Blue glass surface adaptive coating regulation method and system

Abstract

The application provides a blue glass surface adaptive coating regulation method and system, relates to the technical field of blue glass coating, and comprises the following steps: establishing a coating constraint condition; after obtaining a coating process, performing process analysis, configuring variable sensitivity and variable dependence of an optimization variable; after constructing an initial parameter set based on the optimization variable, performing fitness analysis of the initial parameter set through a comprehensive fitness function, and performing optimization update under the coating constraint condition based on the fitness analysis result, the variable sensitivity and the variable dependence; when the optimization update result meets a preset condition, stopping updating, and performing coating regulation through the final optimization result. Through the application, the technical problem that the film layer quality fluctuates greatly due to inaccurate control of process parameters in the prior art can be solved, the process parameters are optimized through the setting of the constraint condition and the comprehensive fitness function, and the overall stability of the film layer quality is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of blue glass coating, in particular to a blue glass surface adaptive coating regulation method and system. BACKGROUND

[0002] Blue glass coating generally refers to adding a blue filter layer or a reflective layer on the surface of the glass through coating technology, so that the glass presents a blue appearance or has specific functionality. By selecting different coating materials and controlling the process parameters of coating, the light transmittance, reflectivity, ultraviolet resistance and other properties of blue glass can be adjusted. However, the current coating process mostly relies on empirical adjustment and manual intervention, and it is difficult to monitor and accurately control multiple process parameters affecting the quality of the film layer in the production process in real time. Especially when facing equipment aging or environmental changes, adaptive adjustments cannot be made in time, resulting in inaccurate control of process parameters, and frequent problems of large fluctuations in film layer quality in the production process, thereby affecting the overall performance of the blue glass.

[0003] In summary, the prior art has the technical problem of large fluctuations in film layer quality due to inaccurate control of process parameters. SUMMARY

[0004] The purpose of the present application is to provide a blue glass surface adaptive coating regulation method and system to solve the technical problem of large fluctuations in film layer quality due to inaccurate control of process parameters in the prior art.

[0005] In view of the above problems, the present application provides a blue glass surface adaptive coating regulation method and system.

[0006] In a first aspect, the present application provides a blue glass surface adaptive coating regulation method, which is realized by a blue glass surface adaptive coating regulation system, wherein the blue glass surface adaptive coating regulation method comprises: establishing a coating constraint condition, the coating constraint condition is constructed by collecting the limit constraint of the film layer and the equipment parameters of the coating equipment, and the coating constraint condition includes the film layer hardness constraint, the film thickness allowable deviation constraint, and the coating equipment working range constraint; after obtaining the coating process, performing process analysis based on the coating process, configuring the variable sensitivity and variable dependence of the optimization variable; after constructing the initial parameter set based on the optimization variable, performing fitness analysis of the initial parameter set by a comprehensive fitness function, and performing optimization update under the coating constraint condition based on the fitness analysis result and the variable sensitivity and variable dependence; when the optimization update result meets the preset condition, stop updating, and perform coating regulation through the final optimization result.

[0007] In the second aspect, the present application also provides a blue glass surface adaptive coating control system for executing the blue glass surface adaptive coating control method as described in the first aspect, wherein the blue glass surface adaptive coating control system includes: a constraint construction module for establishing coating constraints, and the coating constraints are constructed by collecting the limit constraints of the film layer and the equipment parameters of the coating equipment, and the coating constraints include film layer hardness constraints, film thickness allowable deviation constraints, and coating equipment working range constraints; a process analysis module for performing process analysis based on the coating process after obtaining the coating process, and configuring the variable sensitivity and variable dependence of the optimization variables; a fitness analysis module for performing fitness analysis of the initial parameter set through a comprehensive fitness function after constructing the initial parameter set based on the optimization variables, and performing optimization update under the coating constraints based on the fitness analysis results and variable sensitivity and variable dependence; a coating control module for stopping the update when the optimization update result meets the preset conditions, and performing coating control through the final optimization result.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] By establishing coating constraint conditions, the coating constraint conditions are constructed by collecting the limit constraints of the film layer and the equipment parameters of the coating equipment, and the coating constraint conditions include the film layer hardness constraint, the film thickness allowable deviation constraint, and the coating equipment working range constraint; after obtaining the coating process, a process analysis is performed based on the coating process, and the variable sensitivity and variable dependency of the optimization variables are configured; after constructing the initial parameter set based on the optimization variables, the fitness analysis of the initial parameter set is performed through the comprehensive fitness function, and the optimization update is performed under the coating constraint conditions based on the fitness analysis results and the variable sensitivity and variable dependency; when the optimization update result meets the preset conditions, the update is stopped, and the coating is regulated by the final optimization result. Blue glass is generally classified as a kind of advanced inorganic non-metallic material, which is a special glass. After coating, blue glass can achieve optical functions such as sunlight control to meet different application requirements. By establishing coating constraints and conducting process analysis, an initial parameter set is constructed based on the optimization variables. The initial parameters are evaluated through a comprehensive fitness function to determine the pros and cons of the parameters. The initial parameters are optimized and updated until the results meet the preset target conditions. The update is then stopped and the coating is regulated based on the final optimization results. This improves the overall stability of the film quality, thereby improving the quality and long-term stability of the blue glass coating.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, which can be implemented in accordance with the contents of the description, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically listed below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in this application or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without any creative work.

[0012] Figure 1 This is a flow chart of the method for controlling the adaptive coating on the surface of blue glass in this application;

[0013] Figure 2 This is a schematic diagram of the structure of the blue glass surface adaptive coating control system of this application.

