A method for preparing a microcapsule-based photothermal self-healing anti-icing coating

Abstract

This invention discloses a method for preparing a microcapsule-based photothermal self-healing anti-icing coating, belonging to the field of functional coating materials technology. The method includes: obtaining the core material and wall material composition parameters of the microcapsules, and collecting the polyacrylamide concentration and dopamine polymerization time; constructing a morphology-photothermal correlation model of the microcapsules to obtain the predicted photothermal conversion efficiency; performing multi-objective optimization on the core and wall material composition parameters with the target photothermal conversion efficiency as a constraint, and iteratively inverting to obtain the optimal microcapsule configuration; preparing primary microcapsules and composited with a dopamine photothermal modification layer on their surface to obtain a photothermal self-healing microcapsule additive; mixing the photothermal self-healing microcapsule additive with an epoxy resin-based coating at an adaptive mass ratio, coating and curing to obtain an anti-icing coating. This invention achieves accurate prediction of the photothermal conversion efficiency of microcapsules and intelligent optimization of composition parameters, obtaining an anti-icing coating with both high photothermal conversion efficiency and rapid self-healing capability.

Description

Technical Field

[0001] This invention relates to the field of functional coating materials technology, specifically to a method for preparing a microcapsule-based photothermal self-healing anti-icing coating. Background Technology

[0002] Power grid transmission and transformation equipment, such as overhead transmission line conductors and insulator strings, are highly susceptible to severe icing in winter conditions of freezing rain or wet snow. Icing can lead to increased conductor load and decreased flashover voltage of insulator strings, which in turn can cause power accidents such as line breaks, tower collapses, and flashovers, seriously threatening the safe and stable operation of the power grid.

[0003] To address icing issues, traditional de-icing methods primarily employ manual knocking, high-current DC melting, or applying ordinary hydrophobic coatings. However, manual knocking relies on high-altitude operations, which are inefficient and prone to damaging conductors and insulator surfaces; high-current DC melting requires power outages, consumes significant energy, and is only applicable to specific lines; while ordinary hydrophobic coatings can delay icing, their surface microstructure is easily worn, resulting in short-lived anti-icing effects and a lack of active de-icing capability. Therefore, existing methods all suffer from complex operation, high energy consumption, and poor durability, making it difficult to meet the long-term anti-icing requirements of power grid equipment in extreme environments. Summary of the Invention

[0004] This invention addresses the technical problems of slow photothermal response, low self-healing efficiency, and difficulty in simultaneously achieving photothermal and self-healing functions in existing technologies by providing a method for preparing a microcapsule-based photothermal self-healing anti-icing coating.

[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0006] This invention provides a method for preparing a microcapsule-based photothermal self-healing anti-icing coating, comprising: The core material and wall material parameters of the microcapsules were obtained, and the polyacrylamide concentration and dopamine polymerization time were collected. Using the polyacrylamide concentration and the dopamine polymerization time as inputs, a morphology-photothermal correlation model of the microcapsules is constructed, and the predicted photothermal conversion efficiency is obtained as the output. Based on the predicted photothermal conversion efficiency, and with the target photothermal conversion efficiency as a constraint, multi-objective optimization is performed on the core material composition parameters and wall material composition parameters, and the optimal microcapsule configuration is obtained through iterative inversion. Primary microcapsules were prepared according to the optimal microcapsule configuration, and a dopamine photothermal modification layer was composited on the surface of the primary microcapsules to obtain a photothermal self-healing microcapsule additive. The photothermal self-healing microcapsule additive is mixed with an epoxy resin-based coating at an adaptive mass ratio, and the coating is applied and cured to obtain the anti-icing coating.

[0007] The beneficial effects of this invention are: Compared to existing technologies, this invention first constructs a morphology-photothermal correlation model by collecting core and wall material component parameters, polyacrylamide concentration, and dopamine polymerization time, achieving accurate prediction of the photothermal conversion efficiency of microcapsules. Secondly, using the target photothermal conversion efficiency as a constraint, multi-objective optimization is performed on the component parameters, and the optimal microcapsule configuration is obtained through iterative inversion, enabling synergistic optimization of the micro / nano rough structure on the microcapsule surface and the photothermal modification layer. Finally, microcapsule additives are prepared according to the optimal configuration and mixed with epoxy resin-based coatings at an adaptive mass ratio to obtain an anti-icing coating. This invention solves the problems of slow photothermal response, low self-healing efficiency, and difficulty in simultaneously achieving photothermal and self-healing functions in existing technologies, realizing an integrated function of rapid photothermal de-icing and damage self-repair. Attached Figure Description

[0008] Figure 1 A schematic flowchart illustrating a method for preparing a microcapsule-based photothermal self-healing anti-icing coating provided by the present invention; Figure 2 This is a schematic diagram illustrating the calculation process of the adaptive mass ratio provided by the present invention. Detailed Implementation

[0009] Examples, such as Figure 1 As shown, this embodiment of the invention provides a method for preparing a microcapsule-based photothermal self-healing anti-icing coating, comprising: S10: Obtain the core material composition parameters and wall material composition parameters of the microcapsules, and collect the polyacrylamide concentration and dopamine polymerization time; First, the core material and wall material composition parameters of the microcapsules were obtained. Microcapsules are core-shell structured microparticles with a core material composed of epoxy resin and microcrystalline wax, and a wall material composed of inorganic nanomaterials and graphene. The core material composition parameter is the mass ratio of epoxy resin to microcrystalline wax, which determines the melting characteristics of the core material and the release rate of the repair agent. The wall material composition parameter is the mass ratio of inorganic nanomaterials to graphene, which affects the mechanical strength and photothermal absorption capacity of the wall material.

[0010] Simultaneously, the concentration of polyacrylamide and the polymerization time of dopamine were collected. The concentration of polyacrylamide refers to the concentration of the surfactant added during the primary preparation of microcapsules. This concentration controls the degree of self-assembly and aggregation of inorganic nanoparticles and graphene on the surface of microcapsules, directly affecting the formation density of micron-nano-level rough structures on the surface of microcapsules. The polymerization time of dopamine refers to the time for the primary microcapsules to undergo surface polymerization reaction in dopamine hydrochloride solution. This time determines the polymerization thickness and density of the dopamine photothermal modified layer on the surface of microcapsules.

[0011] By collecting the above parameters, input variables can be provided for the construction of subsequent morphology-photothermal correlation models, which can be used to predict the photothermal conversion efficiency of microcapsules under different process conditions.

[0012] Specifically, the core material composition parameters and wall material composition parameters of the microcapsules were obtained, and the polyacrylamide concentration and dopamine polymerization time were collected, including: The core material composition parameters are configured by mixing epoxy resin and microcrystalline wax in a mass ratio ranging from 1:1 to 3:1. Inorganic nanomaterials and graphene were configured in a mass ratio ranging from 1:1 to 10:1 as the wall material composition parameters; The concentration of polyacrylamide was collected within the range of 0.1% to 1.0% based on the required density of micron-level protrusions on the surface of the microcapsules. The polymerization time of dopamine was collected in a dopamine solution of 2 to 5 mg / mL based on the required photothermal modified layer thickness on the surface of the microcapsules, wherein the polymerization time ranged from 4 to 12 hours.

[0013] First, the epoxy resin and microcrystalline wax were configured as core material components in a mass ratio ranging from 1:1 to 3:1. The mass ratio of epoxy resin to microcrystalline wax determines the melting characteristics of the core material and the release behavior of the repair agent. A higher proportion of epoxy resin increases the crosslinking density of the core material, resulting in higher mechanical strength of the repaired coating, but reduces melt flowability. A higher proportion of microcrystalline wax makes the core material more prone to melting and flowing out under photothermal temperatures, accelerating the self-healing response, but reducing the strength after repair. Limiting the mass ratio to the range of 1:1 to 3:1 balances the self-healing response speed and the mechanical properties after repair.

[0014] Secondly, the inorganic nanomaterials and graphene were configured as wall material composition parameters in a mass ratio ranging from 1:1 to 10:1. The inorganic nanomaterials, including silicon dioxide, cerium oxide, and titanium oxide, provided mechanical support and a rough structural framework for the microcapsule wall material; graphene enhanced photothermal absorption. Specifically, a higher proportion of inorganic nanomaterials increased the rigidity of the wall material and made the microcapsule structure more stable, but the photothermal conversion efficiency decreased due to the graphene content. Conversely, a higher proportion of graphene improved the photothermal conversion efficiency, but reduced the mechanical strength of the wall material. Limiting the mass ratio to within the 1:1 to 10:1 range balanced the mechanical strength and photothermal absorption performance of the wall material.

[0015] Furthermore, the concentration of polyacrylamide was determined to be within the range of 0.1% to 1.0% based on the required micron-level protrusion density on the microcapsule surface. Polyacrylamide, as a surface activity modifier, is used to induce the controlled aggregation of inorganic nanoparticles and graphene on the core material surface through physical adsorption, forming a micron-level protrusion structure. Lower polyacrylamide concentrations result in weaker aggregation and lower protrusion density; higher concentrations result in stronger aggregation and higher protrusion density. Limiting the concentration range to 0.1% to 1.0% allows for the control of forming a moderately dense micro / nano rough structure on the microcapsule surface, ensuring a large effective surface area for photothermal conversion while avoiding wall material defects caused by excessive aggregation.

