Coating process control method and system for stress control

Abstract

The present invention discloses a coating process control method and system for stress control, relating to the field of coating process technology. The method comprises: constructing a coating task vector based on substrate feature information and coating requirement information of a lens to be coated, performing a global credible search to obtain a coating control plan; introducing a stress-induced risk prediction framework for risk analysis; establishing a coating optimization guidance space if the stress-induced risk sequence does not satisfy the induced risk constraint sequence; adjusting the coating control plan to establish a first coating adjustment space, performing induced risk optimization to establish a second coating adjustment space; determining a coating control optimization strategy, and performing coating control based on the lens to be coated. The method solves the technical problems in the prior art of stress control in coating processes, which lack global optimization capabilities, make it difficult to balance optical performance and stress risk, and result in poor lens coating quality and reliability, thereby achieving the technical effect of improving lens coating quality and coating process reliability.

Description

Technical Field

[0001] The present application relates to the technical field related to coating processes, and specifically to a coating process control method and system for stress control. Background Art

[0002] The coating process of high-performance lenses has a decisive influence on the performance, reliability and lifespan of optical devices. The stress generated during the coating process is a key factor affecting the quality of the lens. During the optical coating process, thin films of different materials are deposited on the surface of the lens (usually glass, silicon, polymer, etc.), which will generate internal stress. If the stress is not controlled, it may cause the lens to deform, bend, crack or fall off the film layer, or even peel off between multi-layer film systems, causing damage to the substrate, thereby reducing the optical imaging quality and stability; traditional coating process control lacks the ability to globally model and predict the risk of coating stress formation mechanism, especially when facing complex substrate features (such as aspheric surfaces, ultra-thin lenses or special materials) and diversified coating requirements (such as multi-layer high-reflection films, anti-reflection films or environmentally resistant films). It is difficult to effectively control the stress distribution while ensuring optical performance; and the stress-induced risks in the coating process are characterized by multi-factor coupling and nonlinearity. Adjustment of a single process parameter often cannot achieve global optimization, affecting the coating quality, reliability and production efficiency of optical lenses.

[0003] Therefore, in the current relevant technologies, there is a technical problem that the stress control of the coating process lacks global optimization capabilities, making it difficult to balance optical performance and stress risks, resulting in poor lens coating quality and reliability. Summary of the Invention

[0004] This application solves the technical problems in the prior art of lacking global optimization capability for stress control in the coating process, making it difficult to balance optical performance and stress risk, resulting in poor lens coating quality and reliability, by providing a coating process control method and system for stress control. This achieves the technical effect of improving the lens coating quality and coating process reliability.

[0005] The present application provides a coating process control method for stress control, the method comprising: constructing a coating task vector based on substrate feature information and coating requirement information of a lens to be plated, and performing a global credible search on the coating process space based on the coating task vector to obtain a coating control plan; introducing a stress-induced risk prediction framework to perform risk analysis on the coating control plan to obtain a stress-induced risk sequence; if the stress-induced risk sequence does not satisfy an induced risk constraint sequence, performing a stress optimization guidance analysis based on the induced risk constraint sequence to establish a coating optimization guidance space; adjusting the coating control plan based on the coating optimization guidance space to establish a first coating adjustment space, and performing induced risk optimization on the first coating adjustment space based on a stress-induced risk optimization mechanism to establish a second coating adjustment space; performing an optimization search on the second coating adjustment space based on a lens coating evaluation element set to determine a coating control optimization strategy, and performing coating control on the lens to be plated based on the coating control optimization strategy.

[0006] In a possible implementation, the coating process control method for stress control also performs the following processing: searching the coating process space according to the coating task vector to establish a first coating control matching space, wherein the coating process space includes multiple coating process vectors, and each coating process vector includes a sample coating task vector and a sample coating control vector; performing a first-level trigger degree calculation based on each coating control matching vector in the first coating control matching space to obtain a first-level trigger degree of each vector; performing a second-level trigger degree calculation based on each coating control matching vector to obtain a second-level trigger degree of each vector; calculating the ratio of the first-level trigger degree of each vector to the second-level trigger degree of each vector to generate multiple coating control global credibility; performing a global credibility maximization search on the first coating control matching space based on the multiple coating control global credibility to generate the coating control scheme.

[0007] In a possible implementation, the coating process control method for stress control further performs the following processing: performing stress distribution confidence fitting according to the coating control scheme to obtain a coating stress confidence distribution; the stress-induced risk prediction architecture includes an induced substrate risk prediction model, an induced film layer risk prediction model and an induced film-base adhesion risk prediction model; inputting the coating stress confidence distribution into the induced substrate risk prediction model to obtain an induced substrate risk coefficient; inputting the coating stress confidence distribution into the induced film layer risk prediction model to obtain an induced film layer risk coefficient; inputting the coating stress confidence distribution into the induced film-base adhesion risk prediction model to obtain an induced film-base adhesion risk coefficient, and combining the induced substrate risk coefficient and the induced film layer risk coefficient to generate the stress-induced risk sequence.

[0008] In a possible implementation, the coating process control method for stress control further performs the following processing: modeling is performed based on the lens to be plated to obtain a lens model to be plated; the lens model to be plated is simulated coated Q times according to the coating control scheme to obtain Q stress distribution fitting sets, where Q is a positive integer greater than 1; the Q stress distribution fitting sets are clustered at the same position to obtain stress distribution clusters at each position; and the stress distribution clusters at each position are fused within the cluster to generate the coating stress confidence distribution.

[0009] In a possible implementation, the coating process control method for stress control also performs the following processing: optimizing and screening the first coating control matching space according to a predetermined global credibility to establish a second coating control matching space; performing stress-induced risk retrieval according to the second coating control matching space to obtain each control vector induced risk set, and performing similar risk concentration value calculation on each control vector induced risk set to establish each control vector induced risk sequence; based on each control vector induced risk sequence, selecting the second coating control matching space according to the induced risk constraint sequence to generate a third coating control matching space; performing multi-dimensional coating control trigger feature analysis on the third coating control matching space to generate the coating optimization guidance space.