[0014] Explanation of the accompanying symbols: constraint condition construction module 11, process analysis module 12, fitness analysis module 13, coating control module 14. DETAILED DESCRIPTION

[0015] This application solves the technical problem in the prior art of large fluctuations in film quality due to imprecise control of process parameters by providing a method and system for adaptive coating control on blue glass surfaces. By establishing coating constraints, conducting process analysis, constructing an initial parameter set based on optimal variables, and evaluating the initial parameters through a comprehensive fitness function to determine the quality of the parameters, the initial parameters are optimized and updated until the results meet the preset target conditions. The update is then stopped and the coating is controlled based on the final optimized result. This improves the overall stability of the film quality, thereby enhancing the quality and long-term stability of the blue glass coating.

[0016] Below, the technical solutions in this application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited to the example embodiments described herein. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. It should also be noted that, for the convenience of description, only the parts related to this application, rather than all of them, are shown in the accompanying drawings.

[0017] For example, see the attached Figure 1 The present application provides a method for regulating and controlling an adaptive coating on a blue glass surface, wherein the method is applied to a blue glass surface adaptive coating regulating and controlling system, and the method specifically comprises the following steps:

[0018] S100: Establishing coating constraint conditions, wherein the coating constraint conditions are constructed by collecting the limit constraints of the film layer and the equipment parameters of the coating equipment. The coating constraint conditions include the film layer hardness constraint, the film thickness allowable deviation constraint, and the coating equipment working range constraint.

[0019] Specifically, through experimental testing, the ultimate constraint data of the film layer is collected, that is, the maximum or minimum physical or chemical conditions that the film layer can withstand in a specific application, such as the upper and lower limits of the hardness, thickness, durability, etc. of the film layer. The ultimate constraint of the film layer ensures that the film layer does not exceed the limits of its physical and chemical properties while meeting the design requirements. For example, the hardness of the film layer is measured by a hardness tester, and the thickness of the film layer is measured by an ellipsometer or a step meter. At the same time, the equipment parameters of the coating equipment are collected, that is, various operating parameters of the equipment used for coating. These parameters can be adjusted or fixed, and are usually obtained from the equipment manual or equipment control system, such as temperature, deposition rate, vacuum degree, atmosphere composition, operating voltage, etc.

[0020] Based on the collected data, coating constraints are established—those constraints that must be observed during the coating process—to ensure film quality and proper equipment operation. Based on the ultimate constraints on film hardness, a control range for film hardness is established. If the hardness is too low, the film may wear out during long-term use; if the hardness is too high, the film may become brittle, causing the glass to break. Based on the ultimate constraints on film thickness, constraints on allowable film thickness deviation are established to determine the maximum allowable deviation between the actual film thickness and the target thickness, ensuring precise control of film thickness. Based on the coating equipment parameters, operating range constraints are established for the coating equipment—those allowable ranges of parameters during operation, such as temperature range and pressure range, to ensure equipment safety and proper operation. By collecting the ultimate constraints on the film and the equipment parameters of the coating equipment, coating constraints are established to ensure that the film performance meets predetermined quality standards during the coating process while also ensuring the stable operation of the coating equipment.

[0021] S200: After obtaining the coating process, a process analysis is performed based on the coating process, and variable sensitivity and variable dependency of optimization variables are configured.

[0022] Specifically, through experimental data or existing process standards, the coating process flow is obtained, including the selection of coating materials, the setting of coating parameters (such as temperature, pressure, gas flow, etc.), the operation of the coating equipment, and basic requirements such as film thickness. The coating process refers to the specific coating process, including the selection of materials, equipment, operating procedures, etc., to ensure that the film performance reaches the expected target. After obtaining the coating process, a detailed process analysis of the coating process is carried out to evaluate the impact of various process parameters on the film performance. Process analysis refers to a detailed study of the various operations and parameters in the coating process and their impact on the final film quality, aiming to identify the key factors affecting the film quality and how these factors interact with each other.

[0023] Based on the results of the process analysis, variables with a significant impact on coating quality are selected to form the optimization variables. These optimization variables are adjustable parameters that significantly influence objectives such as film quality, thickness, and hardness. By adjusting these optimization variables, the optimal process conditions are found. Designed experiments or numerical simulations are used to adjust the value of each optimization variable, observing changes in film quality and measuring the hardness, thickness, and optical properties of the film for each combination. Based on the experimental results, the sensitivity of each optimization variable is calculated through local or global sensitivity analysis. Local sensitivity analysis analyzes the impact of changes in a single variable (such as deposition rate or temperature) near a specific point on changes in the objective function (such as film hardness or thickness). It focuses on the immediate response of variable changes within a small range and is typically calculated using partial derivatives. Global sensitivity analysis evaluates the impact of each variable on the process outcome by analyzing changes across the entire variable range. It considers interactions between variables and is suitable for situations with large variable changes or multi-variable interactions.

[0024] Small changes are made to the optimization variables near a specific point. The sensitivity of each variable is evaluated by measuring or calculating the change in the objective function (e.g., film hardness or thickness). The local sensitivity of the variable is calculated through partial derivatives. By varying the values ​​of the optimization variables across their entire range, the combined impact of these variables on film quality is evaluated. The variance contribution of each variable and their interactions to film hardness is calculated to assess which variables have the greatest impact on film hardness and determine the global sensitivity. Based on the sensitivity analysis results, the sensitivity coefficient of each variable is calculated, representing the impact of a unit change in each variable on the objective function.

[0025] Variable sensitivity refers to the degree to which a change in a variable affects the final result. Specifically, if a process parameter (variable) changes, how will the film quality (such as hardness, thickness, optical properties, etc.) change accordingly? If a variable has a significant impact on the result, it is considered highly sensitive, while if it does not, it is considered less sensitive. Variables with high sensitivity should be prioritized during optimization.

[0026] Variable dependency analysis identifies the relationships between multiple variables. If dependencies exist between two variables, these dependencies should be considered during the optimization process to avoid undesirable changes in other variables when adjusting one variable. Variable dependency refers to the relationships and influences between multiple variables—that is, changing one variable may affect changes in others. Analyzing variable dependency helps better understand the interactions between process variables and avoid conflicting or unattainable optimization goals during the optimization process.