[0016] Meanwhile, the polymerization time of dopamine was collected in a dopamine solution with a concentration of 2 to 5 mg / mL, ranging from 4 to 12 hours, based on the required photothermal modified layer thickness on the microcapsule surface. Dopamine undergoes a self-polymerization reaction under alkaline conditions, forming a polydopamine photothermal modified layer on the microcapsule surface. The concentration of the dopamine solution determines the amount of material supplied for the polymerization reaction; the higher the concentration, the thicker the modified layer formed per unit time. The polymerization time determines the extent of the reaction; the longer the time, the denser the modified layer. By limiting the concentration to 2 to 5 mg / mL and the time to 4 to 12 hours, the thickness of the modified layer can be controlled to be moderate, ensuring sufficient photothermal absorption capacity while avoiding excessively thick modified layers that could lead to microcapsule aggregation.

[0017] In summary, the above parameter collection provides a complete input variable space for the subsequent construction of the morphology-photothermal correlation model.

[0018] S20: Using the polyacrylamide concentration and the dopamine polymerization time as inputs, construct a morphology-photothermal correlation model of the microcapsules and output the predicted photothermal conversion efficiency. Secondly, a morphology-photothermal correlation model for microcapsules was constructed using polyacrylamide concentration and dopamine polymerization time as inputs. This morphology-photothermal correlation model is a machine learning model based on a fully connected regression network, used to establish a nonlinear mapping relationship between microcapsule preparation process parameters and photothermal conversion efficiency. Polyacrylamide concentration directly affects the formation density of micron-nano-level rough structures on the microcapsule surface, while dopamine polymerization time determines the polymerization thickness and density of the photothermal modification layer. Both factors jointly determine the light absorption capacity and photothermal conversion efficiency of the microcapsules. By constructing the morphology-photothermal correlation model, the photothermal conversion efficiency of microcapsules under different combinations of process parameters can be rapidly predicted without actual preparation and testing, providing an efficient evaluation method for subsequent optimization of component parameters.

[0019] The steps for constructing the morphology-photothermal correlation model of the microcapsules include: Multiple groups of microcapsule samples were collected under different polyacrylamide concentrations and different dopamine polymerization times. The actual photothermal conversion efficiency of each group of microcapsule samples was measured to obtain the sample parameter set and the corresponding photothermal conversion efficiency label. The polyacrylamide concentration and the dopamine polymerization time are encoded into an input feature vector; Construct a fully connected regression network architecture, wherein the fully connected regression network architecture includes an input layer, at least two hidden layers, and an output layer; Using the input feature vector as input and the photothermal conversion efficiency label as supervision signal, the fully connected regression network architecture is trained under supervision until the regression loss converges, thereby obtaining the morphology-photothermal correlation model.

[0020] First, multiple sets of microcapsule samples were collected under different polyacrylamide concentrations and different dopamine polymerization durations. The actual photothermal conversion efficiency of each set of microcapsule samples was measured to obtain a sample parameter set and a corresponding photothermal conversion efficiency label. Specifically, multiple concentration points were selected within the polyacrylamide concentration range of 0.1% to 1.0%, and multiple duration points were selected within the dopamine polymerization duration range of 4 to 12 hours. Multiple sets of microcapsule samples were prepared through combined experiments. After each set of samples was prepared, the samples were irradiated with xenon lamps simulating sunlight under standard light intensity. The surface temperature change of the samples was measured using an infrared thermometer, and the actual photothermal conversion efficiency was calculated based on the heating rate. The polyacrylamide concentration, dopamine polymerization duration, and corresponding actual photothermal conversion efficiency of each set of samples were combined to form a sample. All samples constituted a sample parameter set, and the actual photothermal conversion efficiency served as a monitoring label.

[0021] Secondly, the polyacrylamide concentration and dopamine polymerization time are encoded as input feature vectors. Since both polyacrylamide concentration and dopamine polymerization time are continuous numerical variables, they can be directly concatenated to form a two-dimensional input feature vector, such as [polyacrylamide concentration value, dopamine polymerization time value]. To improve the model training effect, the input features can also be normalized, mapping the concentration and time to the interval between 0 and 1 respectively, to eliminate the influence of dimensional differences on model convergence.

[0022] Furthermore, a fully connected regression network architecture is constructed, comprising an input layer, at least two hidden layers, and an output layer. Specifically, the number of nodes in the input layer equals the dimension of the input feature vector; in this scheme, the input features are two-dimensional, therefore the input layer has two nodes. The hidden layers adopt a fully connected structure, with each neuron in the hidden layer connected to all neurons in the previous layer. For example, two hidden layers can be set, with the first hidden layer containing 64 neurons and the second hidden layer containing 32 neurons. Each neuron uses the ReLU activation function to introduce nonlinear transformation capability. The output layer has one node, uses a linear activation function, and outputs the predicted photothermal conversion efficiency value.

[0023] Furthermore, using the input feature vector as input and the photothermal conversion efficiency label as the supervision signal, the fully connected regression network architecture is trained under supervision until the regression loss converges, thus obtaining the morphology-photothermal correlation model. Specifically, during training, the sample parameter set is divided into a training set and a validation set, for example, in an 8:2 ratio. Mean squared error is used as the regression loss function to measure the deviation between the model's predicted photothermal conversion efficiency and the actual photothermal conversion efficiency label. The network weight parameters are iteratively updated using the backpropagation algorithm and the Adam optimizer. After each training epoch, the loss value on the validation set is calculated. Training stops when the validation set loss value no longer decreases for several consecutive epochs or reaches the preset number of training epochs; the network parameters at this point represent the completed morphology-photothermal correlation model.

[0024] The convergence criteria are set based on the trend of the loss value on the validation set and the upper limit of the number of training epochs. Specifically, when the mean squared error loss value on the validation set no longer decreases after 10 consecutive training epochs, the model is considered to have converged, and training is terminated. If the loss value continues to decrease but the number of training epochs has reached the preset maximum number of training epochs, such as 200 epochs, training is also terminated. By setting the above dual convergence criteria, it is ensured that the model fully learns the nonlinear mapping relationship between polyacrylamide concentration, dopamine polymerization time, and photothermal conversion efficiency, while avoiding overfitting caused by overtraining.

[0025] This morphology-photothermal correlation model allows for rapid prediction of the photothermal conversion efficiency of microcapsules under given polyacrylamide concentration and dopamine polymerization time, eliminating the need for repeated preparation and testing.

[0026] Finally, the predicted photothermal conversion efficiency is output. Specifically, the collected polyacrylamide concentration and dopamine polymerization time are used as input feature vectors and fed into the trained morphology-photothermal correlation model. After forward computation by the model, the corresponding predicted photothermal conversion efficiency is output. This predicted photothermal conversion efficiency reflects the efficiency with which the microcapsules convert absorbed light energy into heat energy under the current process parameters. The value typically ranges from 0 to 1 or is expressed as a percentage, with higher values ​​indicating better photothermal performance. This predicted output allows for a quantitative assessment of the impact of different process parameter combinations on photothermal performance, providing an optimization target for subsequent optimization of component parameters constrained by the target photothermal conversion efficiency.

[0027] S30: Based on the predicted photothermal conversion efficiency, and with the target photothermal conversion efficiency as a constraint, perform multi-objective optimization on the core material composition parameters and wall material composition parameters, and iteratively invert to obtain the optimal microcapsule configuration; Furthermore, based on the predicted photothermal conversion efficiency obtained from the aforementioned morphology-photothermal correlation model, and constrained by the target photothermal conversion efficiency, multi-objective optimization is performed on the core material composition parameters and wall material composition parameters, and the optimal microcapsule configuration is obtained through iterative inversion. The predicted photothermal conversion efficiency is the predicted photothermal performance value under the combination of process parameters output by the morphology-photothermal correlation model. The target photothermal conversion efficiency is a pre-set expected performance index, such as 85% or 90%, representing the minimum photothermal conversion efficiency that the microcapsules need to achieve in practical applications. Constrained by the target photothermal conversion efficiency, only microcapsule configurations that meet or exceed this performance index are considered qualified candidates.

[0028] The core material composition parameter, namely the mass ratio of epoxy resin to microcrystalline wax, and the wall material composition parameter, namely the mass ratio of inorganic nanomaterials to graphene, together determine the self-healing performance, mechanical strength, and photothermal absorption capacity of the microcapsules. Multi-objective optimization refers to simultaneously optimizing both the core and wall material composition parameters to ensure that the microcapsules, while meeting the constraints of photothermal conversion efficiency, also balance self-healing response speed and wall material mechanical strength. Iterative inversion refers to the process of repeatedly adjusting the combination of composition parameters, using a morphology-photothermal correlation model to predict the corresponding photothermal conversion efficiency, selecting configurations that meet the constraints, and continuously optimizing to approach the optimal solution.