[0010] In a possible implementation, the coating process control method for stress control further performs the following processing: extracting the P-th coating adjustment scheme according to the first coating adjustment space, where P is a positive integer; performing risk prediction on the P-th coating adjustment scheme according to the stress-induced risk prediction framework to obtain the P-th induced risk sequence; if the P-th induced risk sequence satisfies the induced risk constraint sequence, adding the P-th coating adjustment scheme to the second coating adjustment space; if the P-th induced risk sequence does not satisfy the induced risk constraint sequence, eliminating the P-th coating adjustment scheme.

[0011] In a possible implementation, the coating process control method for stress control also performs the following processing: constructing a coating evaluation multidimensional constraint based on the lens coating evaluation element set, the lens coating evaluation element set including lens coating quality, lens coating efficiency and film bonding degree; evaluating the second coating adjustment space according to the lens coating evaluation element set to obtain a coating evaluation space; based on the coating evaluation space, performing evaluation constraint optimization on the second coating adjustment space according to the coating evaluation multidimensional constraint to generate a coating adjustment third space; performing energy consumption minimization optimization on the third coating adjustment space to generate the coating control optimization strategy.

[0012] In a possible implementation, the stress-control-oriented coating process control method further performs the following processing: constructing a substrate feature vector based on the substrate feature information; constructing a coating requirement vector based on the coating requirement information; and generating the coating task vector based on the substrate feature vector and the coating requirement vector.

[0013] In a possible implementation, the coating process control method for stress control further performs the following processing: constructing a coating process expansion vector based on the coating task vector and the coating control optimization strategy, and expanding the coating process space based on the coating process expansion vector.

[0014] The present application also provides a coating process control system for stress control, the system comprising: a coating control scheme acquisition module for constructing a coating task vector according to substrate feature information and coating requirement information of a lens to be plated, and performing a global credible search on the coating process space based on the coating task vector to obtain a coating control scheme; a scheme risk analysis module for introducing a stress-induced risk prediction framework to perform risk analysis on the coating control scheme to obtain a stress-induced risk sequence; a stress optimization guidance analysis module for, if the stress-induced risk sequence does not satisfy the induced risk constraint sequence, performing a stress-induced risk analysis on the coating control scheme according to the induced risk constraint sequence; A risk constraint sequence is used to perform stress optimization guidance analysis to establish a coating optimization guidance space; a coating adjustment space establishment module is used to adjust the coating control scheme according to the coating optimization guidance space, establish a first coating adjustment space, and perform induced risk optimization on the first coating adjustment space according to the stress-induced risk optimization mechanism to establish a second coating adjustment space; a coating control optimization strategy determination module is used to perform optimization search on the second coating adjustment space according to a lens coating evaluation element set, determine a coating control optimization strategy, and perform coating control on the lens to be plated according to the coating control optimization strategy.

[0015] The stress-controlled coating process control method and system proposed in this application constructs a coating task vector based on the substrate feature information and coating requirement information of the lens to be coated, performs a global credible search to obtain a coating control plan, introduces a stress-induced risk prediction framework for risk analysis, establishes a coating optimization guidance space if the stress-induced risk sequence does not satisfy the induced risk constraint sequence, adjusts the coating control plan, establishes a first coating adjustment space, performs induced risk optimization, and establishes a second coating adjustment space, determines a coating control optimization strategy, and performs coating control based on the lens to be coated. This solves the technical problem in the prior art that stress control in the coating process lacks global optimization capabilities, making it difficult to balance optical performance and stress risk, resulting in poor lens coating quality and reliability, thereby achieving the technical effect of improving lens coating quality and coating process reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] To more clearly illustrate the technical solutions of the embodiments of the present disclosure, the accompanying drawings of the embodiments of the present disclosure are briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in precise order. Instead, various steps may be processed in reverse order or simultaneously as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0017] Figure 1 A schematic flow chart of a coating process control method for stress control provided in an embodiment of the present application.

[0018] Figure 2 Schematic diagram of the structure of the coating process control system for stress control provided in an embodiment of the present application.

[0019] Explanation of the reference numerals: coating control scheme obtaining module 10 , scheme risk analysis module 20 , stress optimization guidance analysis module 30 , coating adjustment space establishment module 40 , coating control optimization strategy determination module 50 . DETAILED DESCRIPTION

[0020] 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, it can be implemented in accordance with the contents of the specification. In order 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 listed below.

[0021] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting 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.

[0022] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict, and the terms “first\second” involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. The terms “including” and “having” and any variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or that are inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.

[0023] The present application provides a coating process control method for stress control, such as Figure 1 As shown, the method includes:

[0024] Step S100: constructing a coating task vector according to substrate feature information and coating requirement information of the lens to be coated, and performing a global credible search on the coating process space based on the coating task vector to obtain a coating control solution.

[0025] Preferably, substrate characteristic information and coating requirement information of the lens to be plated are obtained, wherein the substrate characteristic information is a key factor affecting the coating stress distribution, and may include the substrate material (such as fused quartz, BK7, sapphire, etc.) and its mechanical properties (refractive index, elastic modulus, Abbe number, thermal expansion coefficient, etc.), substrate geometric parameters (such as curvature radius, thickness, aspheric coefficient, etc.) and surface condition (polishing quality, cleanliness, etc.). For example, different glass materials, such as crown glass and flint glass, have different refractive indices and dispersion characteristics. The size of the curvature radius will affect the stress distribution of the film layer during the coating process. When coating a lens surface with a small curvature, the film layer is more susceptible to bending stress in the edge area. Coating requirement information is a key indicator that determines the design of the film layer. It may include optical performance requirements (such as reflectivity, transmittance, laser damage threshold, etc.), coating type (anti-reflection film, anti-reflection film, reflective film, etc.), and film structure (such as single-layer film, multi-layer film, gradient refractive index film, etc.). The thickness of the film layer is the core parameter of the coating requirements. For example, for anti-reflection films in the visible light region, the optical thickness of a single-layer film is usually designed to be about one-quarter of the wavelength.