[0027] By analyzing experimental data, we can determine the dependencies between variables. Specifically, we can analyze whether changes in one variable affect other variables and whether these variables exhibit a specific relationship. For example, in some cases, temperature and atmosphere composition may interact, meaning that changes in atmosphere composition have a more pronounced effect on film hardness as temperature increases. Using experimental data, we can construct a variable dependency model using a quadratic regression model to capture these interactions. When optimizing a process, we must consider not only the sensitivity of individual variables but also the dependencies between them. By adjusting these combinations of variables, we can achieve superior film quality.

[0028] By identifying the key variables that affect film quality, these variables can be accurately adjusted to ensure that the film quality reaches the optimal state. Through a comprehensive analysis of the sensitivity and dependence of the variables, the coating process is ensured to be more controllable, thereby improving the stability of the overall process and reducing the fluctuation of film quality caused by improper variable adjustment.

[0029] S300: After constructing an initial parameter set based on the optimization variables, a fitness analysis of the initial parameter set is performed through a comprehensive fitness function, and optimization updates are performed under coating constraints based on the fitness analysis results and variable sensitivity and variable dependency.

[0030] Specifically, based on the determined optimization variables, a reasonable range and initial value are set for each optimization variable to construct an initial parameter set. A fitness analysis is performed on the initial parameter set using a comprehensive fitness function to obtain fitness analysis results. The comprehensive fitness function is a mathematical function used to evaluate coating process performance, comprehensively considering multiple indicators such as film thickness, film adhesion, and optical properties. Based on the fitness analysis results, independent fitness sets are obtained, including film thickness fitness, film adhesion fitness, and optical performance fitness.

[0031] Trigger analysis is performed on independent fitness sets. If the fitness evaluation of certain parameter combinations is poor and the changed parameters have a significant impact on the film quality, the joint gene adjustment is triggered based on this evaluation mechanism. For example, the uneven film thickness of a certain combination may be due to uncoordinated changes in temperature and gas flow, which can be solved by adjusting the joint gene of these two variables in a coordinated manner.

[0032] Based on variable dependencies, analyze the inter-variable dependencies and determine which variables need to be adjusted together. Based on the results of the trigger analysis, determine which variable combinations should be adjusted in conjunction. Match them to obtain joint combination values; for any unmatched combinations, determine independent combination values. Combine the joint and independent combination values ​​and filter them together to identify the optimal combination, which is then configured as a joint gene. The joint gene will serve as the optimization target and be updated in the next round of optimization. Based on variable sensitivity analysis, determine the baseline adjustment amplitude. For applications with high sensitivity to film thickness, a larger adjustment amplitude can be set.

[0033] The target parameter in the initial parameter set is selected as the optimization target, and the other parameters in the initial parameter set are updated using the baseline adjustment amplitude. Next, adaptive mutation updates of the target parameter (for example, fine-tuning the film thickness) are performed, along with crossover mutation. Through these mutation and crossover processes, an updated parameter set is established for the next round of optimization updates. After each round of optimization updates, the new parameter set undergoes another fitness analysis to assess whether it has improved the film quality. If the overall fitness value continues to improve, the optimization process is effective.

[0034] Ensure that the entire optimization process satisfies coating constraints. Coating constraints refer to the limiting conditions that must be met during coating optimization, including film hardness, film thickness deviation, and equipment operating range. Optimal update refers to the process of improving the results by adjusting parameter combinations during the optimization process. After each update, the optimization algorithm re-evaluates fitness based on the new parameter values, gradually finding an optimal parameter set that meets the preset goals. By performing optimal updates based on fitness analysis results and variable sensitivity and dependency, coating parameters are optimized to improve coating quality and efficiency. Furthermore, by considering the dependencies between variables, a comprehensive understanding of the impact of each variable on coating quality is achieved, allowing for more accurate parameter adjustments.

[0035] S400: When the optimization update result meets the preset conditions, the update is stopped, and the coating is regulated based on the final optimization result.

[0036] Specifically, before starting the optimization process, a set of target performance indicators and corresponding thresholds are defined, including specific values ​​of film thickness, adhesion, optical properties, etc., to determine whether the optimization goal has been achieved. Preset conditions may include the fitness value reaching a certain threshold, the number of iterations reaching an upper limit, or other performance indicators reaching the expected target. The optimization process evaluates and adjusts parameters multiple times, with the goal of optimizing indicators such as film thickness, adhesion, and optical properties. In each round of optimization update, the fitness function calculates a comprehensive fitness value based on the performance of the current parameter set. When the optimization update result meets the preset conditions, the update process stops. For example, when the comprehensive fitness value reaches or approaches the preset target, it means that the quality of the film has met the requirements; or when the predetermined maximum number of optimization iterations is reached, but the fitness value fails to reach the target, the update will automatically stop and the current best result will be given; or, if the fitness value changes little in multiple rounds of iterations, it is considered that convergence has been reached and the update will stop.

[0037] After the update stops, the final optimization results will be used as control parameters for the coating process and applied to actual production coating control. Based on the final optimization results, the operating parameters of the coating equipment, such as the coating machine's operating temperature, gas flow rate, and deposition rate, are adjusted. The final results obtained through the optimization update ensure that all performance indicators in the coating process meet the design requirements and ensure the coating process achieves optimal performance in terms of film thickness, adhesion, optical properties, etc., thereby achieving the goal of improving product quality and production efficiency.