[0029] Specifically, this step stems from the fact that the component ratio of the core material and the wall material has a complex coupled influence on the overall performance of microcapsules. Relying solely on empirical trial and error makes it difficult to find the globally optimal ratio. By leveraging the rapid predictive capabilities provided by the morphology-photothermal correlation model, combined with a multi-objective optimization algorithm, the optimal microcapsule configuration can be efficiently searched within a large parameter space, achieving synergistic optimization of photothermal conversion efficiency, self-healing response speed, and wall material mechanical strength. The ultimately obtained optimal microcapsule configuration includes the optimal mass ratio of epoxy resin to microcrystalline wax and the optimal mass ratio of inorganic nanomaterials to graphene, providing precise formulation guidance for subsequent microcapsule preparation.

[0030] Specifically, based on the predicted photothermal conversion efficiency, and with the target photothermal conversion efficiency as a constraint, multi-objective optimization is performed on the core material composition parameters and wall material composition parameters, and the optimal microcapsule configuration is obtained through iterative inversion, including: Multiple candidate configurations are randomly initialized within the mass ratio range of the core material component parameters and the mass ratio range of the wall material component parameters. Each candidate configuration includes a mass ratio of epoxy resin to microcrystalline wax and a mass ratio of inorganic nanomaterial to graphene. Each candidate configuration is input into the morphology-photothermal correlation model, and the predicted photothermal conversion efficiency corresponding to each candidate configuration is output. The fitness of multiple candidate configurations is evaluated by minimizing the deviation between the predicted photothermal conversion efficiency and the target photothermal conversion efficiency. Calculate the upper quartile of the fitness values ​​of all candidate configurations, retain candidate configurations with fitness values ​​greater than the upper quartile, and generate the next generation of candidate configurations through crossover and mutation operations; Repeatedly perform fitness evaluation and iterative updates until the rate of change of the optimal fitness for multiple consecutive generations is lower than the dynamic convergence threshold, and the candidate configuration with the best fitness is taken as the optimal microcapsule configuration.

[0031] First, multiple candidate configurations are randomly initialized within the mass ratio ranges of the core material component parameters and the wall material component parameters. Specifically, the mass ratio range of the core material component parameters is 1:1 to 3:1 for the epoxy resin to microcrystalline wax ratio, and the mass ratio range of the wall material component parameters is 1:1 to 10:1 for the inorganic nanomaterial to graphene ratio. Random sampling is performed within each range to generate multiple candidate configurations, each containing one epoxy resin to microcrystalline wax mass ratio and one inorganic nanomaterial to graphene mass ratio. For example, the initial population size can be set to 50 groups, i.e., 50 different mass ratio combinations are randomly generated.

[0032] Secondly, each candidate configuration is input into the morphology-photothermal correlation model, which outputs the predicted photothermal conversion efficiency for each configuration. The morphology-photothermal correlation model has been trained using historical samples, establishing a mapping relationship between component parameters and photothermal conversion efficiency. Specifically, for each candidate configuration, the morphology-photothermal correlation model can output the corresponding predicted photothermal conversion efficiency value based on the mass ratio of epoxy resin to microcrystalline wax and the mass ratio of inorganic nanomaterials to graphene.

[0033] Furthermore, the fitness of multiple candidate configurations is evaluated by minimizing the deviation between the predicted and target photothermal conversion efficiencies. The target photothermal conversion efficiency is a pre-defined desired performance index, representing the minimum photothermal conversion efficiency level that the microcapsules need to achieve in practical applications. This target efficiency is set comprehensively based on the requirements of the anti-icing coating application scenario regarding the ice-melting response speed and the technical feasibility of the photothermal material; for example, it can be set to 85%. The smaller the absolute deviation between the predicted and target values, the closer the configuration is to the desired performance, and the higher its fitness. By calculating the deviation value for each candidate configuration and converting the deviation into a fitness value, the merits of each configuration can be quantitatively evaluated.

[0034] Specifically, the fitness of multiple candidate configurations is evaluated using the minimization of the deviation between the predicted and target photothermal conversion efficiencies as the fitness metric, including: Calculate the absolute deviation between the predicted photothermal conversion efficiency and the target photothermal conversion efficiency for each candidate configuration, and calculate the mean and standard deviation of the absolute deviations for all candidate configurations. The ratio of the absolute deviation to the mean is used as the relative deviation coefficient, and the relative deviation coefficient is multiplied by the standard deviation to obtain the weighted deviation; The candidate configurations are sorted in ascending order of weighted bias, and each candidate configuration is assigned a fitness that decays exponentially according to its ranking. The fitness calculation steps include: Subtract one from the sorting position and take the negative value, which is used as the exponent of the exponent power; Using the natural constant as the base, calculate the value of the current exponent and use it as the fitness value of the current candidate configuration; The fitness value of each candidate configuration is calculated sequentially from 1 to N according to the sorting position, where N is the total number of candidate configurations.

[0035] First, the absolute deviation between the predicted and target photothermal conversion efficiencies for each candidate configuration is calculated, and the mean and standard deviation of the absolute deviations for all candidate configurations are also calculated. Specifically, the absolute deviation is the absolute value of the difference between the predicted and target photothermal conversion efficiencies, reflecting the gap between the configuration and the desired performance. For example, if the target photothermal conversion efficiency is 85%, and a certain configuration predicts a photothermal conversion efficiency of 82%, then the absolute deviation is 3%, the mean is the average of the absolute deviations for all candidate configurations, and the standard deviation reflects the dispersion of the absolute deviation.

[0036] Secondly, the ratio of the absolute deviation to the mean is used as the relative deviation coefficient, and the relative deviation coefficient is multiplied by the standard deviation to obtain the weighted deviation. The relative deviation coefficient reflects the magnitude of the deviation of this configuration relative to the overall average deviation; a value greater than 1 indicates a deviation higher than the average level, and a value less than 1 indicates a deviation lower than the average level. Multiplying the relative deviation coefficient by the standard deviation yields the weighted deviation, which comprehensively considers both the degree of deviation of the configuration itself and the dispersion of the overall deviation, making the fitness assessment more discriminative.

[0037] Furthermore, multiple candidate configurations are ranked in ascending order of weighted bias, and each configuration is assigned a fitness level that decays exponentially according to its ranking. A smaller weighted bias indicates that the predicted photothermal conversion efficiency of the configuration is closer to the target value, signifying a superior configuration and thus a higher ranking. Specifically, the fitness decays exponentially, meaning that configurations with higher rankings receive higher fitness, and the fitness value decreases exponentially with increasing ranking, thus highlighting the difference between excellent and ordinary configurations.

[0038] The fitness calculation steps include: subtracting one from the ranking position and taking the negative value as the exponent of the power; calculating the value of the current power using the natural constant as the base, which is the fitness value of the current candidate configuration. Specifically, if the ranking position of a configuration is k, then its fitness is calculated as e raised to the power of -k-1. When k equals 1, the fitness is e raised to the power of 0, which equals 1; when k equals 2, the fitness is e raised to the power of -1, which equals approximately 0.3679; when k equals 3, the fitness is e raised to the power of -2, which equals approximately 0.1353; when k equals 4, the fitness is e raised to the power of -3, which equals approximately 0.0498. The fitness value is calculated sequentially from ranking position 1 to N, yielding the fitness value for each group of candidate configurations, where N is the total number of candidate configurations.

[0039] Through the aforementioned exponentially decaying fitness allocation method, the top-ranked excellent configurations can obtain significantly higher fitness and are preferentially retained in subsequent crossover and mutation operations, thereby guiding the optimization process to quickly converge to the optimal microcapsule configuration.

[0040] Furthermore, the upper quartiles of the fitness values ​​of all candidate configurations are calculated. Candidate configurations with fitness values ​​greater than the upper quartile are retained, and the next generation of candidate configurations are generated through crossover and mutation operations. The upper quartile refers to the value at the 75th percentile after sorting all candidate configurations by fitness values ​​from smallest to largest. Configurations with fitness values ​​higher than this threshold are retained as parents, while configurations with lower fitness values ​​are discarded.

[0041] Specifically, crossover recombination refers to randomly selecting two sets of configurations from the retained parent generation, exchanging and combining their component parameters to generate new offspring configurations; mutation operation refers to fine-tuning the component parameters in the newly generated offspring configurations with a certain probability, such as increasing or decreasing the mass ratio of epoxy resin to microcrystalline wax by a small random amount. Through the above operations, a new generation of candidate configurations with the same population size as the parent generation are generated.

[0042] Optionally, let M be the number of parental configurations retained, for example, M equals half the original population size. The specific operation of crossover recombination is as follows: randomly select two configurations from the retained M parental configurations, denoted as parent A and parent B. Parent A contains epoxy resin and microcrystalline wax in a mass ratio R. A The mass ratio of inorganic nanomaterials to graphene S A The parent generation B contains a quality ratio of R. B and S B Using a single-point crossover method, the mass ratios of the two configurations are swapped with a 50% probability to generate two offspring configurations: Offspring 1 is (R... A ,S B ), Child 2 is (R B ,S ARepeat the random selection and crossover operation described above until a new generation of candidate configurations with the same number of offspring as the original population is generated. For example, if the original population size is 50, then 50 offspring configurations need to be generated.