[0026] Preferably, the substrate feature information and coating requirement information are converted into a numerical vector through mathematical modeling (such as tensor representation, feature encoding), which is like an array containing multiple parameters, and each element corresponds to different information. For example, the elements of the vector may include the code of the substrate material (to distinguish different types of optical glass, etc.), the desired coating thickness, the upper and lower limits of the required transmittance range, etc.; then, a global credible search is performed on the coating process space based on the coating task vector, that is, in the coating process parameter space (such as deposition temperature, rate, air pressure, film thickness, etc.), combined with the constraints of the task vector, an intelligent optimization algorithm (such as The team uses genetic algorithms, particle swarm optimization, Bayesian optimization, etc. to conduct efficient searches, combined with physical simulations (such as finite element analysis (FEM) and optical thin film simulation) to evaluate the stress distribution and optical performance of different parameter combinations. Credibility assessments (such as uncertainty quantification and Monte Carlo sampling) are also introduced to ensure that the optimization results are within the feasible range of the process. Ultimately, the team seeks the optimal or near-optimal coating control solution, that is, a set of optimized process parameter combinations, such as deposition parameters (sputtering power, deposition rate, substrate temperature, etc.), post-processing parameters (annealing temperature, time, etc.), and film structure optimization (adjusting the film thickness distribution or material combination to reduce stress).

[0027] Furthermore, step S100 also includes step S101, constructing a substrate feature vector based on the substrate feature information; step S102, constructing a coating requirement vector based on the coating requirement information; and step S103, generating the coating task vector based on the substrate feature vector and the coating requirement vector.

[0028] Preferably, the parameters contained in the lens substrate feature information are quantified into numerical values ​​to construct a substrate feature vector. Specifically, the substrate material is divided into different categories, such as optical glass, resin, etc., and different materials are assigned different numerical representations; physical properties such as hardness are measured using standard values ​​such as Mohs hardness; the thermal expansion coefficient is directly represented by a numerical value measured by scientific experiments; the shape parameters (spherical radius, aspheric coefficient, etc.) and size parameters (diameter, center thickness, edge thickness, etc.) of the lens are described with specific numerical values, and the parameters are arranged and combined in order to form a substrate feature vector.

[0029] Preferably, the coating requirement information mainly includes the functions that the lens needs to achieve and the performance requirements that need to be met after coating. For example, if an anti-reflection coating is coated, the transmittance requirements for different wavelengths of light are used as parameters, and the transmittance at a specific wavelength (such as 550nm in the visible light band) is required to reach 99%. If it is a reflective coating, the reflection wavelength range and reflectivity requirements are specified, such as the reflectivity at a wavelength of 1064nm must be greater than 99.5%. For film thickness, different functional coatings have different requirements. For example, some multi-layer film structures have precise numerical requirements for the thickness of each layer as a coating thickness parameter. Similarly, the various parameters of the coating requirements are arranged and combined in order to construct a coating requirement vector. Finally, the substrate feature vector and the coating requirement vector are combined to generate a coating task vector to comprehensively describe the coating task, which helps to accurately understand the lens substrate and coating goals, thereby formulating an appropriate coating control plan.

[0030] Furthermore, step S100 also includes step S110, searching the coating process space according to the coating task vector to establish a first coating control matching space, wherein the coating process space includes multiple coating process vectors, and each coating process vector includes a sample coating task vector and a sample coating control vector; step S120, performing a first-level trigger degree calculation according to each coating control matching vector in the first coating control matching space to obtain a first-level trigger degree of each vector; step S130, performing a second-level trigger degree calculation according to each coating control matching vector to obtain a second-level trigger degree of each vector; step S140, calculating the ratio of the first-level trigger degree of each vector to the second-level trigger degree of each vector to generate multiple coating control global credibility; step S150, performing a global credibility maximization search on the first coating control matching space according to the multiple coating control global credibility to generate the coating control scheme.

[0031] Preferably, a credible process matching space is constructed based on historical coating data, and the optimal coating control solution is screened through multi-level triggering degree calculation. The coating process space includes multiple coating process vectors, each of which includes a sample coating task vector (historical lens substrate characteristics and coating requirements) and a sample coating control vector (corresponding successful coating process parameters). Specifically, based on the coating task vector of the current lens to be coated, similar historical cases are retrieved from the coating process space through similarity calculations (e.g., cosine similarity or Euclidean distance), and multiple matching coating control matching vectors (i.e., candidate process solutions) are output as the first coating control matching space. A first-level triggering degree calculation is then performed on each coating control matching vector in the first coating control matching space. This involves counting the number of times each coating control matching vector appears in the historical data (the first coating control matching space) to measure the overall reliability of the process solution (the greater the number of appearances, the higher the credibility). Ultimately, a first-level triggering degree is obtained for each vector. For example, if a coating control matching vector appears five times in the historical data, its first-level triggering degree is 5.

[0032] Preferably, a second-level triggering degree calculation is performed for each coating control match vector. This involves calculating the sum of the independent credibility of each parameter within each coating control match vector. Specifically, the coating control match vector is decomposed to extract each parameter (e.g., sputtering power, deposition rate, substrate temperature). The number of occurrences of each parameter in the first coating control matching space (i.e., the coating parameter triggering degree) is then counted. The triggering degrees of all parameters are summed to obtain a second-level triggering degree, which is used to assess the rationality of the parameter combination of the scheme. (Even if the overall scheme occurs infrequently, it may still be reliable if all parameters are common.) The ratio of the first-level triggering degree to the second-level triggering degree of each coating control match vector is then calculated as the global credibility of the coating control. Multiple global credibility scores for coating control are then obtained. A ratio close to 1 indicates that the scheme is highly standardized (both the overall scheme and the parameters appear frequently). A lower ratio (e.g., 5 / 45) indicates that the parameter combination of the scheme is common, but the overall scheme is rare, and further verification may be required to balance the reliability of the overall scheme and the rationality of the parameters. Finally, from the first coating control matching space, the coating control matching vector with the highest global credibility is selected. For example, all candidate coating control matching vectors are sorted according to the global credibility of coating control, and the solution with the highest credibility is directly selected as the final coating control solution (i.e., the optimal process parameter combination).