[0038] Furthermore, the present application S300 includes:

[0039] The comprehensive fitness function is as follows: ;in, Represents the fitness value, is the film thickness weight, Characterize the film thickness fitness function, , Characterize the total number of measurement points, Characterize any measurement point, Characterization The thickness value of each measuring point, Characterizes the average thickness of the film layer, Characterize the film adhesion weight, Characterize the film adhesion fitness function, , Characterize the film adhesion test value, Characterize the best adhesion value, Characterize the optical performance weight, Characterize the optical performance fitness function, , is the transmittance weight, is the transmittance, is the reflection weight, is the reflectivity.

[0040] Specifically, the comprehensive fitness function is to quantify the comprehensive effect of multiple optimization objectives, so that different film properties (such as film thickness, adhesion, optical properties, etc.) can be comprehensively considered and process parameters can be adjusted according to these properties to achieve the best effect. Using a film thickness measuring instrument, the thickness of the film is measured at multiple points (such as the four corners and center of the film), and the average thickness of the film is calculated. According to the formula Calculate the difference between the film thickness at multiple points and the average thickness to obtain the film thickness fitness function, where: Characterize the total number of measurement points, Characterize any measurement point, Characterization The thickness value of each measuring point, Characterizes the average thickness of the film.

[0041] The adhesion of the film layer is tested by an adhesion tester, and an optimal adhesion value is pre-set based on experience or literature data. By measuring the adhesion of the film layer and comparing it with the optimal adhesion value, according to the formula The film adhesion fitness function is calculated, where Characterize the film adhesion test value, Characterizes the optimal adhesion value. The closer the film adhesion test value is to the optimal adhesion value, that is, the closer the film adhesion fitness function is to 1, the better the film adhesion.

[0042] The transmittance T and reflectance R of the film are measured by optical measuring instruments. The transmittance is the proportion of light allowed to pass through the film, and the reflectance is the proportion of light reflected by the film. According to the formula The optical performance fitness function is calculated, where is the transmittance weight, is the transmittance, is the reflection weight, For example, assuming the transmittance is 80%, the reflectance is 10%, the transmittance weight is 0.7, and the reflectance weight is 0.3, the calculated optical performance fitness is 59%.

[0043] is the film thickness weight, Characterize the film adhesion weight, Characterize the weights of optical properties and indicate how much they affect the final fitness. Characterize the film thickness fitness function, Characterize the film adhesion fitness function, The fitness function of the optical performance is characterized according to the formula The comprehensive fitness function is calculated to represent the overall effect of the current process parameter combination on the film thickness, adhesion and optical performance.

[0044] If the fitness evaluation of some parameter combination is poor, and the changed parameter has a greater impact on the film quality, the adjustment of the joint gene is triggered according to this evaluation mechanism, for example, the film thickness of the current combination is uneven, which may be due to the uncoordinated change of temperature and gas flow, and the joint gene of the two variables can be adjusted to solve the problem. Through multiple experiments and calculations, the parameters are gradually adjusted until a satisfactory fitness value is reached. The fitness analysis of the initial parameter set is performed by the comprehensive fitness function, the performance of each parameter combination is quantified and compared, and the film quality is evaluated and optimized in all directions.

[0045] Further, the application further comprises the following steps:

[0046] The fitness analysis result is analyzed to obtain an independent fitness set, the independent fitness set including film thickness fitness, film adhesion fitness and optical performance fitness; the fitness trigger analysis is performed on the independent fitness set to establish a trigger analysis result; the variable linkage combination is performed according to the trigger analysis result to configure a joint gene, and the next round of optimization update is performed based on the joint gene.

[0047] Specifically, the fitness analysis of the initial parameter set is performed by the comprehensive fitness function to obtain the fitness analysis result, and the overall fitness value is divided into an independent fitness set, each part reflecting different quality indicators of the film, including film thickness fitness, film adhesion fitness and optical performance fitness. The film thickness fitness measures whether the film thickness distribution is uniform, and the film thickness is too thin or too thick, which will affect the performance of the film and result in a lower fitness value; the film adhesion fitness reflects the adhesion strength between the film and the substrate, and the low adhesion may cause the film to fall off, thereby affecting its performance; the optical performance fitness is used to evaluate the optical performance of the film, and the higher the fitness value, the better the optical performance of the film.

[0048] By observing the changes in individual fitness values, fitness trigger analysis is performed on independent fitness sets to determine which indicators have reached the trigger point. By setting thresholds for each independent fitness value, it is determined when that fitness value triggers adjustments to other variables. Each fitness value is analyzed to see if it falls below its threshold, and the parameter combinations that require adjustment are recorded. Based on the analysis results, a trigger record is created to indicate which parameter combinations fail to meet the criteria. For example, suppose the independent fitness set includes a film thickness fitness of 0.012, an adhesion fitness of 0.85, and an optical performance fitness of 0.68, corresponding to a temperature of 320°C and a gas flow rate of 60 sccm. If the film thickness fitness trigger point is set to greater than 0.015, the adhesion fitness trigger point is equal to 1, and the optical performance fitness trigger point is less than 0.65, the film thickness fitness is below the threshold of 0.015, indicating uneven film thickness. Adjustments to the temperature and gas flow rate are made, lowering the temperature to 310°C and adjusting the gas flow rate to 55 sccm. Low adhesion fitness may be due to poor base cleanliness and gas flow.

[0049] Based on the trigger analysis results, variable linkage is performed. Based on the dependencies between different parameters, variables with significant impact on film quality are selected for linkage adjustment. Variable linkage refers to determining which variables require joint adjustment during the optimization process based on their mutual influence. For example, if the trigger analysis results indicate that temperature and gas flow rate have a strong influence on film thickness, the decision to jointly adjust these two variables may be made.

[0050] Based on the results of the variable linkage combination, an appropriate parameter combination is selected as a joint gene. This joint gene becomes one of the initial solutions in the genetic algorithm and participates in the next round of optimization updates. A joint gene is a combination of multiple related parameters in the genetic algorithm. Through the combination and optimization of genes, an optimal parameter set can be found to achieve optimal film performance.