[0043] The specific operation of the mutation is as follows: In each progeny configuration generated by cross-recombination, the component parameters are fine-tuned with a preset mutation probability, such as 10%. When a progeny configuration triggers mutation, the mass ratio R of epoxy resin to microcrystalline wax or the mass ratio S of inorganic nanomaterials to graphene is randomly adjusted. The adjustment method is as follows: based on the original mass ratio, a mutation coefficient is multiplied, which is randomly generated in a uniform distribution within the range of 0.8 to 1.2. After mutation, it is necessary to check whether the mass ratio is still within the specified range, that is, the mass ratio of epoxy resin to microcrystalline wax should be between 1:1 and 3:1, and the mass ratio of inorganic nanomaterials to graphene should be between 1:1 and 10:1. If it exceeds the range, the value exceeding the boundary is adjusted to the nearest boundary value. For example, if the mass ratio of epoxy resin to microcrystalline wax in a certain progeny configuration is 2:1, after triggering mutation, multiplying by the random coefficient of variation of 1.15 yields 2.3:1, which is still within the range of 1:1 to 3:1 and is valid; if multiplied by 1.3, it yields 2.6:1, which is still within the range; if multiplied by 0.7, it yields 1.4:1, which is still within the range.

[0044] Through the aforementioned crossover and mutation operations, the new generation of candidate configurations inherits the superior genes of the parent generation's excellent configurations while introducing new random mutations, maintaining population diversity and preventing the optimization process from prematurely falling into local optima. These operations can generate a new generation of candidate configurations with the same population size as the parent generation.

[0045] Furthermore, the fitness evaluation and iterative update are repeated until the rate of change of the optimal fitness for multiple consecutive generations is lower than the dynamic convergence threshold. The candidate configuration with the best fitness is then selected as the optimal microcapsule configuration. In each iteration, the rate of change between the current generation's optimal fitness and the previous generation's optimal fitness is calculated, i.e., the difference divided by the previous generation's optimal fitness. When the absolute value of the rate of change is lower than the dynamic convergence threshold for multiple consecutive generations, the optimization process is considered to have converged, and the iteration stops. At this point, the candidate configuration with the highest fitness is selected from all generations, and the mass ratio of epoxy resin to microcrystalline wax and the mass ratio of inorganic nanomaterials to graphene contained in it are the optimal microcapsule configuration.

[0046] The dynamic convergence threshold is an adaptive convergence criterion dynamically calculated based on the dispersion of the population fitness in the current iteration and the geometric mean rate of change of fitness over the last three generations. It represents the minimum rate of change critical value for determining convergence in the current population state, and is used to replace the fixed threshold to adaptively determine whether the iteration should terminate. When the absolute value of the rate of change of the optimal fitness for several consecutive generations is lower than this dynamic convergence threshold, it indicates that the population has stabilized and fitness no longer significantly improves. At this point, the optimization process is considered convergent, and the iteration stops.

[0047] Specifically, the calculation steps for the dynamic convergence threshold include: Starting from the third iteration, after each iteration, the optimal fitness value of each generation in the most recent three iterations is extracted, and the rate of change between the optimal fitness values ​​of two adjacent generations is calculated to obtain two rate of change values. Take the product of the two rate of change values ​​to obtain the rate of change product, and take the square root of the rate of change product to obtain the geometric mean rate of change; Extract the fitness values ​​of all candidate configurations in the current iteration, calculate the interquartile range of the fitness values, and multiply them by the geometric mean rate of change to obtain the dynamic convergence threshold.

[0048] First, starting from the third iteration, after each iteration, the optimal fitness value of each generation from the most recent three iterations is extracted, and the rate of change between the optimal fitness values ​​of two adjacent generations is calculated, resulting in two rate of change values. Since the optimal fitness values ​​from three consecutive iterations are needed to calculate two adjacent rate of change, thus providing sufficient data for the subsequent geometric mean calculation, this condition cannot be met in the first two iterations. Therefore, the calculation of the dynamic convergence threshold begins from the third iteration. For example, suppose the optimal fitness value of the previous generation is A, the optimal fitness value of the current generation is B, and the optimal fitness value of the generation before that is C. Then, the first rate of change is equal to the absolute value of the difference between A and C divided by C, and the second rate of change is equal to the absolute value of the difference between B and A divided by A.

[0049] The geometric mean rate of change is obtained by multiplying the two rate of change values ​​and taking the square root of the product. The geometric mean rate of change is equal to the square root of the product of the rate of change values. Specifically, this geometric mean rate of change comprehensively reflects the average magnitude of fitness change between the most recent two generations. Compared to the arithmetic mean, the geometric mean is more robust to extreme values.

[0050] Next, the fitness values ​​of all candidate configurations in the current iteration are extracted, the interquartile range of the fitness values ​​is calculated, and multiplied by the geometric mean rate of change to obtain the dynamic convergence threshold. The interquartile range is equal to the upper quartile minus the lower quartile. The upper quartile refers to the value at the 75th percentile after sorting all candidate configurations' fitness values ​​from smallest to largest, and the lower quartile refers to the value at the 25th percentile. The interquartile range reflects the dispersion of the population's fitness values; a higher dispersion indicates higher population diversity and that the optimization has not yet converged. Multiplying the interquartile range by the geometric mean rate of change yields the dynamic convergence threshold. This dynamic convergence threshold is positively correlated with both the population dispersion and the magnitude of fitness change: when the population still has high diversity or fitness is still changing significantly, the threshold is larger, and the iteration continues; when the population tends to stabilize and fitness changes gradually, the threshold is smaller, and the iteration terminates after the convergence condition is met.

[0051] In summary, by calculating the dynamic convergence threshold, the convergence state of the optimization process can be adaptively determined, avoiding premature termination or iterative redundancy caused by using a fixed threshold. For example, suppose the optimal fitness values ​​for the last three generations are 0.85, 0.87, and 0.88, respectively. The first rate of change equals the difference between 0.87 and 0.85 divided by 0.85, approximately 0.0235; the second rate of change equals the difference between 0.88 and 0.87 divided by 0.87, approximately 0.0115. The product of the two rates of change is approximately 0.000270, and the square root yields a geometric mean rate of change of approximately 0.0164. Let the upper quartile of the fitness values ​​for all candidate configurations in the current iteration be 0.75, the lower quartile be 0.63, and the interquartile range be 0.12. Multiplying the interquartile range of 0.12 by the geometric mean rate of change of 0.0164 yields a dynamic convergence threshold of approximately 0.00197. When the absolute value of the rate of change of the optimal fitness is below 0.00197 for five consecutive generations, the optimization process is considered converged, and iteration is stopped.

[0052] Finally, the fitness evaluation and iterative update are repeated until the rate of change of the optimal fitness is lower than the dynamic convergence threshold for several consecutive generations. The candidate configuration with the best fitness is then selected as the optimal microcapsule configuration. Here, "several consecutive generations" refers to the number of iterations in which the rate of change of the optimal fitness is continuously lower than the dynamic convergence threshold, as required for convergence. This number of consecutive generations is set based on a combination of stability requirements and computational efficiency in the optimization process; for example, it can be set to 5 consecutive generations. Finally, when the absolute value of the rate of change of the optimal fitness is lower than the corresponding dynamic convergence threshold for 5 consecutive generations, the optimization process is considered to have converged, iteration stops, and the candidate configuration with the highest fitness from all generations is selected as the optimal microcapsule configuration.

[0053] S40: Prepare primary microcapsules according to the optimal microcapsule configuration, and composite a dopamine photothermal modification layer on the surface of the primary microcapsules to obtain a photothermal self-healing microcapsule additive; Furthermore, primary microcapsules were prepared according to the aforementioned optimal microcapsule configuration, and a dopamine photothermal modification layer was composited onto the surface of the primary microcapsules to obtain a photothermal self-healing microcapsule additive. The optimal microcapsule configuration includes the optimal mass ratio of epoxy resin to microcrystalline wax and the optimal mass ratio of inorganic nanomaterials to graphene. This configuration was obtained through the aforementioned multi-objective optimization iterative inversion, which can balance self-healing response speed and wall material mechanical strength while meeting the target photothermal conversion efficiency constraint.

[0054] Specifically, the preparation process of the primary microcapsules is as follows: According to the mass ratio of epoxy resin to microcrystalline wax in the optimal microcapsule configuration, epoxy resin and microcrystalline wax are mixed and heated to above the melting point of microcrystalline wax, and dispersed evenly under mechanical stirring to obtain a core material premix; According to the mass ratio of inorganic nanomaterials to graphene in the optimal microcapsule configuration, inorganic nanomaterials, graphene and hexadecyltrimethylammonium bromide are dispersed in formamide, stirred and mixed evenly to obtain a wall material premix; The core material premix is ​​added to the wall material premix, and an aqueous solution of polyacrylamide is added, and primary microcapsules are formed through homogenization and emulsification.

[0055] Furthermore, a dopamine photothermal modification layer is composited onto the surface of the primary microcapsules. Although the wall material of the primary microcapsules contains graphene, which has a certain light absorption capacity, the effective area of ​​graphene exposed on the microcapsule surface after dispersion in the wall material is limited, and there is still considerable room for improvement in photothermal conversion efficiency. At the same time, the interfacial compatibility between the unmodified microcapsule surface and the epoxy resin matrix is ​​poor, and high addition levels can easily lead to a decrease in coating adhesion. Therefore, it is necessary to composite a dopamine photothermal modification layer onto the surface of the primary microcapsules. This dopamine photothermal modification layer, on the one hand, utilizes the broad-spectrum light absorption characteristics of polydopamine itself to form a synergistic photothermal effect with the graphene in the wall material, significantly improving the photothermal conversion efficiency of the microcapsules; on the other hand, the catechol groups of polydopamine can chemically react with the epoxy resin matrix or form hydrogen bonds, enhancing the interfacial adhesion between the microcapsules and the coating matrix, avoiding the degradation of the coating's mechanical properties caused by the addition of additives, thereby achieving a dual improvement in photothermal function and interfacial compatibility.