[0033] In step S200 , a stress-induced risk prediction framework is introduced to perform risk analysis on the coating control solution to obtain a stress-induced risk sequence.

[0034] Step S200 further includes step S210, performing stress distribution confidence fitting according to the coating control scheme to obtain a coating stress confidence distribution; step S220, the stress-induced risk prediction framework includes an induced substrate risk prediction model, an induced film layer risk prediction model and an induced film-base adhesion risk prediction model; step S230, inputting the coating stress confidence distribution into the induced substrate risk prediction model to obtain an induced substrate risk coefficient; step S240, inputting the coating stress confidence distribution into the induced film layer risk prediction model to obtain an induced film layer risk coefficient; step S250, inputting the coating stress confidence distribution into the induced film-base adhesion risk prediction model to obtain an induced film-base adhesion risk coefficient, and generating the stress-induced risk sequence by combining the induced substrate risk coefficient and the induced film layer risk coefficient.

[0035] Preferably, stress distribution confidence fitting is performed according to the coating control scheme, that is, the distribution results of the stress magnitude and its credible fluctuation range introduced by the coating process in the lens substrate and film layer are predicted through quantitative analysis to evaluate the reliability and risk of the process. Specifically, the coating control scheme (optimized process parameter combination), substrate material parameters (elastic modulus, thermal expansion coefficient, etc.) and film layer structure data (thickness, material, stacking sequence, etc.) are used as input, and the thermal stress and intrinsic stress distribution in the coating process are simulated by finite element analysis. Then, Gaussian process regression (GPR) or Bayesian neural network is used to predict the stress distribution, and the confidence interval (such as the stress fluctuation range at 95% confidence level) is output. Finally, the coating stress confidence distribution is output, which may include the stress mean, variance and spatial distribution (such as the stress gradient on the substrate surface).

[0036] Preferably, the stress-induced risk prediction architecture includes an induced substrate risk prediction model, an induced film layer risk prediction model, and an induced film-substrate adhesion risk prediction model, which are respectively used to evaluate different failure modes. Specifically, the induced substrate risk prediction model is trained using random forest or gradient boosting tree to predict the risk of deformation or rupture of the substrate due to coating stress, wherein the model input is the stress confidence distribution (maximum tensile / compressive stress, stress gradient), the mechanical properties of the substrate material (fracture toughness, yield strength), and the geometric characteristics of the substrate (thickness, curvature radius), the training label is the substrate failure case in the historical data, and the output result is the induced substrate risk coefficient (0~1, the higher the value, the greater the risk); the induced film layer risk prediction model uses a convolutional neural network (CNN) to process the stress spatial distribution, combined with the training of the historical data of film layer failure, to predict the risk of cracking or peeling of the film layer itself due to stress, wherein the model input is the film layer stress distribution (internal stress, interface shear stress), the film layer material properties (hardness, brittleness), and the deposition process parameters (deposition rate, temperature), and the output result is the induced film layer risk coefficient (such as the probability of film cracking).

[0037] Preferably, the induced film-substrate adhesion risk prediction model uses a support vector machine (SVM) or graph neural network (GNN) to analyze the nonlinear relationship between interfacial stress and adhesion, and is used to predict the risk of adhesion failure (e.g., delamination) between the film layer and the substrate. The model inputs include interfacial stress (shear stress, peel stress), interface treatment process (plasma cleaning, transition layer design), and material compatibility (difference in thermal expansion coefficient). The output is an induced film-substrate adhesion risk coefficient. The induced film-substrate adhesion risk coefficient, induced substrate risk coefficient, and induced film layer risk coefficient are integrated to generate a stress-induced risk sequence. If any risk coefficient exceeds a threshold (e.g., substrate risk > 0.3), the solution is deemed infeasible.

[0038] Furthermore, step S210 also includes step S211, modeling the lens to be plated to obtain a lens model to be plated; step S212, performing Q simulated coating on the lens model to be plated according to the coating control scheme to obtain Q stress distribution fitting sets, where Q is a positive integer greater than 1; step S213, clustering the Q stress distribution fitting sets at the same position to obtain stress distribution clusters at each position; step S214, performing cluster-in-cluster confidence fusion on the stress distribution clusters at each position to generate the coating stress confidence distribution.

[0039] Preferably, the substrate geometric parameters (CAD model, curvature, thickness, etc.) and material properties (Young's modulus, thermal expansion coefficient, etc.) are used as input, a parameterized three-dimensional model is established through finite element analysis, and the surface morphology is reconstructed using point cloud scanning to generate a digital twin of the lens that can be simulated including mesh division, that is, the lens model to be plated; within the process fluctuation range allowed by the coating control scheme (such as deposition rate ±5%, temperature ±3°C), Monte Carlo sampling is performed to generate Q groups of parameter combinations, which are input into the lens model to be plated for simulated coating, and Q full-lens stress distributions (such as the stress value of each grid point) are output, thereby obtaining Q stress distribution fitting sets, where Q is a positive integer. For example, if Q=100, the stress distribution fitting set includes 100 stress cloud maps.

[0040] Preferably, Q stress distribution fitting sets are clustered at the same position, that is, the stress value distribution pattern at the same position on the lens surface (such as grid point A) in Q coating simulations is analyzed to identify potential multimodal distribution characteristics (such as principal stress intervals and outlier intervals). Specifically, the stress data of the Q simulations are aligned according to the grid points, and a data set is generated for each point. The high-density area is divided into clusters using DBSCAN density clustering, and the low-density points are regarded as noise. Outliers (such as erroneous simulation results caused by numerical divergence) are eliminated, and principal stress clusters and outlier clusters are identified to obtain stress distribution clusters at each position, that is, each grid point corresponds to a clustering result; then, the clustering result of each position point is probabilistically modeled to generate a stress confidence interval and quantify process uncertainty, that is, the probability distribution is independently fitted to each cluster, and the confidence interval is calculated, and finally a stress confidence thermodynamic map of the entire lens is generated as the coating stress confidence distribution.