[0051] The resulting joint gene is used for the next round of optimization updates. Adjustment ranges are determined based on variable sensitivity. Based on the fitness analysis results, target parameters are determined from the initial parameter set. Genetic operations such as crossover and mutation are used to generate new parameter combinations to further optimize film quality. Based on the joint gene, the adjusted parameter values ​​are used in the next round of optimization. Through fitness-triggered analysis and variable linkage, film thickness, adhesion, and optical properties are optimized to ensure that film quality meets predetermined requirements. The optimal coating parameter combination is determined to improve coating quality and efficiency.

[0052] Furthermore, the present application further comprises the following steps:

[0053] A variable combination pair set is obtained according to the variable dependency screening; linkage combination matching of the variable combination pair set is performed according to the trigger analysis result, and a joint combination value is established according to the linkage combination matching result; independent combination values ​​are established for unmatched variable combination pairs in the variable combination pair set; combination screening is performed using the joint combination value and the independent combination value, and a joint gene is configured according to the combination screening result.

[0054] Specifically, based on the variable dependencies of the optimization variables, we determine which variables need to be adjusted together. Based on the strength of the dependencies, we then identify variable combinations. Based on these determined variable combinations, we construct a set of variable combinations that includes all variable pairs that need to be adjusted together. For example, based on the variable dependencies, we screened out variable influence pairs. The correlation coefficient between pressure and deposition rate was 0.78, the correlation coefficient between substrate cleanliness and adhesion was 0.50, and the correlation coefficient between temperature and gas flow rate was 0.85.

[0055] Based on the results of the trigger analysis, the variables in the variable combination pair set are matched in a linked combination to determine which variable combinations should be adjusted together to optimize the process. Linked combination matching refers to adjusting the variable combinations based on the results of the trigger analysis so that they match within the optimal working range. Rules for linked combination matching are established. If the fitness of a variable combination pair is poor (such as large film thickness deviation) and the dependence between the variables is high, linked optimization is performed; if the fitness of a variable combination pair is good, no adjustment is made; combinations with a small impact on other variables but still requiring optimization are optimized separately and do not participate in the linked combination. For example, the current fitness between temperature and gas flow is 0.68, and the target fitness is set to 0.75. Therefore, the variable combination temperature and gas flow needs to be optimized.

[0056] A joint combination value is established based on the results of the linkage combination matching. That is, if a variable triggers the optimization condition, the variables that depend on it are also taken into account to form a joint combination value. For unmatched variable combination pairs in the variable combination pair set, independent combination values ​​are established, indicating the variable combinations that can be considered independently during the optimization process. If certain variable combinations are not identified as objects requiring linkage optimization by the trigger analysis, they will be adjusted separately and form independent combination values. For example, if a variable (such as substrate rotation speed) has a small impact on other variables, its parameters can be optimized separately without affecting the overall process.

[0057] Combination screening is performed using joint and independent combination values, calculating the fitness of each combination. Crossover, selection, and mutation are then used to perform combination screening. Combination screening involves selecting the optimal set of parameter combinations from multiple variable combinations to achieve the optimal optimization objective (such as film thickness uniformity, adhesion, and optical performance). Combinations with higher fitness are prioritized, and some variables are swapped between different combinations, with some parameters randomly adjusted to increase diversity. After crossover, selection, and mutation, multiple candidate parameter sets are generated, and the one with the highest fitness is selected as the joint gene. The joint gene is the optimal set of parameters formed during the optimization process to ensure the optimal combination of variables. The role of the joint gene is to mimic biological genetic mechanisms, allowing variables to continuously evolve during the optimization process, ultimately achieving the optimal solution. The optimal variable combination values ​​are then applied to the optimization algorithm for the next round of updates.

[0058] For example, in the film coating process, substrate temperature T, sputtering power P, and gas flow rate F are interrelated variables. Their combination must meet certain process conditions and, after optimization, form an optimal set of parameters. Substrate rotation speed R has a minimal impact on film thickness and can therefore be optimized independently. After crossover, selection, and mutation, multiple candidate parameter sets were obtained: Combination 1: sputtering power 150W, temperature 290°C, gas flow rate 15sscm, substrate rotation speed 25rpm, with a fitness of 0.89; Combination 2: sputtering power 155W, temperature 292°C, gas flow rate 17sscm, substrate rotation speed 30rpm, with a fitness of 0.92; and sputtering power 158W, temperature 295°C, gas flow rate 18sscm, substrate rotation speed 35rpm, with a fitness of 0.91. Combination 2 has the highest fitness and is therefore selected as the joint gene.

[0059] By matching and screening the linkage combinations of sets through variable combinations, considering the interactions between variables and taking these variables into account simultaneously in the optimization process, it is helpful to determine the best combination of coating parameters, improve the accuracy of process parameter optimization, and thus improve the coating quality and efficiency.

[0060] Furthermore, the present application further comprises the following steps:

[0061] A benchmark adjustment amplitude is configured based on the variable sensitivity; a target parameter in the initial parameter set is selected according to the fitness analysis result, the target parameter is used as the optimization target, and the remaining parameters in the initial parameter set are updated based on the benchmark adjustment amplitude; an adaptive mutation update of the target parameter is performed, and after cross-mutating all the update results, an updated parameter set is established to complete a round of optimization update.

[0062] Specifically, a baseline adjustment amplitude, or the step size for adjusting parameters during the optimization process, is determined based on the sensitivity of the variables. For variables with higher sensitivities, a larger adjustment amplitude is set to allow for greater adjustments to these parameters during the optimization process; for variables with lower sensitivities, a smaller adjustment amplitude is set. Based on the fitness analysis results, the parameter with the greatest impact on the process outcome is selected as the target parameter, which is the primary parameter to be adjusted. Based on the baseline adjustment amplitude, the remaining parameters in the initial parameter set are updated, adjusting their values ​​based on each parameter's sensitivity and the baseline adjustment amplitude to better optimize the target parameter. For example, if the target parameter is temperature, parameters such as gas flow rate and deposition rate are adjusted based on the baseline adjustment amplitude.