[0056] Specifically, primary microcapsules are prepared according to the optimal microcapsule configuration, and a dopamine photothermal modification layer is composited on the surface of the primary microcapsules to obtain a photothermal self-healing microcapsule additive, comprising: According to the mass ratio of epoxy resin to microcrystalline wax in the optimal microcapsule configuration, the epoxy resin and microcrystalline wax are mixed and heated to above the melting point of the microcrystalline wax. The mixture is then stirred and dispersed at a speed of 300 to 500 rpm to obtain the core material premix. According to the mass ratio of inorganic nanomaterials to graphene in the optimal microcapsule configuration, inorganic nanomaterials, graphene and hexadecyltrimethylammonium bromide are dispersed in formamide and stirred at a speed of 300 to 500 rpm to obtain a wall material premix. The core material premix is ​​added to the wall material premix, and a polyacrylamide aqueous solution with a concentration of 0.1% to 1.0% is added according to the collected polyacrylamide concentration. The mixture is then homogenized and emulsified at a speed of 5000 to 15000 rpm to form primary microcapsules. The primary microcapsules were dispersed in a dopamine hydrochloride solution with a concentration of 2 to 5 mg / mL, and the stirring time was controlled in the dark within the range of 4 to 12 hours according to the collected dopamine polymerization time. By adjusting the pH of the solution to 8.5 to 9.0, the dopamine is polymerized on the surface of the microcapsules to form a photothermal modified layer. After filtration and vacuum drying, a photothermal self-healing microcapsule additive is obtained.

[0057] The optimal microcapsule configuration comprises two key mass ratio parameters: the mass ratio of epoxy resin to microcrystalline wax and the mass ratio of inorganic nanomaterials to graphene. According to the core material composition parameters in this optimal microcapsule configuration, epoxy resin and microcrystalline wax are mixed at the optimal mass ratio, heated above the melting point of the microcrystalline wax to ensure complete melting, and then mechanically stirred and dispersed at a speed of 300 to 500 rpm to obtain a homogeneous core material premix.

[0058] Meanwhile, according to the wall material component parameters in the optimal configuration, inorganic nanomaterials and graphene are dispersed in formamide at the optimal mass ratio with hexadecyltrimethylammonium bromide, and stirred and mixed at a speed of 300 to 500 rpm to obtain a wall material premix.

[0059] Next, the core material premix is ​​added to the wall material premix, and a 0.1% to 1.0% polyacrylamide aqueous solution is added according to the previously collected polyacrylamide concentration. The mixture is then emulsified for 5 to 15 minutes using a homogenizing emulsifier at a speed of 5000 to 15000 rpm in a water bath at 40 to 70 degrees Celsius to form primary microcapsules. During this process, polyacrylamide, acting as a surface activity modifier, induces the controlled aggregation of inorganic nanoparticles and graphene on the core material surface through physical adsorption, self-assembling to form a multi-level rough structure combining micron-scale and nano-scale protrusions.

[0060] Next, the primary microcapsules were dispersed in a dopamine hydrochloride solution with a concentration of 2 to 5 mg / mL. The stirring time was controlled within the light-protected range of 4 to 12 hours, following the previously obtained dopamine polymerization time. By adjusting the solution pH to 8.5 to 9.0, dopamine underwent a self-polymerization reaction on the microcapsule surface, forming a polydopamine photothermal modified layer. This dopamine photothermal modified layer, in synergy with the micro / nano rough structure on the microcapsule surface, broadens the light absorption range through the π-π conjugation between the catechol groups of polydopamine and graphene, and increases the effective surface area for light absorption through the micro / nano rough structure, thereby significantly improving the photothermal conversion efficiency. Simultaneously, the catechol groups of polydopamine also enhance the interfacial compatibility between the microcapsules and the subsequent epoxy resin matrix. After filtration and vacuum drying, a photothermal self-healing microcapsule additive with a micro / nano rough surface and a composite dopamine photothermal modified layer was obtained.

[0061] Through the above steps, the resulting photothermal self-healing microcapsule additive possesses high photothermal conversion efficiency, rapid self-healing response, and good interfacial compatibility, providing a core functional filler for the subsequent preparation of anti-icing coatings. Specifically, this step transforms the theoretically optimal configuration obtained through optimization into a practically achievable microcapsule product, and further enhances the photothermal function and interfacial compatibility through dopamine surface modification, achieving an integrated function of photothermal de-icing and damage self-repair.

[0062] S50: The photothermal self-healing microcapsule additive is mixed with epoxy resin-based coating at an adaptive mass ratio, and the coating is applied and cured to obtain the anti-icing coating.

[0063] Finally, the photothermal self-healing microcapsule additive and epoxy resin-based coating were mixed at an adaptive mass ratio, coated, and cured to obtain an anti-icing coating. The photothermal self-healing microcapsule additive is the core functional filler prepared in the aforementioned steps, and its addition amount directly affects the anti-icing performance and mechanical properties of the coating: if the addition amount is too low, the photothermal conversion efficiency and self-healing ability are insufficient, making it difficult to meet the ice-melting requirements; if the addition amount is too high, the microcapsules may agglomerate in the coating, leading to a decrease in coating density and adhesion. Therefore, the mass ratio of the microcapsule additive to the epoxy resin-based coating needs to be adaptively determined based on the component parameters corresponding to the optimal microcapsule configuration and the process parameters during preparation.

[0064] Determining this adaptive mass ratio requires comprehensive consideration of multiple factors, including core material composition parameters, wall material composition parameters, polyacrylamide concentration, and dopamine polymerization time. The mass ratio of epoxy resin to microcrystalline wax in the core material determines the self-healing response speed and post-healing strength; the mass ratio of inorganic nanomaterials to graphene in the wall material determines the photothermal conversion efficiency and the mechanical properties of the wall material; the polyacrylamide concentration affects the density of the rough structure on the microcapsule surface; and the dopamine polymerization time affects the thickness and density of the photothermal modified layer.

[0065] Specifically, such as Figure 2 As shown, the steps for determining the adaptive mass ratio mixing include: The mass ratio of epoxy resin to microcrystalline wax in the optimal microcapsule configuration is extracted as the first factor, and the mass ratio of inorganic nanomaterials to graphene is extracted as the second factor. Calculate the geometric mean of the first factor and the second factor to obtain the component matching index; The harmonic mean of the polyacrylamide concentration and the dopamine polymerization time is calculated to obtain the process compatibility index; Based on the component matching index and the process coordination index, a comprehensive fit coefficient is calculated, wherein the comprehensive fit coefficient is directly proportional to the component matching index and inversely proportional to the process coordination index. The relative position of the comprehensive adaptation coefficient between the theoretical minimum and the theoretical maximum is linearly mapped to the lower and upper limits of the allowable mass ratio of microcapsule additives in epoxy resin-based coatings, and the mapping result is output as the adaptive mass ratio.

[0066] First, the mass ratio of epoxy resin to microcrystalline wax in the optimal microcapsule configuration was extracted as the primary factor, and the mass ratio of inorganic nanomaterials to graphene was extracted as the secondary factor. The mass ratio of epoxy resin to microcrystalline wax determines the self-healing performance of the core material; a higher ratio results in greater strength after repair but a slower response speed. The mass ratio of inorganic nanomaterials to graphene determines the photothermal absorption capacity and mechanical strength of the wall material; a higher graphene ratio results in better photothermal performance but a decrease in wall material strength. Using these two as core factors comprehensively reflects the formulation characteristics of the microcapsules.

[0067] Next, the geometric mean of the first and second factors is calculated to obtain the component matching index. The geometric mean is equal to the square root of the product of the first and second factors. The component matching index comprehensively reflects the degree of coordination between the core material and the wall material. A higher index indicates a more balanced formulation of the microcapsules in terms of both self-healing and photothermal dimensions, and a better expected overall performance.

[0068] Furthermore, the harmonic mean of polyacrylamide concentration and dopamine polymerization time was calculated to obtain the process coordination index. The harmonic mean is equal to 2 divided by (the reciprocal of the polyacrylamide concentration plus the reciprocal of the dopamine polymerization time). The polyacrylamide concentration controls the formation density of the micro / nano rough structures on the microcapsule surface, while the dopamine polymerization time controls the polymerization thickness of the photothermal modification layer. Together, they determine the morphology and photothermal properties of the microcapsules. Therefore, this process coordination index reflects the degree of matching between the two process parameters; a higher index indicates more coordinated process conditions and a more reasonable combination of microcapsule surface morphology and modification layer thickness.