[0041] Step S300 : If the stress induced risk sequence does not satisfy the induced risk constraint sequence, a stress optimization guidance analysis is performed according to the induced risk constraint sequence to establish a coating optimization guidance space.

[0042] Step S300 further includes step S310, performing optimal screening on the first coating control matching space according to a predetermined global credibility to establish a second coating control matching space; step S320, performing stress-induced risk retrieval according to the second coating control matching space to obtain each control vector induced risk set, and performing similar risk concentration value calculation on each control vector induced risk set to establish each control vector induced risk sequence; step S330, based on each control vector induced risk sequence, selecting the second coating control matching space according to the induced risk constraint sequence to generate a third coating control matching space; step S340, performing multi-dimensional coating control trigger feature analysis on the third coating control matching space to generate the coating optimization guidance space.

[0043] Preferably, when the stress-induced risk sequence generated by the coating control scheme exceeds the induced risk constraint sequence set by the user, a coating optimization guidance space is established. Specifically, a predetermined global credibility (e.g., 0.7) is set based on historical data, and the triggering degree ratio of each coating control matching vector in the first coating control matching space is calculated to obtain the global credibility. All coating control matching vectors with global credibility greater than the predetermined global credibility are retained to form the second coating control matching space. Then, a stress-induced risk search is performed on each coating control scheme in the second coating control matching space, that is, the induced substrate risk coefficient, induced film layer risk coefficient, and induced film-base adhesion risk coefficient in the historical data are retrieved to form each control vector induced risk set. Then, a risk concentration value calculation of the same type is performed on each control vector induced risk set, including calculating the median of the induced substrate risk coefficient, the median of the induced film layer risk coefficient, and the median of the induced film-base adhesion risk coefficient. Then, the induced risk sequence of each control vector is constructed using the median of each risk coefficient. Example stress-induced risk search calculation data is shown in Table 1:

[0044] Table 1 Stress-induced risk retrieval calculation data table (example)

[0045]

[0046] Preferably, the induced risk constraint sequence is a risk coefficient threshold defined according to historical data, and then the second matching space of coating control is selected according to the induced risk constraint sequence, and the coating control scheme that satisfies the induced risk constraint in all items of the risk sequence is retained to establish a third matching space of coating control; a multi-dimensional coating control trigger feature analysis is performed on the third matching space of coating control, specifically, principal component analysis is used to extract key influencing factors, which may include process parameter sensitivity, risk coupling relationship and historical optimization trajectory, and finally the parameter boundary and optimization direction of the feasible process area are structured to represent, and the coating optimization guidance space is output, thereby ensuring the reliability and optimization efficiency of the coating process.

[0047] Step S400, adjusting the coating control scheme according to the coating optimization guiding space to establish a first coating adjustment space, and performing induced risk optimization on the first coating adjustment space according to a stress induced risk optimization mechanism to establish a second coating adjustment space.

[0048] Preferably, when the stress risk of the initial coating control scheme exceeds the safety threshold, the original coating control scheme is adjusted based on the coating optimization guidance space, that is, under the premise of ensuring the core process goals, the parameters that exceed the safety range are gradually adjusted to the reasonable range recommended by the guidance space. For example, the coating optimization guidance space indicates that the safety range of the deposition rate is 0.3~0.5nm / s, and the coating control scheme uses 0.6nm / s, so a new rate value in the range of 0.3~0.5nm / s is generated, and then the first coating adjustment space is established; then, the first coating adjustment space is induced risk optimization based on the stress induced risk optimization mechanism, and the various parameter groups are deeply analyzed. Specifically, for each candidate parameter combination in the first coating adjustment space, the three types of stress risks (induced substrate risk coefficient, induced film layer risk coefficient and induced film-base adhesion risk coefficient) that may be generated are predicted. A multi-objective optimization algorithm is used to comprehensively evaluate the candidate parameter combinations, and the schemes with excellent performance in multiple risk dimensions are determined. The feasibility of each scheme is evaluated in combination with historical successful case data, and finally the second coating adjustment space is screened and formed, in which all process parameters are within the safety range set by the coating optimization guidance space, and various stress risk indicators are controlled within the allowable threshold. While ensuring safety, the optical performance requirements are met as much as possible.

[0049] Furthermore, step S400 also includes step S410, extracting the P-th coating adjustment scheme according to the first coating adjustment space, where P is a positive integer; step S420, performing risk prediction on the P-th coating adjustment scheme according to the stress-induced risk prediction framework to obtain the P-th induced risk sequence; step S430, if the P-th induced risk sequence satisfies the induced risk constraint sequence, adding the P-th coating adjustment scheme to the second coating adjustment space; step S440, if the P-th induced risk sequence does not satisfy the induced risk constraint sequence, eliminating the P-th coating adjustment scheme.

[0050] Preferably, the Pth candidate process scheme is extracted from the first space of coating adjustment, where P is a positive integer, indicating which candidate scheme is currently being evaluated. Risk prediction is performed on the Pth coating adjustment scheme according to the stress-induced risk prediction framework, that is, the process parameters (such as deposition temperature, rate, etc.) of the Pth coating adjustment scheme are input into the induced substrate risk prediction model, the induced film layer risk prediction model and the induced film-base adhesion risk prediction model, and the risks of lens substrate deformation or rupture, thin film cracking or peeling, and insufficient bonding between the film and the substrate are evaluated respectively. The corresponding induced substrate risk coefficient, induced film layer risk coefficient and induced film-base adhesion risk coefficient are output and combined to form the Pth induced risk sequence.

[0051] Preferably, an induced risk constraint sequence is set according to historical data and actual needs, that is, the upper limit value of each risk indicator is pre-set. When the Pth induced risk sequence satisfies the induced risk constraint sequence, that is, all risk coefficients of the Pth scheme are ≤ the corresponding risk constraint value, the Pth coating adjustment scheme is added to the second coating adjustment space. If the Pth induced risk sequence does not satisfy the induced risk constraint sequence, that is, any risk coefficient is greater than the corresponding risk constraint value, the Pth coating adjustment scheme is eliminated, and finally a second coating adjustment space that fully meets the risk requirements is formed.