[0063] Adaptive mutation of the target parameters dynamically adjusts the mutation amplitude based on the current state of the target parameters and the optimization progress. Adaptive mutation update is an optimization strategy that adapts the parameter adjustment amplitude to the optimization process. For example, if the target parameter performed well in the previous optimization round, the update amplitude can be reduced to avoid over-optimization; conversely, if the performance was poor, the update amplitude can be increased to speed up optimization.

[0064] Crossover mutation is performed on all update results, combining the update results of different parameters to generate new parameter combinations. Crossover involves exchanging some information between two sets of parameters (parents) to generate new parameter combinations. The purpose of crossover is to combine two excellent solutions to produce a potentially superior solution. Mutation involves making small, random adjustments to a solution to increase the diversity of the search space and avoid local optimal solutions. The specific process involves selecting multiple optimized parameter sets as parents for crossover. After the crossover operation, the parameters generated by the crossover are mutated to increase the diversity of solutions. The range of mutation is adjusted according to the adaptive mutation strategy to ensure that the mutation does not deviate from the reasonable range.

[0065] After cross-mutation, a new set of parameter combinations is obtained, including the optimized solution, new target parameter values ​​(such as temperature, gas flow, and power), and the parameters adjusted after mutation. The cross-mutation results are combined to form a new parameter set, known as the updated parameter set, completing one round of update optimization and used for the next round of optimization. The new parameter set will result in higher film quality, gradually bringing film performance closer to the preset target.

[0066] Through the combination of adaptive variation and cross variation, the quality of the film layer has been significantly improved, and indicators such as film thickness uniformity, adhesion and optical properties have been optimized. The combination of variation amplitude and cross operation not only accelerates the optimization process, but also ensures diversity and avoids falling into local optimal solutions.

[0067] Furthermore, the present application S400 includes:

[0068] The film coating monitoring is performed to establish a monitoring data set. Based on the monitoring data set, a matching analysis of the actual film coating effect and the optimization result is performed to establish a matching error. A backtracking identifier is established according to the matching error, and the film coating process backtracking self-check is performed by using the backtracking identifier to establish a backtracking self-check result. The film coating process control compensation is performed based on the backtracking self-check result.

[0069] Specifically, during the film coating process, the key parameters and performance indicators are monitored in real time or periodically. Data during the film coating process is collected through various sensors and measuring instruments to obtain a monitoring data set, i.e., various data during the film coating process, which usually includes a data set of indicators such as the thickness, adhesion, light transmittance, and reflectivity of the film layer. Through film coating monitoring, the changes in the film layer are continuously tracked, and parameters related to the quality of the film layer are recorded.

[0070] The actual result in the monitoring data set is compared with the optimized optimization result to evaluate the gap between them. The matching error includes thickness error (the gap between the actual thickness and the target thickness), adhesion error (the gap between the actual adhesion and the target adhesion), optical performance error (the gap between the light transmittance and reflectivity and the target value), etc.

[0071] According to the matching error, a backtracking identifier is established for data with large errors to indicate which links or parameters have problems. The backtracking identifier is an identifier generated according to the matching error analysis during the film coating process, which is used to backtrack the key parameters and steps in the process. Through the film coating process backtracking self-check by using the backtracking identifier, the specific film coating step or parameter setting is backtracked to check whether there is a process link or control parameter deviation, including checking the equipment state, operation steps, environmental conditions, etc., which leads to the film layer quality not meeting the expected value. According to the problems and their causes determined by the backtracking self-check, a backtracking self-check result is generated. For example, if there is an error in the film layer thickness, the backtracking analysis will indicate that it is caused by the gas flow, temperature, or deposition rate parameter deviation.

[0072] According to the backtracking self-check result, the process parameters affecting the film layer quality are adjusted, such as changing the deposition rate, adjusting the temperature, etc., to optimize the film coating effect. Through the film coating process control compensation, after the process error is found, measures are taken to adjust or compensate to ensure that the film layer quality in the subsequent film coating process meets the expected target. Through the matching analysis of the actual film coating effect and the optimization result, any process deviation can be found and corrected to determine the root cause of the potential problem, thereby optimizing the control parameters to ensure the stability of the film layer quality.

[0073] Further, the present application further includes the following steps:

[0074] A self-check cycle is established, a coating effect self-check is performed during the self-check cycle, and a cycle deviation is established; a cycle compensation is generated using the cycle deviation, and self-optimization management of the coating process is performed based on the cycle compensation.

[0075] Specifically, a self-test cycle—the interval between self-tests—is determined based on the stability of the coating process and the length of the production cycle. This is used to regularly check the coating results. A process check is performed at the end of each cycle to ensure that the film quality meets expectations. During each self-test cycle, online inspection equipment and various sensors (such as thickness gauges, adhesion testers, and optical sensors) are used to collect data on key parameters and performance indicators during the coating process. This data is then compared with expected values ​​to calculate cycle deviation. Cycle deviation typically refers to the difference between the target values ​​for film thickness, adhesion, optical properties, and other parameters.

[0076] Based on the cycle deviation analysis results, cycle compensation is generated to adjust the coating process to compensate for the deviation and bring the subsequent film quality closer to the target. This may include adjusting process parameters (such as deposition rate, temperature, and pressure) or equipment maintenance (such as cleaning and calibration). By regularly monitoring cycle deviations and dynamically adjusting key parameters in the coating process (such as temperature, gas flow, and pressure), the process is self-optimized to ensure that the film quality continues to meet requirements.