[0069] Furthermore, based on the component matching index and the process coordination index, a comprehensive fit coefficient is calculated. The comprehensive fit coefficient is directly proportional to the component matching index and inversely proportional to the process coordination index. A higher component matching index indicates better formulation performance, allowing for a reduction in the amount of microcapsules added. A higher process coordination index indicates better process conditions and stronger photothermal properties of the microcapsules, also allowing for a reduction in the amount added. Specifically, the comprehensive fit coefficient can be calculated by dividing the component matching index by the process coordination index. A larger ratio indicates better microcapsule quality and a smaller required addition amount.

[0070] Secondly, the relative position of the comprehensive fit coefficient between the theoretical minimum and maximum values ​​is linearly mapped to the lower and upper limits of the allowable mass ratio range for microcapsule additives in epoxy resin-based coatings. The mapped result is then used as an adaptive mass ratio. The allowable mass ratio range is determined based on the critical aggregation concentration of microcapsules in epoxy resin-based coatings, for example, a lower limit of 5% and an upper limit of 20%. The closer the comprehensive fit coefficient is to the theoretical minimum, the worse the quality of the microcapsules themselves, and the closer the mapped result is to the upper limit of the mass ratio range, requiring more microcapsules to ensure coating performance. Conversely, the closer the comprehensive fit coefficient is to the theoretical maximum, the better the quality of the microcapsules themselves, and the closer the mapped result is to the lower limit of the mass ratio range, allowing for a reduction in the amount added to lower costs. Through adaptive mapping, the amount of microcapsules added can be dynamically adjusted according to the advantages and disadvantages of their own formulation and process conditions, achieving an optimal balance between coating performance and cost.

[0071] Specifically, the relative position of the comprehensive fit coefficient between the theoretical minimum and the theoretical maximum is linearly mapped to the lower and upper limits of the allowable mass ratio range for microcapsule additives in epoxy resin-based coatings, and the mapping result is output as an adaptive mass ratio, including: Based on the absolute deviation between the target photothermal conversion efficiency and the predicted photothermal conversion efficiency, the allowable mass ratio range of microcapsule additives for epoxy resin-based coatings is determined. The minimum value of the component matching index is calculated based on the minimum mass ratio of the core material component parameters and the minimum mass ratio of the wall material component parameters. The maximum value of the process coordination index is calculated based on the maximum value of polyacrylamide concentration and the maximum value of dopamine polymerization time. The minimum value of the component matching index is obtained by dividing the minimum value of the process coordination index by the maximum value of the component matching index. The maximum value of the component matching index is calculated based on the maximum mass ratio of the core material component parameters and the maximum mass ratio of the wall material component parameters. The minimum value of the process coordination index is calculated based on the minimum value of polyacrylamide concentration and the minimum value of dopamine polymerization time. The theoretical maximum value is obtained by dividing the maximum value of the component matching index by the minimum value of the process coordination index. Calculate the difference between the comprehensive adaptation coefficient and the theoretical minimum value, and divide it by the difference between the theoretical maximum value and the theoretical minimum value to obtain the normalized position parameter; Calculate the difference between the upper and lower limits of the mass ratio range, and multiply the difference by the normalized position parameter to obtain the mass ratio offset; The lower limit of the mass ratio range is added to the mass ratio offset, and the sum is output as the adaptive mass ratio.

[0072] First, based on the absolute deviation between the target photothermal conversion efficiency and the predicted photothermal conversion efficiency, the allowable mass ratio range of microcapsule additives for epoxy resin-based coatings is determined. The target photothermal conversion efficiency is the desired performance indicator, while the predicted photothermal conversion efficiency is the predicted value of the optimal configuration using a morphology-photothermal correlation model. The absolute deviation between the two reflects the gap between the optimal configuration and the desired performance; a larger deviation indicates that the performance of the microcapsules themselves is insufficient, requiring a wider range of addition amounts to ensure the overall performance of the coating.

[0073] Specifically, based on the absolute deviation between the target photothermal conversion efficiency and the predicted photothermal conversion efficiency, the allowable mass ratio range for adding microencapsulated additives to epoxy resin-based coatings is determined, including: Based on the critical agglomeration concentration of microcapsule additives in epoxy resin-based coatings, the amount of additive corresponding to the critical agglomeration concentration is taken as the maximum mass ratio, and the minimum mass ratio allowed during mixing is obtained. The ratio of the absolute deviation to the target photothermal conversion efficiency is calculated and used as the deviation coefficient. Calculate the difference between the maximum mass ratio and the minimum mass ratio, and multiply the difference by the deviation coefficient to obtain the range reduction amount; The maximum mass ratio minus the range reduction amount is used as the upper limit of the mass ratio range, and the minimum mass ratio is used as the lower limit of the mass ratio range.

[0074] First, based on the critical agglomeration concentration of microcapsule additives in epoxy resin-based coatings, the addition amount corresponding to the critical agglomeration concentration is taken as the maximum mass ratio, and the minimum mass ratio allowed during mixing is obtained.

[0075] The critical agglomeration concentration refers to the minimum amount of microencapsulated additive added to an epoxy resin-based coating at which agglomeration begins. Exceeding this concentration, the microcapsules aggregate, leading to decreased coating density and adhesion. The critical agglomeration concentration is determined by preparing a series of epoxy resin-based coating samples with progressively increasing mass fractions of microencapsulated additive, for example, samples with mass fractions of 5%, 8%, 10%, 12%, 15%, 18%, 20%, 22%, and 25%. After thorough mixing, the samples are allowed to stand or centrifuged, and the dispersion state of the additive is observed. The concentration at which significant agglomeration, precipitation, or a sudden change in coating viscosity begins to appear is the critical agglomeration concentration. This critical value can be determined experimentally; for example, significant particle agglomeration occurs in the coating when the addition exceeds 20%, therefore 20% is considered the maximum mass ratio.

[0076] The minimum mass ratio is the lowest addition amount required to ensure the coating possesses basic photothermal and self-healing functions. The minimum mass ratio is determined by preparing a series of epoxy resin-based coating samples with increasing mass fractions of microcapsule additives, for example, samples with mass fractions of 1%, 2%, 3%, 4%, 5%, and 6%, respectively. After coating and curing, observe whether the coating surface is continuous and whether there are pinholes or exposed substrate defects. The lowest addition amount that forms a complete, defect-free coating is taken as the minimum mass ratio. For example, if the coating surface is continuous and complete without pinholes at 5%, while exposed substrate occurs at 4%, then 5% is taken as the minimum mass ratio. Below this value, the photothermal conversion efficiency and self-healing ability are insufficient to meet the ice-melting requirements, and the integrity of the coating cannot be guaranteed.

[0077] Secondly, the ratio of the absolute deviation to the target photothermal conversion efficiency is calculated as the deviation coefficient. Assuming the target photothermal conversion efficiency is 85% and the predicted efficiency is 80%, the absolute deviation is 5%, and the deviation coefficient equals 5% divided by 85%, approximately 0.0588. The deviation coefficient reflects the relative gap between the optimal configuration and the desired performance. A larger deviation coefficient indicates that the microcapsule's own performance is insufficient, requiring a wider range of addition amounts to compensate.

[0078] Next, the difference between the maximum and minimum mass ratios is calculated, and this difference is multiplied by a deviation coefficient to obtain the range reduction. Assuming the maximum mass ratio is 20% and the minimum mass ratio is 5%, the difference is 15%. The range reduction equals 15% multiplied by 0.0588, approximately 0.88%. The range reduction reflects the degree to which the mass ratio range needs to be compressed due to the performance of the microcapsules themselves: a larger deviation coefficient indicates worse microcapsule performance, a smaller range reduction, and a wider mass ratio range, allowing for more addition to ensure performance; a smaller deviation coefficient indicates better microcapsule performance, a larger range reduction, and a narrower mass ratio range, allowing for less addition to reduce costs.

[0079] Furthermore, the maximum mass ratio minus the range reduction is used as the upper limit of the mass ratio range, and the minimum mass ratio is used as the lower limit. For example, subtracting the range reduction of 0.88% from the maximum mass ratio of 20% yields an upper limit of 19.12%, while the lower limit remains unchanged at 5%. Through this dynamic adjustment, the upper limit of the mass ratio range shrinks downward as the performance of the microcapsules improves, making the optimization space for the addition amount more concentrated in the low addition amount region. When the microcapsule performance fully reaches the target photothermal conversion efficiency, the absolute deviation is zero, the deviation coefficient is zero, and the range reduction is zero. The upper limit of the mass ratio range is the maximum mass ratio, and the addition amount can be optimized within the entire range of 5% to 20%. When the microcapsule performance is poor, the absolute deviation is large, the deviation coefficient is large, and the range reduction is small. The upper limit of the mass ratio range is close to the maximum mass ratio, allowing for a higher addition amount to ensure coating performance.

[0080] In summary, through this adaptive range determination mechanism, the mass ratio range can be dynamically adjusted according to the gap between the optimal configuration and the desired performance, providing a basis for the accurate calculation of the subsequent adaptive mass ratio.

[0081] Furthermore, based on the minimum mass ratios of the core material component parameters and the wall material component parameters, the minimum value of the component matching index is calculated. The minimum mass ratio of the core material component parameters refers to the minimum value within the range of the mass ratio of epoxy resin to microcrystalline wax, i.e., 1:1; the minimum mass ratio of the wall material component parameters refers to the minimum value within the range of the mass ratio of inorganic nanomaterials to graphene, i.e., 1:1. The component matching index is equal to the geometric mean of the first factor and the second factor. Therefore, the minimum value of the component matching index is equal to the square root of the product of the minimum mass ratio of the core material and the minimum mass ratio of the wall material, i.e., 1 multiplied by the square root of 1 equals 1.