[0052] Step S500: performing an optimization search on the second coating adjustment space according to the lens coating evaluation element set, determining a coating control optimization strategy, and performing coating control on the lens to be coated according to the coating control optimization strategy.

[0053] Step S500 further includes step S510, constructing a coating evaluation multidimensional constraint based on the lens coating evaluation element set, wherein the lens coating evaluation element set includes lens coating quality, lens coating efficiency and film bonding degree; step S520, evaluating the second coating adjustment space according to the lens coating evaluation element set to obtain a coating evaluation space; step S530, based on the coating evaluation space, performing evaluation constraint optimization on the second coating adjustment space according to the coating evaluation multidimensional constraint to generate a coating adjustment third space; step S540, performing energy consumption minimization optimization on the third coating adjustment space to generate the coating control optimization strategy.

[0054] Preferably, the lens coating evaluation element set includes lens coating quality, lens coating efficiency, and film bonding. Lens coating quality includes key parameters directly affecting optical effects, such as optical performance indicators (e.g., transmittance, reflectivity), surface roughness, and film uniformity. Lens coating efficiency includes production efficiency indicators such as process time cost, material utilization, and equipment occupancy time. Film bonding includes reliability indicators such as the bonding strength between the film layer and the substrate, interface stability, and resistance to environmental aging. Based on historical coating data, reasonable evaluation criteria are set for each dimension to form a multi-dimensional constraint for coating evaluation. The lens coating evaluation element set is used to evaluate the second coating adjustment space. Specifically, for each candidate coating adjustment solution, historical production records under similar process conditions are retrieved. The median of the retrieved historical coating data is taken as the expected performance of the solution. A coating evaluation sequence containing the evaluation results of the three dimensions is generated for each coating adjustment solution. Ultimately, a coating evaluation space is obtained, which not only includes process parameters but also records lens coating quality parameters, lens coating efficiency parameters, and film bonding parameters.

[0055] Preferably, based on the coating evaluation space, the second coating adjustment space is evaluated and constrained for optimization according to the coating evaluation multi-dimensional constraints, that is, the evaluation sequence of each coating adjustment scheme is compared with the preset coating evaluation multi-dimensional constraints, and according to pre-set standards (such as coating quality > 98 points, efficiency > 90 points, and bonding > 95 points), the coating adjustment schemes that meet the requirements of the three dimensions are screened out to form the third coating adjustment space, ensuring that the coating adjustment scheme can meet the optical performance, production efficiency and reliability, etc.; finally, the third coating adjustment space is optimized for energy minimization, that is, the production energy consumption data of each scheme (including electricity, gas, target material consumption, etc.) is analyzed, and the coating adjustment scheme with the lowest energy consumption is given priority, and finally the coating control optimization strategy is output, which not only ensures the high quality and high reliability of the coating products, but also takes into account the production cost and environmental protection requirements, such as a coating adjustment scheme in which the deposition temperature is reduced by 20°C but extended by 10 minutes, or a new pulse deposition mode is used instead of continuous deposition. The process parameters included in the coating control optimization strategy are then automatically loaded into the equipment control unit to establish the process recipe for each chamber (such as the electron gun scanning path). For example, the deposition time is adjusted in real time through in-situ spectral monitoring, the substrate temperature is adjusted using infrared thermal imager feedback, and the gas flow is dynamically optimized based on mass spectrometer data. After the coating is completed, full-band scanning with a spectrophotometer and laser confocal surface morphology analysis are performed, and verification feedback is performed to ensure that the coating process achieves the best balance in all aspects, as well as the coating quality and reliability, and to ensure the manufacture of high-performance optical devices.

[0056] Furthermore, step S500 also includes constructing a coating process expansion vector according to the coating task vector and the coating control optimization strategy, and expanding the coating process space according to the coating process expansion vector.

[0057] Preferably, the coating task vector and the coating control optimization strategy are integrated, that is, on the basis of the original coating task vector, the key process parameter adjustment information in the coating control optimization strategy is introduced to construct a coating process expansion vector, which not only includes the basic information and coating requirements of the lens, but also includes the adjustment direction and range of the process parameters after optimization; the coating process space is expanded according to the coating process expansion vector, that is, a derivative parameter combination is generated at the edge of the safety interval of the optimization strategy and added to the coating process space, providing more choices and possibilities for the optimization of the coating process, ensuring that in the face of complex coating tasks, especially when it is necessary to accurately control quality factors such as film layer stress, the probability of determining the optimal coating solution is increased, which helps to improve the quality and performance of the coated products and enables the coating process to better adapt to different lens substrates and the diversity of coating needs.

[0058] In the above, refer to Figure 1 The coating process control method for stress control according to an embodiment of the present invention is described in detail. Figure 2A coating process control system for stress control according to an embodiment of the present invention is described.

[0059] The coating process control system for stress control according to the embodiment of the present invention is used to solve the technical problems in the prior art that stress control in the coating process lacks global optimization capabilities, making it difficult to balance optical performance and stress risks, resulting in poor lens coating quality and reliability, thereby achieving the technical effect of improving lens coating quality and coating process reliability. Figure 2 As shown, the coating process control system for stress control includes: a coating control plan acquisition module 10, a plan risk analysis module 20, a stress optimization guidance analysis module 30, a coating adjustment space establishment module 40, and a coating control optimization strategy determination module 50.