[0077] The specific implementation process is as follows: Based on the cycle deviation within each self-inspection cycle, cycle compensation is generated by adjusting key process parameters. For example, if the film thickness deviation is large, the gas flow rate or temperature may need to be adjusted; if the transmittance deviation is large, the material evaporation rate or reflectivity may need to be adjusted. After the cycle compensation is implemented, the next round of production will be carried out based on the compensated and adjusted process parameters. After each cycle, the process parameter adjustment will take effect in real time and will be used in subsequent coating production. Through continuous self-inspection and cycle compensation, self-optimization management of the coating process is achieved, so that the production process is gradually stabilized, the film quality is continuously improved, and it can cope with the impact of changes in equipment status, raw material differences, etc.

[0078] Periodic self-inspection and deviation analysis can promptly identify process issues and reduce film quality fluctuations caused by external factors (such as equipment aging and material batch changes), thereby maintaining stable production quality. Through the implementation of self-inspection cycles and periodic compensation mechanisms, the coating process can be continuously optimized, the stability of film quality can be improved, and abnormal fluctuations in the production process can be reduced, thus ensuring that film quality continues to meet standards and improving production efficiency.

[0079] In summary, the method for controlling the adaptive coating on the blue glass surface provided by this application has the following technical effects:

[0080] By establishing coating constraint conditions, the coating constraint conditions are constructed by collecting the limit constraints of the film layer and the equipment parameters of the coating equipment, and the coating constraint conditions include the film layer hardness constraint, the film thickness allowable deviation constraint, and the coating equipment working range constraint; after obtaining the coating process, a process analysis is performed based on the coating process, and the variable sensitivity and variable dependency of the optimization variables are configured; after constructing the initial parameter set based on the optimization variables, the fitness analysis of the initial parameter set is performed through the comprehensive fitness function, and the optimization update is performed under the coating constraint conditions based on the fitness analysis results and the variable sensitivity and variable dependency; when the optimization update result meets the preset conditions, the update is stopped, and the coating is regulated by the final optimization result. Blue glass is generally classified as a kind of advanced inorganic non-metallic material, which is a special glass. After coating, blue glass can achieve optical functions such as sunlight control to meet different application requirements. By establishing coating constraints and conducting process analysis, an initial parameter set is constructed based on the optimization variables. The initial parameters are evaluated through a comprehensive fitness function to determine the pros and cons of the parameters. The initial parameters are optimized and updated until the results meet the preset target conditions. The update is then stopped and the coating is regulated based on the final optimization results. This improves the overall stability of the film quality, thereby improving the quality and long-term stability of the blue glass coating.

[0081] Example 2: Based on the same inventive concept as the blue glass surface adaptive coating control method in the above-mentioned Example 1, this application also provides a blue glass surface adaptive coating control system, please refer to the attached Figure 2 , the blue glass surface adaptive coating control system includes:

[0082] A constraint condition construction module 11 is used to establish coating constraint conditions, and the coating constraint conditions are constructed by collecting the limit constraints of the film layer and the equipment parameters of the coating equipment. The coating constraint conditions include film layer hardness constraint, film thickness allowable deviation constraint, and coating equipment working range constraint; a process analysis module 12 is used to perform process analysis based on the coating process after obtaining the coating process, and configure the variable sensitivity and variable dependence of the optimization variables; a fitness analysis module 13 is used to perform fitness analysis of the initial parameter set through a comprehensive fitness function after constructing the initial parameter set based on the optimization variables, and perform optimization update under the coating constraint conditions based on the fitness analysis results and variable sensitivity and variable dependence; a coating control module 14 is used to stop updating when the optimization update result meets the preset conditions, and perform coating control based on the final optimization result.

[0083] Furthermore, the fitness analysis module 13 in the blue glass surface adaptive coating control system is also used to:

[0084] The comprehensive fitness function is as follows: ;in, characterize a fitness value, is a film layer thickness weight, characterize a film layer thickness fitness function, , characterize a total number of measurement points, characterize an arbitrary measurement point, characterize a thickness value of the characterize a thickness value of the characterize a film layer thickness average value, characterize a film layer adhesion weight, characterize a film layer adhesion fitness function, , characterize a film layer adhesion test value, characterize an optimal adhesion value, characterize an optical performance weight, characterize an optical performance fitness function, , is a light transmission weight, is a light transmission rate, is a reflection weight, is a reflectivity.

[0085] Further, the fitness analysis module 13 in the blue glass surface adaptive coating regulation system is further used to:

[0086] analyze the fitness analysis result to obtain an independent fitness set, the independent fitness set including a film layer thickness fitness, a film layer adhesion fitness, and an optical performance fitness; perform fitness trigger analysis on the independent fitness set to establish a trigger analysis result; perform variable linkage combination according to the trigger analysis result to configure a joint gene, and perform next round optimization update based on the joint gene.

[0087] Further, the fitness analysis module 13 in the blue glass surface adaptive coating regulation system is further used to:

[0088] obtain a variable combination pair set according to the variable dependence screening; perform linkage combination matching of the variable combination pair set according to the trigger analysis result, establish a joint combination value according to the linkage combination matching result; establish an independent combination value for a variable combination pair that is not matched in the variable combination pair set; perform combination screening using the joint combination value and the independent combination value, and configure a joint gene according to the combination screening result.

[0089] Further, the fitness analysis module 13 in the blue glass surface adaptive coating regulation system is further used to:

[0090] A benchmark adjustment amplitude is configured based on the variable sensitivity; a target parameter in the initial parameter set is selected according to the fitness analysis result, the target parameter is used as the optimization target, and the remaining parameters in the initial parameter set are updated based on the benchmark adjustment amplitude; an adaptive mutation update of the target parameter is performed, and after cross-mutating all the update results, an updated parameter set is established to complete a round of optimization update.