[0082] The maximum process compatibility index is calculated based on the maximum polyacrylamide concentration and the maximum dopamine polymerization time. The maximum polyacrylamide concentration is 1.0%, and the maximum dopamine polymerization time is 12 hours. The process compatibility index equals 2 divided by (the reciprocal of the polyacrylamide concentration plus the reciprocal of the dopamine polymerization time). The process compatibility index is the maximum value when both concentration and time are at their maximum values.

[0083] Furthermore, the minimum value of the component matching index is divided by the maximum value of the process compatibility index to obtain the theoretical minimum value. This theoretical minimum value is used to reflect the lower limit of the overall compatibility coefficient under the worst formulation combination and the optimal process conditions.

[0084] Furthermore, based on the maximum mass ratio of the core material component parameters to the maximum mass ratio of the wall material component parameters, the maximum value of the component matching index is calculated. The maximum mass ratio of the core material component parameters is 3:1, and the maximum mass ratio of the wall material component parameters is 10:1. The maximum value of the component matching index is equal to the square root of the product of 3 and 10, that is, the square root of 30 is approximately equal to 5.48.

[0085] Simultaneously, the minimum process compatibility index is calculated based on the minimum polyacrylamide concentration and the minimum dopamine polymerization time. The minimum polyacrylamide concentration is 0.1%, and the minimum dopamine polymerization time is 4 hours. The process compatibility index is minimized when both concentration and time are at their minimum values.

[0086] Furthermore, dividing the maximum value of the component matching index by the minimum value of the process compatibility index yields the theoretical maximum value. The theoretical maximum value reflects the upper limit of the comprehensive compatibility coefficient under optimal formulation combination and worst process conditions.

[0087] Furthermore, the difference between the overall fitness coefficient and the theoretical minimum value is calculated, and then divided by the difference between the theoretical maximum value and the theoretical minimum value to obtain the normalized position parameter. This normalized position parameter reflects the relative position of the overall fitness coefficient of the current optimal configuration within the theoretical value range, and its value ranges from 0 to 1.

[0088] Secondly, the difference between the upper and lower limits of the mass ratio range is calculated, and this difference is multiplied by the aforementioned normalized position parameter to obtain the mass ratio offset. The closer the normalized position parameter is to 1, the better the quality of the microcapsules themselves; the larger the mass ratio offset, the closer the mapping result is to the lower limit of the mass ratio range, meaning a smaller addition amount is required. Conversely, the closer the normalized position parameter is to 0, the worse the quality of the microcapsules themselves; the smaller the mass ratio offset, the closer the mapping result is to the upper limit of the mass ratio range, meaning a larger addition amount is required.

[0089] Finally, the lower limit of the mass ratio range is added to the mass ratio offset, and the sum is output as the adaptive mass ratio. Through the above linear mapping, the amount of microcapsules added can be dynamically adjusted according to the advantages and disadvantages of their own formulation and process conditions: when the photothermal and self-healing properties of the microcapsules are excellent, the amount added is automatically reduced to reduce costs; when the performance is relatively insufficient, the amount added is automatically increased to ensure the overall performance of the coating. This adaptive mechanism achieves the optimal balance between coating performance and cost.

[0090] Finally, the photothermal self-healing microcapsule additive and epoxy resin-based coating are mixed at an adaptive mass ratio, coated, and cured to obtain an anti-icing coating. Specifically, the preparation of the coating includes the following steps: The first step involves mixing the photothermal self-healing microcapsule additive, determined according to an adaptive mass ratio, with the base epoxy resin. The amount of microcapsule additive added is 5% to 20% of the total mass of the epoxy resin-based coating. The mixture is stirred for 10 to 20 minutes at 300 to 500 rpm using a mechanical stirrer to ensure uniform dispersion of the microcapsule additive in the epoxy resin. Subsequently, a polyamide curing agent, defoamer, leveling agent, and other additives are added. The amount of polyamide curing agent added is calculated based on the epoxy value of the epoxy resin. The amounts of defoamer and leveling agent added are 0.3% to 0.8% and 0.2% to 0.5% of the total mass, respectively. The mixture is then stirred at a high speed of 500 to 800 rpm for 30 minutes to ensure thorough and uniform dispersion of all components, forming a homogeneous coating slurry.

[0091] The second step involves applying the mixed slurry to the substrate surface via spraying, brushing, or dipping. Taking spraying as an example, an air spraying process is used, with a nozzle diameter of 2.0 to 2.5 mm and an air pressure of 0.3 to 0.5 MPa. The wet film thickness is controlled according to the dry film thickness requirements. After coating, allow it to level at room temperature for 10 to 20 minutes to allow the coating surface to naturally smooth. Then, place the coated substrate in a forced-air oven and cure it at 50 to 100 degrees Celsius for 2 to 4 hours, allowing the epoxy resin and curing agent to fully cross-link and form a dense coating. The cured coating thickness is controlled within the range of 150 to 1500 micrometers, with the specific thickness determined according to the application requirements. For example, for the protection of power grid transmission and transformation equipment, the coating thickness can be set to 300 to 500 micrometers; for the protection of wind turbine blades, the coating thickness can be set to 500 to 1000 micrometers.

[0092] In summary, through the above coating and curing process, the photothermal self-healing microcapsule additive is uniformly dispersed in the epoxy resin coating. Under light irradiation, the micro-nano rough structure on the microcapsule surface and the dopamine photothermal modified layer synergistically absorb light energy and convert it into heat energy, causing the coating surface to rapidly heat up and melt ice. When the coating is mechanically damaged, the microcapsule wall material ruptures, releasing the epoxy resin and microcrystalline wax from the core material. The epoxy resin then undergoes a cross-linking reaction with moisture in the air or the curing agent on the substrate surface, while the microcrystalline wax melts simultaneously under photothermal temperature. Both work together to repair coating damage and restore the coating's integrity and protective function.

[0093] In summary, the embodiments of this application have at least the following technical effects: Compared to existing technologies, this invention first collects the core and wall material composition parameters of microcapsules and introduces polyacrylamide concentration and dopamine polymerization time as key process parameters, providing a multi-dimensional data foundation for subsequent morphology-photothermal correlation modeling. Secondly, using polyacrylamide concentration and dopamine polymerization time as inputs, a morphology-photothermal correlation model is constructed, enabling accurate prediction of the photothermal conversion efficiency of microcapsules and allowing evaluation of photothermal performance under different process conditions without repeated trial and error. Thirdly, with the target photothermal conversion efficiency as a constraint, multi-objective optimization is performed on the core and wall material composition parameters. The optimal microcapsule configuration is obtained through iterative inversion, enabling synergistic optimization of the micro / nano rough structure on the microcapsule surface and the photothermal modification layer, thereby improving photothermal conversion efficiency and self-healing response speed.

[0094] Finally, the present invention prepares microcapsule additives according to the optimal configuration, mixes them with epoxy resin-based coatings at an adaptive mass ratio, and obtains an anti-icing coating after coating and curing. This achieves the integrated function of rapid photothermal de-icing and self-repair of damage, solving the problems of slow photothermal response, performance degradation after damage, and difficulty in synergistic multi-functionality of traditional photothermal coatings.

Claims

1. A method for preparing a microcapsule-based photothermal self-healing anti-icing coating, characterized in that, The method includes: The core material and wall material parameters of the microcapsules were obtained, and the polyacrylamide concentration and dopamine polymerization time were collected. Using the polyacrylamide concentration and the dopamine polymerization time as inputs, a morphology-photothermal correlation model of the microcapsules is constructed, and the predicted photothermal conversion efficiency is obtained as the output. Based on the predicted photothermal conversion efficiency, and with the target photothermal conversion efficiency as a constraint, multi-objective optimization is performed on the core material composition parameters and wall material composition parameters, and the optimal microcapsule configuration is obtained through iterative inversion. Primary microcapsules were prepared according to the optimal microcapsule configuration, and a dopamine photothermal modification layer was composited on the surface of the primary microcapsules to obtain a photothermal self-healing microcapsule additive. The photothermal self-healing microcapsule additive is mixed with an epoxy resin-based coating at an adaptive mass ratio, and the coating is applied and cured to obtain the anti-icing coating.

2. The method for preparing a microcapsule-based photothermal self-healing anti-icing coating according to claim 1, characterized in that, The core material and wall material parameters of the microcapsules were obtained, and the polyacrylamide concentration and dopamine polymerization time were collected, including: The core material composition parameters are configured by mixing epoxy resin and microcrystalline wax in a mass ratio ranging from 1:1 to 3:

1. Inorganic nanomaterials and graphene were configured in a mass ratio ranging from 1:1 to 10:1 as the wall material composition parameters; The concentration of polyacrylamide was collected within the range of 0.1% to 1.0% based on the required density of micron-level protrusions on the surface of the microcapsules. The polymerization time of dopamine was collected in a dopamine solution of 2 to 5 mg / mL based on the required photothermal modified layer thickness on the surface of the microcapsules, wherein the polymerization time ranged from 4 to 12 hours.

