index, establishing a logical link between the input features and the final output as intermediate variables. The prediction results of the intermediate parameters help to better understand the model's behavior, thereby improving the transparency and reliability of the prediction process. The independent hidden layer further processes the features of the intermediate parameters and the shared hidden layer, extracting high-order features specific to the weather resistance index. The independent hidden layer provides customized features for the weather resistance index prediction task without sharing them with other intermediate tasks, avoiding interference. The design of independent hidden layers allows the final output to fully consider the characteristics of various complex inputs and intermediate parameters, improving the overall predictive ability of the model. The prediction output layer is the final layer of the neural network, used to output the predicted result of the comprehensive weather resistance index. Based on the high-order features extracted by the independent hidden layers, the comprehensive weather resistance index is calculated.

[0032] The training methods for the second neural network include: The initialization of several shared hidden layers and intermediate parameter output layers is completed through transfer learning, and the training of the second neural network is completed through supervised learning.

[0033] Transfer learning is a method that improves the training efficiency and performance of a target task by transferring model parameters, structure, or knowledge from a source task to a target task. In this invention, considering the high cost and long time required to obtain complete experimental data, which may result in insufficient training data, transfer learning is used to leverage existing data and model experience, reducing the target task's dependence on large-scale training data. Therefore, considering the generalizability between the features learned in the first neural network and the intermediate layers of the multi-task learning model, the low-level feature extraction layer in the first neural network has successfully captured the relationship between input features and solar reflectivity. These features are also valuable for predicting ultraviolet absorption and surface temperature changes. Furthermore, the output (solar reflectivity) of the first neural network has a close physical relationship with the intermediate parameter predictions; for example, high solar reflectivity usually corresponds to lower ultraviolet absorption and smaller surface temperature changes. Therefore, initializing the shared hidden layer and intermediate parameter output layer through transfer learning can significantly reduce the training time required for random parameter initialization, accelerate convergence, and improve prediction performance.

[0034] In some preferred embodiments, a superior method is provided for initializing several shared hidden layers and intermediate parameter output layers through transfer learning. Specifically, this method includes: transferring the network parameters of the first neural network to the shared hidden layers by freezing weights; wherein, freezing weights means retaining some model weights in the first neural network during the transfer learning process, so that they are not updated during the training of the second neural network. Freezing weights reduces the number of parameters that need to be updated during training, thereby reducing the computational cost and time of training.

[0035] Adaptive initialization is performed on the weights of the intermediate parameter output layer to complete the transfer learning. Adaptive initialization refers to adjusting or re-initializing some model weights according to the requirements of the target task in transfer learning, so as to better adapt to the prediction requirements of the target task. The reason for this design is that the intermediate parameter output layer needs to predict ultraviolet absorption and surface temperature changes, while the first neural network only outputs solar reflectivity. Although there is a physical correlation among the three, directly using the weights of the first neural network may lead to errors, so adaptive initialization is necessary. Compared with completely random initialization, adaptive initialization utilizes some transferred weights and structural experience to transfer the knowledge of solar reflectivity prediction to the prediction task of ultraviolet absorption and surface temperature changes, avoiding the errors caused by completely random initialization and making the model converge more easily and quickly.

[0036] Supervised learning is a machine learning method that trains a model to accurately predict target values ​​on unknown data by learning the mapping relationship between input features and target outputs. Its core idea is to provide a clearly labeled training dataset to guide the model in progressively optimizing its parameters, ultimately achieving effective modeling of complex tasks. Training data typically consists of pairs of input features X (such as experimental parameters or recipe proportions) and corresponding target outputs Y (such as experimental results or performance metrics). The model's goal is to minimize the error between the predicted and actual values ​​(such as mean squared error or cross-entropy), thereby obtaining a predictive model with good generalization ability.

[0037] In this invention, a supervised learning method is used to train a second neural network. Its core objective is to construct an intelligent predictive model for the weather resistance of coatings based on weather resistance evaluation indicators obtained from standard weather resistance tests. Specifically, through standard weather resistance tests, the input features of the coating (proportion of three primary color pigments, proportion of acrylic substrate, and coating thickness) and the corresponding weather resistance evaluation results are obtained (the specific evaluation indicators depend on the standards used and are not further limited here). These standard test data not only possess high scientific validity and authority but also provide clear supervisory signals for the model, guiding the adjustment of model parameters.

[0038] To fully utilize this high-quality experimental data, this invention uses data from standard experimental procedures as input to the model and the final weather resistance assessment results obtained from the experiment as the target output. By defining a reasonable loss function (such as the error of the comprehensive weather resistance index), the difference between the model's predicted results and the experimentally labeled values ​​is calculated, and the model's weights and bias parameters are gradually optimized. This approach ensures the scientific validity and reliability of the model's predictions while significantly improving its robustness within the standard experimental range.