[0091] Furthermore, the coating control module 14 in the blue glass surface adaptive coating control system is also used to:

[0092] Perform coating monitoring and establish a monitoring data set; perform matching analysis between the actual coating effect and the optimization result based on the monitoring data set to establish a matching error; establish a retrospective identifier based on the matching error, use the retrospective identifier to perform retrospective self-inspection of the coating process, and establish a retrospective self-inspection result; perform coating process control compensation based on the retrospective self-inspection result.

[0093] Furthermore, the blue glass surface adaptive coating control system further includes a self-checking and compensation module, which is further used to:

[0094] A self-check cycle is established, a coating effect self-check is performed during the self-check cycle, and a cycle deviation is established; a cycle compensation is generated using the cycle deviation, and self-optimization management of the coating process is performed based on the cycle compensation.

[0095] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The blue glass surface adaptive coating control method and specific examples in Example 1 are also applicable to the blue glass surface adaptive coating control system in this embodiment. Through the detailed description of the blue glass surface adaptive coating control method above, those skilled in the art can clearly understand the blue glass surface adaptive coating control system in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here. As for the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple. For relevant details, please refer to the method description.

[0096] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

[0097] Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A method for controlling adaptive coating on a blue glass surface, characterized in that: include: Establishing coating constraint conditions, wherein the coating constraint conditions are constructed by collecting the limit constraints of the film layer and the equipment parameters of the coating equipment, and the coating constraint conditions include the film hardness constraint, the film thickness allowable deviation constraint, and the coating equipment working range constraint; After obtaining the coating process, a process analysis is performed based on the coating process, and variable sensitivity and variable dependency of the optimization variables are configured; After constructing the initial parameter set based on the optimization variables, the fitness analysis of the initial parameter set is performed through the comprehensive fitness function, and the optimization update is performed under the coating constraint conditions based on the fitness analysis results and variable sensitivity and variable dependency; When the optimization update result meets the preset conditions, the update is stopped and the coating is regulated based on the final optimization result; Wherein, the comprehensive fitness function is as follows: in, Represents the fitness value, is the film thickness weight, Characterize the film thickness fitness function, , Characterize the total number of measurement points, Characterize any measurement point, Characterization The thickness value of each measuring point, Characterize the average thickness of the film layer, Characterize the film adhesion weight, Characterize the film adhesion fitness function, , Characterize the film adhesion test value, Characterize the best adhesion value, Characterize the optical performance weight, Characterize the optical performance fitness function, , is the transmittance weight, is the transmittance, is the reflection weight, is the reflectivity; The optimization update based on the fitness analysis results, variable sensitivity, and variable dependency under the coating constraint conditions includes: Analyzing the fitness analysis result to obtain an independent fitness set, wherein the independent fitness set includes film thickness fitness, film adhesion fitness, and optical performance fitness; Performing fitness trigger analysis on the independent fitness set to establish a trigger analysis result; Perform variable linkage combination according to the trigger analysis result to configure joint genes, and perform the next round of optimization update based on the joint genes; The step of performing variable linkage combination according to the trigger analysis result to configure joint genes includes: Obtain a set of variable combination pairs according to the variable dependency screening; Perform linkage combination matching of the variable combination pair set according to the trigger analysis result, and establish a joint combination value according to the linkage combination matching result; Establishing independent combination values ​​for unmatched variable combination pairs in the variable combination pair set; Combination screening is performed using the joint combination value and the independent combination value, and the joint gene is configured according to the combination screening result.

2. The method for controlling the adaptive coating of blue glass surface according to claim 1, wherein: The next round of optimization update based on the combined gene includes: configuring a benchmark adjustment amplitude based on the variable sensitivity; Selecting a target parameter in the initial parameter set according to the fitness analysis result, taking the target parameter as the optimization target, and updating the remaining parameters in the initial parameter set based on the benchmark adjustment amplitude; Perform adaptive mutation updates of target parameters, perform cross-mutation on all update results, establish an updated parameter set, and complete a round of optimization update.

3. The method for controlling the adaptive coating on the surface of blue glass according to claim 1, wherein: After the coating is regulated by the final optimization result, the following steps are included: Perform coating monitoring and establish monitoring data sets; Performing a matching analysis between the actual coating effect and the optimization result based on the monitoring data set, and establishing a matching error; Establishing a retrospective mark according to the matching error, performing a retrospective self-inspection of the coating process using the retrospective mark, and establishing a retrospective self-inspection result; Coating process control compensation is performed based on the retrospective self-test result.

4. The method for controlling the adaptive coating of blue glass surface according to claim 1, wherein: Also includes: Establish a self-inspection cycle, conduct self-inspection of the coating effect during the self-inspection cycle, and establish the cycle deviation; The cycle deviation is used to generate a cycle compensation, and self-optimization management of the coating process is performed based on the cycle compensation.

5. The blue glass surface adaptive coating control system is characterized by: The method for controlling the adaptive coating on the surface of blue glass according to any one of claims 1 to 4 is used to implement the steps, wherein the adaptive coating on the surface of blue glass comprises: A constraint condition construction module is used to establish coating constraint conditions. The coating constraint conditions are constructed by collecting the limit constraints of the film layer and the equipment parameters of the coating equipment. The coating constraint conditions include the film hardness constraint, the film thickness allowable deviation constraint, and the coating equipment working range constraint; A process analysis module is used to perform process analysis based on the coating process after obtaining the coating process, and configure variable sensitivity and variable dependency of optimization variables; The fitness analysis module is used to construct an initial parameter set based on the optimization variables, conduct fitness analysis of the initial parameter set through a comprehensive fitness function, and perform optimization and update under coating constraints based on the fitness analysis results and variable sensitivity and variable dependency; The coating control module is used to stop updating when the optimization update result meets the preset conditions, and to perform coating control based on the final optimization result.

Images