3. The method for preparing a microcapsule-based photothermal self-healing anti-icing coating according to claim 1, characterized in that, The steps for constructing the morphology-photothermal correlation model of the microcapsules include: Multiple groups of microcapsule samples were collected under different polyacrylamide concentrations and different dopamine polymerization times. The actual photothermal conversion efficiency of each group of microcapsule samples was measured to obtain the sample parameter set and the corresponding photothermal conversion efficiency label. The polyacrylamide concentration and the dopamine polymerization time are encoded into an input feature vector; Construct a fully connected regression network architecture, wherein the fully connected regression network architecture includes an input layer, at least two hidden layers, and an output layer; Using the input feature vector as input and the photothermal conversion efficiency label as supervision signal, the fully connected regression network architecture is trained under supervision until the regression loss converges, thereby obtaining the morphology-photothermal correlation model.

4. The method for preparing a microcapsule-based photothermal self-healing anti-icing coating according to claim 1, characterized in that, Based on the predicted photothermal conversion efficiency, and with the target photothermal conversion efficiency as a constraint, multi-objective optimization is performed on the core material composition parameters and wall material composition parameters, and the optimal microcapsule configuration is obtained through iterative inversion, including: Multiple candidate configurations are randomly initialized within the mass ratio range of the core material component parameters and the mass ratio range of the wall material component parameters. Each candidate configuration includes a mass ratio of epoxy resin to microcrystalline wax and a mass ratio of inorganic nanomaterial to graphene. Each candidate configuration is input into the morphology-photothermal correlation model, and the predicted photothermal conversion efficiency corresponding to each candidate configuration is output. The fitness of multiple candidate configurations is evaluated by minimizing the deviation between the predicted photothermal conversion efficiency and the target photothermal conversion efficiency. Calculate the upper quartile of the fitness values ​​of all candidate configurations, retain candidate configurations with fitness values ​​greater than the upper quartile, and generate the next generation of candidate configurations through crossover and mutation operations; Repeatedly perform fitness evaluation and iterative updates until the rate of change of the optimal fitness for multiple consecutive generations is lower than the dynamic convergence threshold, and the candidate configuration with the best fitness is taken as the optimal microcapsule configuration.

5. The method for preparing a microcapsule-based photothermal self-healing anti-icing coating according to claim 4, characterized in that, Fitness was evaluated on multiple candidate configurations, using the minimization of the deviation between the predicted and target photothermal conversion efficiencies as the fitness criterion. This included: Calculate the absolute deviation between the predicted photothermal conversion efficiency and the target photothermal conversion efficiency for each candidate configuration, and calculate the mean and standard deviation of the absolute deviations for all candidate configurations. The ratio of the absolute deviation to the mean is used as the relative deviation coefficient, and the relative deviation coefficient is multiplied by the standard deviation to obtain the weighted deviation; The candidate configurations are sorted in ascending order of weighted bias, and each candidate configuration is assigned a fitness that decays exponentially according to its ranking. The fitness calculation steps include: Subtract one from the sorting position and take the negative value, which is used as the exponent of the exponent power; Using the natural constant as the base, calculate the value of the current exponent and use it as the fitness value of the current candidate configuration; The fitness value of each candidate configuration is calculated sequentially from 1 to N according to the sorting position, where N is the total number of candidate configurations.

6. The method for preparing a microcapsule-based photothermal self-healing anti-icing coating according to claim 4, characterized in that, The calculation steps for the dynamic convergence threshold include: Starting from the third iteration, after each iteration, the optimal fitness value of each generation in the most recent three iterations is extracted, and the rate of change between the optimal fitness values ​​of two adjacent generations is calculated to obtain two rate of change values. Take the product of the two rate of change values ​​to obtain the rate of change product, and take the square root of the rate of change product to obtain the geometric mean rate of change; Extract the fitness values ​​of all candidate configurations in the current iteration, calculate the interquartile range of the fitness values, and multiply them by the geometric mean rate of change to obtain the dynamic convergence threshold.

7. The method for preparing a microcapsule-based photothermal self-healing anti-icing coating according to claim 1, characterized in that, Primary microcapsules were prepared according to the optimal microcapsule configuration, and a dopamine photothermal modification layer was composited on the surface of the primary microcapsules to obtain a photothermal self-healing microcapsule additive, comprising: According to the mass ratio of epoxy resin to microcrystalline wax in the optimal microcapsule configuration, the epoxy resin and microcrystalline wax are mixed and heated to above the melting point of the microcrystalline wax. The mixture is then stirred and dispersed at a speed of 300 to 500 rpm to obtain the core material premix. According to the mass ratio of inorganic nanomaterials to graphene in the optimal microcapsule configuration, inorganic nanomaterials, graphene and hexadecyltrimethylammonium bromide are dispersed in formamide and stirred at a speed of 300 to 500 rpm to obtain a wall material premix. The core material premix is ​​added to the wall material premix, and a polyacrylamide aqueous solution with a concentration of 0.1% to 1.0% is added according to the collected polyacrylamide concentration. The mixture is then homogenized and emulsified at a speed of 5000 to 15000 rpm to form primary microcapsules. The primary microcapsules were dispersed in a dopamine hydrochloride solution with a concentration of 2 to 5 mg / mL, and the stirring time was controlled in the dark within the range of 4 to 12 hours according to the collected dopamine polymerization time. By adjusting the pH of the solution to 8.5 to 9.0, the dopamine is polymerized on the surface of the microcapsules to form a photothermal modified layer. After filtration and vacuum drying, a photothermal self-healing microcapsule additive is obtained.

8. The method for preparing a microcapsule-based photothermal self-healing anti-icing coating according to claim 1, characterized in that, The steps for determining the adaptive mass ratio mixing include: The mass ratio of epoxy resin to microcrystalline wax in the optimal microcapsule configuration is extracted as the first factor, and the mass ratio of inorganic nanomaterials to graphene is extracted as the second factor. Calculate the geometric mean of the first factor and the second factor to obtain the component matching index; The harmonic mean of the polyacrylamide concentration and the dopamine polymerization time is calculated to obtain the process compatibility index; Based on the component matching index and the process coordination index, a comprehensive fit coefficient is calculated, wherein the comprehensive fit coefficient is directly proportional to the component matching index and inversely proportional to the process coordination index. The relative position of the comprehensive adaptation coefficient between the theoretical minimum and the theoretical maximum is linearly mapped to the lower and upper limits of the allowable mass ratio of microcapsule additives in epoxy resin-based coatings, and the mapping result is output as the adaptive mass ratio.

9. The method for preparing a microcapsule-based photothermal self-healing anti-icing coating according to claim 8, characterized in that, The relative position of the comprehensive fit coefficient between the theoretical minimum and the theoretical maximum value is linearly mapped to the lower and upper limits of the allowable mass ratio range for microencapsulated additives in epoxy resin-based coatings. The mapping result is output as the adaptive mass ratio, including: Based on the absolute deviation between the target photothermal conversion efficiency and the predicted photothermal conversion efficiency, the allowable mass ratio range of microcapsule additives for epoxy resin-based coatings is determined. The minimum value of the component matching index is calculated based on the minimum mass ratio of the core material component parameters and the minimum mass ratio of the wall material component parameters. The maximum value of the process coordination index is calculated based on the maximum value of polyacrylamide concentration and the maximum value of dopamine polymerization time. The minimum value of the component matching index is obtained by dividing the minimum value of the process coordination index by the maximum value of the component matching index. The maximum value of the component matching index is calculated based on the maximum mass ratio of the core material component parameters and the maximum mass ratio of the wall material component parameters. The minimum value of the process coordination index is calculated based on the minimum value of polyacrylamide concentration and the minimum value of dopamine polymerization time. The theoretical maximum value is obtained by dividing the maximum value of the component matching index by the minimum value of the process coordination index. Calculate the difference between the comprehensive adaptation coefficient and the theoretical minimum value, and divide it by the difference between the theoretical maximum value and the theoretical minimum value to obtain the normalized position parameter; Calculate the difference between the upper and lower limits of the mass ratio range, and multiply the difference by the normalized position parameter to obtain the mass ratio offset; The lower limit of the mass ratio range is added to the mass ratio offset, and the sum is output as the adaptive mass ratio.

10. The method for preparing a microcapsule-based photothermal self-healing anti-icing coating according to claim 9, characterized in that, Based on the absolute deviation between the target photothermal conversion efficiency and the predicted photothermal conversion efficiency, the allowable mass ratio range for adding microencapsulated additives to epoxy resin-based coatings is determined, including: Based on the critical agglomeration concentration of microcapsule additives in epoxy resin-based coatings, the amount of additive corresponding to the critical agglomeration concentration is taken as the maximum mass ratio, and the minimum mass ratio allowed during mixing is obtained. The ratio of the absolute deviation to the target photothermal conversion efficiency is calculated and used as the deviation coefficient. Calculate the difference between the maximum mass ratio and the minimum mass ratio, and multiply the difference by the deviation coefficient to obtain the range reduction amount; The maximum mass ratio minus the range reduction amount is used as the upper limit of the mass ratio range, and the minimum mass ratio is used as the lower limit of the mass ratio range.

Images