[0039] Furthermore, by utilizing supervised learning methods, the model can not only learn the direct relationship between input features and target output, but also capture the inherent patterns and latent dynamics of the data through the training process. By learning the correlation between intermediate parameters such as pigment ratio, coating thickness, UV absorption, and surface temperature changes, the predictive ability of the comprehensive weather resistance index is further improved. Ultimately, the trained second neural network can accurately and efficiently predict the weather resistance of coatings with unknown formulations.

[0040] In some preferred embodiments, a preferred method for training a second neural network through supervised learning is provided, specifically including: S301. Freeze the shared hidden layer and intermediate parameter output layer, and pre-train the second neural network using a publicly available material property dataset. The shared hidden layer and intermediate parameter output layer already have weights from the first neural network during the transfer learning phase. Freezing these layers prevents their parameters from being updated during pre-training, reducing the number of parameters that need optimization and lowering training complexity. This allows computational resources to be concentrated on the newly introduced independent hidden layer and prediction output layer in the second neural network, enabling more efficient model training.

[0041] Because obtaining weather resistance test data relies on standard experiments, requiring complex experimental conditions and long experimental cycles, the amount of experimental data is limited, making it difficult to meet the needs of deep learning models. Furthermore, small sample data may lead to ineffective model training, resulting in overfitting or inaccurate predictions. Therefore, this invention considers pre-training the model on publicly available large-scale material property datasets. This allows the model to learn the optical, thermal, and other properties of the materials in advance, compensating for the limitations of insufficient experimental data. After completing initial feature learning on the publicly available datasets, the model only requires a small amount of high-quality experimental data for fine-tuning to adapt to the target task.

[0042] Publicly available material performance datasets refer to structured collections of data containing material performance parameters released by research institutions, standards organizations, industry associations, or other public sources. These datasets are typically used to study the physical, chemical, and thermal properties of materials and are widely applied in materials design, performance optimization, and the development of predictive models. Examples include the Open Quantum Materials Database (OQMD), the Matbench Dataset, and the Materials Project database. This invention considers utilizing authoritative material databases containing optical, thermal, and environmental performance data to provide a scientific basis for the pre-training of a second neural network, helping the model achieve higher initial performance and generalization ability in coating performance prediction, while reducing experimental costs and data requirements.

[0043] S302. Obtain the quality ratio data and coating thickness data used to train the first neural network as the first training set, and fine-tune the second neural network. The quality ratio data and coating thickness data used to train the first neural network are the basic data for supervised training in the preceding steps. These data are derived from standard experimental results, have high reliability, simplify the data preparation process, and fully utilize existing resources, achieving high efficiency and consistency in model training.

[0044] Fine-tuning refers to the process of retraining an existing pre-trained model using a small amount of specific data from the target task. Its goal is to adjust some or all of the model's parameters to improve its accuracy and adaptability on the target task. In this invention, fine-tuning uses the mass ratio and coating thickness data from the first neural network training. This data is highly reliable and contains clear weather resistance evaluation results, enabling accurate prediction of the overall weather resistance index based on the coating's mass ratio and coating thickness.

[0045] In some preferred embodiments, specific fine-tuning methods are provided, including: The mass ratio and coating thickness data used to train the first neural network are used as the experimental basis. Standard weather resistance tests are conducted to obtain weather resistance evaluation indicators. The mass ratio data, coating thickness data, and corresponding weather resistance evaluation indicators are used as the first training set to supervise the training of the second neural network. Specifically, standard weather resistance tests are conducted on paint samples with different ratios of three primary color pigments to acrylic substrates and different coating thicknesses to obtain key performance data related to weather resistance. The mass ratio data, coating thickness data, and corresponding weather resistance evaluation indicators obtained in these experiments are summarized to construct the first training set, which not only reflects the real physical and chemical relationship between paint formulation parameters and weather resistance indicators, but also provides accurate and high-quality supervision signals for the training of the second neural network. Based on this, supervised learning methods are used to train the second neural network, enabling it to effectively learn the mapping relationship between input features (mass ratio, coating thickness) and output results (weather resistance evaluation indicators). Through this training process, the second neural network can gradually optimize its internal parameters and accurately predict the weather resistance performance of new paint formulations under specific experimental conditions. This method not only makes full use of the scientific rigor and authority of experimental data, but also ensures the reliability and practicality of model prediction results, providing solid data and technical support for intelligent coating performance evaluation.