[0060] The coating control scheme acquisition module 10 is used to construct a coating task vector based on the substrate feature information and coating requirement information of the lens to be coated, and perform a global credible search on the coating process space based on the coating task vector to obtain a coating control scheme; the scheme risk analysis module 20 is used to introduce a stress-induced risk prediction framework to perform risk analysis on the coating control scheme to obtain a stress-induced risk sequence; the stress optimization guidance analysis module 30 is used to perform stress optimization guidance according to the induced risk constraint sequence if the stress-induced risk sequence does not meet the induced risk constraint sequence. Analysis, establishing a coating optimization guidance space; a coating adjustment space establishment module 40, used to adjust the coating control scheme according to the coating optimization guidance space, establish a first coating adjustment space, and induce risk optimization for the first coating adjustment space according to the stress-induced risk optimization mechanism, and establish a second coating adjustment space; a coating control optimization strategy determination module 50, used to perform optimization search for the second coating adjustment space according to the lens coating evaluation element set, determine the coating control optimization strategy, and perform coating control on the lens to be plated according to the coating control optimization strategy.

[0061] The specific configuration of the coating control scheme acquisition module 10 will be described in detail below. The coating control scheme acquisition module 10 further includes: searching the coating process space according to the coating task vector to establish a first coating control matching space, wherein the coating process space includes multiple coating process vectors, each coating process vector includes a sample coating task vector and a sample coating control vector; performing a first-level triggering degree calculation according to each coating control matching vector in the first coating control matching space to obtain a first-level triggering degree of each vector; performing a second-level triggering degree calculation according to each coating control matching vector to obtain a second-level triggering degree of each vector; performing a ratio calculation of the first-level triggering degree of each vector with the second-level triggering degree of each vector to generate multiple coating control global credibility; performing a global credibility maximization search on the first coating control matching space according to the multiple coating control global credibility to generate the coating control scheme.

[0062] The specific configuration of the solution risk analysis module 20 will be described in detail below. The solution risk analysis module 20 further includes: performing stress distribution confidence fitting according to the coating control solution to obtain a coating stress confidence distribution; the stress-induced risk prediction architecture includes an induced substrate risk prediction model, an induced film layer risk prediction model, and an induced film-substrate adhesion risk prediction model; inputting the coating stress confidence distribution into the induced substrate risk prediction model to obtain an induced substrate risk coefficient; inputting the coating stress confidence distribution into the induced film layer risk prediction model to obtain an induced film layer risk coefficient; inputting the coating stress confidence distribution into the induced film-substrate adhesion risk prediction model to obtain an induced film-substrate adhesion risk coefficient, and combining the induced substrate risk coefficient and the induced film layer risk coefficient to generate the stress-induced risk sequence.

[0063] The specific configuration of the solution risk analysis module 20 will be described in detail below. The solution risk analysis module 20 further includes: modeling the lens to be plated to obtain a lens model; performing Q simulated coatings on the lens model according to the coating control scheme to obtain Q stress distribution fitting sets, where Q is a positive integer greater than 1; clustering the Q stress distribution fitting sets at the same location to obtain stress distribution clusters at each location; and performing cluster-in-cluster confidence fusion on the stress distribution clusters at each location to generate the coating stress confidence distribution.

[0064] The specific configuration of the stress optimization guidance analysis module 30 will be described in detail below. The stress optimization guidance analysis module 30 further includes: optimizing and screening the first coating control matching space according to a predetermined global credibility to establish a second coating control matching space; performing stress-induced risk retrieval based on the second coating control matching space to obtain each control vector induced risk set, and performing similar risk concentration value calculation on each control vector induced risk set to establish each control vector induced risk sequence; based on each control vector induced risk sequence, selecting the second coating control matching space according to the induced risk constraint sequence to generate a third coating control matching space; and performing multi-dimensional coating control trigger feature analysis on the third coating control matching space to generate the coating optimization guidance space.

[0065] The specific configuration of the coating adjustment space establishment module 40 will be described in detail below. The coating adjustment space establishment module 40 further includes: extracting a Pth coating adjustment solution based on the first coating adjustment space, where P is a positive integer; performing risk prediction on the Pth coating adjustment solution based on the stress-induced risk prediction framework to obtain a Pth induced risk sequence; if the Pth induced risk sequence satisfies the induced risk constraint sequence, adding the Pth coating adjustment solution to the second coating adjustment space; if the Pth induced risk sequence does not satisfy the induced risk constraint sequence, eliminating the Pth coating adjustment solution.

[0066] The specific configuration of the coating control optimization strategy determination module 50 will be described in detail below. The coating control optimization strategy determination module 50 further includes: constructing a coating evaluation multidimensional constraint based on the lens coating evaluation element set, wherein the lens coating evaluation element set includes lens coating quality, lens coating efficiency, and film bonding; evaluating the coating control second space based on the lens coating evaluation element set to obtain a coating evaluation space; based on the coating evaluation space, performing evaluation constraint optimization on the coating control second space according to the coating evaluation multidimensional constraint to generate a coating control third space; and performing energy consumption minimization optimization on the coating control third space to generate the coating control optimization strategy.

[0067] The specific configuration of the coating control solution acquisition module 10 will be described in detail below. The coating control solution acquisition module 10 further includes: constructing a substrate feature vector based on the substrate feature information; constructing a coating requirement vector based on the coating requirement information; and generating the coating task vector based on the substrate feature vector and the coating requirement vector.

[0068] The following will further describe the specific configuration of the coating control optimization strategy determination module 50. The coating control optimization strategy determination module 50 further includes: constructing a coating process expansion vector based on the coating task vector and the coating control optimization strategy, and expanding the coating process space based on the coating process expansion vector.

[0069] The coating process control system for stress control provided by the embodiment of the present invention can execute the coating process control method for stress control provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0070] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other and are not used to limit the scope of protection of the present invention.

[0071] The above specific embodiments do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application shall be included within the scope of protection of this application.