[0046] It should be understood that the first training set data mainly comes from standard weathering tests, which are costly and time-consuming, resulting in limited large-scale data that is insufficient to meet the training needs of deep learning models. Furthermore, insufficient data may lead to incomplete coverage of input features by the model, resulting in overfitting or inaccurate predictions. Therefore, in some preferred embodiments, new data can be generated by randomly perturbing existing data, which can quickly expand the dataset size without increasing experimental costs. This randomly perturbed data simulates small changes in input parameters in real-world scenarios, enabling the model to better adapt to inputs under different conditions. Specifically, the random perturbation expansion is implemented as follows: The mass ratio and coating thickness data are subject to random perturbation within a certain range. For example, ±5% random deviation is added to the proportion of the three primary color pigments; ±10% random fluctuation is added to the coating thickness. This perturbation range is ensured to be within reasonable physical and process constraints to avoid generating unreasonable samples.

[0047] The generated perturbation data is initially verified to ensure that the expanded sample distribution is reasonable and consistent with actual production or experimental conditions.

[0048] By combining the perturbation-generated data with the original experimental data, a more diverse and comprehensive training set is constructed for model training.

[0049] In specific embodiments of the present invention, an apparatus, a storage medium, and a processor are also provided based on the above-mentioned method for predicting the weather resistance of coatings based on multi-task supervised neural networks. Their specific configurations and functions in practical applications are described for different implementation requirements.

[0050] The device is used to implement a coating weather resistance prediction method based on a multi-task supervised neural network. Specifically, it includes a hardware module and a software module, and can support the entire process of calculation and output from input coating parameters to predicted comprehensive weather resistance index.

[0051] Secondly, the storage medium (e.g., hard disk, solid-state drive, USB flash drive or other readable storage device) stores the code of a running program, which is designed to perform operations such as data input, model calculation, and result output in accordance with the method of the present invention when running in a computing device (such as a computer, server or embedded system), ensuring the automated execution of the method and the accuracy of the results.

[0052] Furthermore, the processor (such as a CPU, GPU, or dedicated AI chip) is used to run the aforementioned program. By executing the program, the processor completes the core tasks of data processing and model prediction, including feature extraction, model inference, and result generation, ensuring the efficient execution of the weather resistance prediction task.

[0053] These implementation methods, through the collaborative design of hardware and software, constitute a complete implementation system of the method of the present invention, providing a flexible and efficient solution that can be widely applied to practical application scenarios of intelligent coating performance prediction.

[0054] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for predicting the weather resistance of coatings based on a multi-task supervised neural network, characterized in that, Including the following steps: Obtain the mass ratio of the three primary color pigments to the acrylic substrate and the coating thickness of the target coating; A first neural network is constructed and trained to predict the solar reflectance of a coating, taking the mass ratio as input and outputting the solar reflectance of the target coating. A second neural network for predicting the weather resistance of coatings is constructed and trained based on a multi-task supervised neural network. The solar reflectance and coating thickness of the target coating are input, and the ultraviolet absorption and surface temperature change of the target coating are output as intermediate parameters. The comprehensive weather resistance index is output as the weather resistance prediction result of the target coating. The second neural network includes a shared hidden layer, an intermediate parameter output layer, an independent hidden layer, and a prediction output layer connected in sequence. The training methods for the second neural network include: The initialization of several shared hidden layers and intermediate parameter output layers is completed through transfer learning, and the training of the second neural network is completed through supervised learning.

2. The coating weather resistance prediction method based on a multi-task supervised neural network as described in claim 1, characterized in that, The method for initializing several shared hidden layers and intermediate parameter output layers through transfer learning includes: transferring the network parameters of the first neural network to the shared hidden layers by freezing the weights; and performing adaptive initialization on the weights of the intermediate parameter output layers to complete the transfer.

3. The coating weather resistance prediction method based on a multi-task supervised neural network as described in claim 1 or 2, characterized in that, The method for training the second neural network through supervised learning includes: Freeze the shared hidden layer and intermediate parameter output layer, and pre-train the second neural network using a publicly available material property dataset; The quality ratio data and coating thickness data used to train the first neural network are obtained as the first training set, and the second neural network is fine-tuned.

4. The coating weather resistance prediction method based on a multi-task supervised neural network as described in claim 3, characterized in that, The fine-tuning method includes: The mass ratio data and coating thickness data used to train the first neural network are used as the experimental basis. Standard weather resistance tests are conducted to obtain weather resistance evaluation indicators. The mass ratio data, coating thickness data and corresponding weather resistance evaluation indicators are used as the first training set to supervise the training of the second neural network.

5. The coating weather resistance prediction method based on a multi-task supervised neural network as described in claim 4, characterized in that, The method for obtaining the first training set further includes expanding the first training set by random perturbation.

6. An apparatus, characterized in that, The apparatus is used to implement the method according to any one of claims 1 to 5.

7. A storage medium, characterized in that, The storage medium includes a stored program, wherein the program, when executed, performs the method according to any one of claims 1 to 5.

8. A processor, characterized in that, The processor is used to run a program, wherein the program executes the method of any one of claims 1 to 5 when it runs.

Images