Claims

1. A coating process control method for stress control, characterized in that: include: Constructing a coating task vector based on the substrate feature information and coating requirement information of the lens to be coated, and performing a global credible search on the coating process space based on the coating task vector to obtain a coating control solution; Introducing a stress-induced risk prediction framework to perform risk analysis on the coating control scheme and obtain a stress-induced risk sequence; If the stress-induced risk sequence does not satisfy the induced risk constraint sequence, performing stress optimization guidance analysis according to the induced risk constraint sequence to establish a coating optimization guidance space; Adjusting the coating control scheme according to the coating optimization guiding space to establish a first coating adjustment space, and performing induced risk optimization on the first coating adjustment space according to a stress-induced risk optimization mechanism to establish a second coating adjustment space; performing an optimization search on the second coating adjustment space according to the lens coating evaluation element set, determining a coating control optimization strategy, and performing coating control on the lens to be coated according to the coating control optimization strategy; Wherein, performing a global credible search on the coating process space based on the coating task vector to obtain a coating control solution includes: Searching the coating process space according to the coating task vector to establish a first coating control matching space, wherein the coating process space includes a plurality of coating process vectors, and each coating process vector includes a sample coating task vector and a sample coating control vector; Performing a first-level triggering degree calculation on each coating control matching vector in the first coating control matching space to obtain a first-level triggering degree of each vector; Perform secondary trigger degree calculation according to each coating control matching vector to obtain the secondary trigger degree of each vector; Calculate the ratio of the first-level trigger degree of each vector to the second-level trigger degree of each vector to generate multiple global credibility of coating control; Performing a global credibility maximization search on the first matching space of the coating control according to the multiple coating control global credibility to generate the coating control solution; The stress-induced risk prediction framework is introduced to analyze the risk of the coating control scheme to obtain a stress-induced risk sequence, including: Perform stress distribution confidence fitting according to the coating control scheme to obtain coating stress confidence distribution; The stress-induced risk prediction framework includes an induced substrate risk prediction model, an induced film layer risk prediction model, and an induced film-substrate attachment risk prediction model; Inputting the coating stress confidence distribution into the induced substrate risk prediction model to obtain an induced substrate risk coefficient; Inputting the coating stress confidence distribution into the induced film layer risk prediction model to obtain the induced film layer risk coefficient; Inputting the coating stress confidence distribution into the induced film-substrate adhesion risk prediction model to obtain the induced film-substrate adhesion risk coefficient, and combining the induced substrate risk coefficient and the induced film layer risk coefficient to generate the stress induced risk sequence; The process of performing stress distribution confidence fitting according to the coating control scheme to obtain coating stress confidence distribution includes: Modeling the lens to be plated to obtain a model of the lens to be plated; Performing Q simulated coatings on the lens model to be coated according to the coating control scheme to obtain Q stress distribution fitting sets, where Q is a positive integer greater than 1; Performing co-location clustering on the Q stress distribution fitting sets to obtain stress distribution clusters at each position; Performing intra-cluster confidence fusion on the stress distribution clusters at each position to generate the coating stress confidence distribution; The stress optimization guidance analysis is performed according to the induced risk constraint sequence to establish a coating optimization guidance space, including: Optimizing and screening the first matching space for coating control according to a predetermined global credibility, and establishing a second matching space for coating control; Performing stress-induced risk retrieval according to the second matching space of the coating control to obtain risk sets induced by each control vector, and performing similar risk concentration value calculation on the risk sets induced by each control vector to establish a risk sequence induced by each control vector; Based on the risk sequences induced by the control vectors, the second matching space for coating control is selected according to the induced risk constraint sequence to generate a third matching space for coating control; Performing multi-dimensional coating control trigger feature analysis on the coating control third matching space to generate the coating optimization guiding space; The stress-induced risk optimization mechanism includes: Extracting a P-th coating adjustment scheme according to the coating adjustment first space, where P is a positive integer; Performing risk prediction on the Pth coating adjustment scheme according to the stress-induced risk prediction framework to obtain a Pth induced risk sequence; If the Pth induced risk sequence satisfies the induced risk constraint sequence, adding the Pth coating adjustment solution to the second coating adjustment space; If the Pth induced risk sequence does not satisfy the induced risk constraint sequence, eliminating the Pth coating adjustment scheme; The method of performing an optimization search on the second coating adjustment space according to the lens coating evaluation element set to determine the coating control optimization strategy includes: Constructing a multi-dimensional constraint for coating evaluation based on the lens coating evaluation element set, wherein the lens coating evaluation element set includes lens coating quality, lens coating efficiency, and film bonding degree; Evaluate the second coating adjustment space according to the lens coating evaluation element set to obtain a coating evaluation space; Based on the coating evaluation space, performing evaluation constraint optimization on the second coating adjustment space according to the coating evaluation multi-dimensional constraint to generate a third coating adjustment space; The energy consumption of the coating adjustment third space is minimized to generate the coating control optimization strategy.

2. The coating process control method for stress control according to claim 1, characterized in that: According to the substrate feature information and coating requirement information of the lens to be coated, a coating task vector is constructed, including: Constructing a basis feature vector according to the basis feature information; Constructing a coating demand vector according to the coating demand information; The coating task vector is generated according to the substrate feature vector and the coating requirement vector.

3. The coating process control method for stress control according to claim 1, characterized in that: According to the coating task vector and the coating control optimization strategy, a coating process expansion vector is constructed, and the coating process space is expanded according to the coating process expansion vector.

4. The coating process control system for stress control is characterized by: The system is used to implement the coating process control method for stress control according to any one of claims 1 to 3, and the system comprises: A coating control scheme acquisition module is used to construct a coating task vector based on the substrate feature information and coating requirement information of the lens to be coated, and perform a global credible search on the coating process space based on the coating task vector to obtain a coating control scheme; A scheme risk analysis module is used to introduce a stress-induced risk prediction framework to perform risk analysis on the coating control scheme and obtain a stress-induced risk sequence; A stress optimization guidance analysis module is used to perform stress optimization guidance analysis according to the induced risk constraint sequence if the stress induced risk sequence does not satisfy the induced risk constraint sequence, and establish a coating optimization guidance space; a coating adjustment space establishment module, configured to adjust the coating control scheme according to the coating optimization guidance space to establish a first coating adjustment space, and to perform induced risk optimization on the first coating adjustment space according to a stress-induced risk optimization mechanism to establish a second coating adjustment space; The coating control optimization strategy determination module is used to perform optimization search on the second coating adjustment space according to the lens coating evaluation element set, determine the coating control optimization strategy, and perform coating control on the lens to be coated according to the coating control optimization strategy.

Images