Digital twinning-based substrate process adaptive control method and system, and medium

Abstract

The invention discloses a digital twinning-based substrate process adaptive control method and system and a medium, and relates to the technical field of substrate process control, and the method comprises the steps: obtaining a substrate strength demand, a substrate structure hierarchy demand and a substrate material cost demand, and determining a substrate polymer combination candidate set; carrying out a twin test, and determining a target substrate polymer combination, a combination mode between combination layers and combination thickness distribution; actual production and simulated production are carried out, a real-time monitoring data set and a simulated data set are obtained, comparison is carried out to determine items to be optimized and targets to be optimized, then real-time substrate process control parameters are obtained and optimized, and optimized substrate process control parameters are obtained; and performing process adaptive control on the substrate production line based on the optimized substrate process control parameters. The technical problem that in the optical glass lens substrate manufacturing process in the prior art, process control lacks an accurate dynamic response mechanism is solved, and the technical effect of achieving substrate process dynamic closed-loop control is achieved.

Description

Technical Field

[0001] The present invention relates to the field of substrate process control technology, and in particular to a substrate process adaptive control method, system and medium based on digital twins. Background Art

[0002] In the production process of optical glass lens substrates, the adjustment of process control parameters usually relies on experience and manual monitoring, and lacks a real-time feedback mechanism. In particular, in processing scenarios involving the composite of multi-layer polymer structures and glass materials, temperature control, pressing time and interlayer bonding behavior are highly coupled, which can easily lead to problems such as uneven interlayer bonding and insufficient heat treatment during the production process. These problems are difficult to detect and effectively optimize in a timely manner, often resulting in unstable substrate quality and increased production deviations, affecting the performance and reliability of the final product. At the same time, traditional methods cannot dynamically adjust control parameters during the production process, and cannot fully utilize the performance of the substrate material, resulting in low production efficiency and yield rate. Summary of the Invention

[0003] The present application provides a substrate process adaptive control method, system and medium based on digital twins, which are used to solve the technical problem of the existing technology in the manufacturing process of optical glass lens substrates, that is, the lack of a precise dynamic response mechanism for process control.

[0004] In view of the above problems, the present application provides a substrate process adaptive control method, system and medium based on digital twin.

[0005] A first aspect of the present application provides a substrate process adaptive control method based on digital twins, the method comprising:

[0006] Obtain substrate strength requirements, substrate structure level requirements and substrate material cost requirements, perform multi-dimensional target screening combination in the polymer candidate library, and determine the substrate polymer combination candidate set; input the substrate polymer combination candidate set and the glass substrate into the digital twin platform respectively to perform twin tests on the combination interlayer bonding mode and combination thickness distribution, and determine the target substrate polymer combination, as well as the combination interlayer bonding mode and combination thickness distribution, wherein the digital twin platform includes a virtual test scenario library, which is used to build test scenarios for different combination interlayer bonding modes and combination thickness distributions of substrate polymer combinations in different application environments; substrate production line and The digital twin platform performs actual production and simulated production based on the target substrate polymer combination and glass substrate, as well as the combination interlayer bonding mode and the combination thickness distribution, and uses a real-time data acquisition module to monitor the production process in real time, obtain a real-time monitoring data set, and compare the real-time monitoring data set with the simulation data set of simulated production to determine the items to be optimized and the targets to be optimized; obtain the real-time substrate process control parameters of the substrate production line, optimize the real-time substrate process control parameters based on the items to be optimized and the targets to be optimized, obtain optimized substrate process control parameters, and perform process adaptive control on the substrate production line based on the optimized substrate process control parameters.

[0007] A second aspect of the present application provides a substrate process adaptive control system based on digital twins, the system comprising:

[0008] A demand acquisition module is used to obtain substrate strength requirements, substrate structure level requirements and substrate material cost requirements, perform multi-dimensional target screening combinations in the polymer candidate library, and determine the substrate polymer combination candidate set; a twin test module is used to input the substrate polymer combination candidate set and the glass substrate into the digital twin platform respectively to perform twin tests on the combination interlayer bonding mode and combination thickness distribution, and determine the target substrate polymer combination, as well as the combination interlayer bonding mode and combination thickness distribution, wherein the digital twin platform includes a virtual test scenario library, and the virtual test scenario library is used to build test scenarios for different combination interlayer bonding modes and combination thickness distributions of substrate polymer combinations in different application environments; a monitoring comparison module A block is used for the substrate production line and the digital twin platform to perform actual production and simulated production based on the target substrate polymer combination and glass substrate, as well as the combined interlayer bonding mode and the combined thickness distribution, and use the real-time data acquisition module to monitor the production process in real time, obtain a real-time monitoring data set, compare the real-time monitoring data set with the simulation data set of simulated production, and determine the items to be optimized and the targets to be optimized; a control module is used to obtain the real-time substrate process control parameters of the substrate production line, optimize the real-time substrate process control parameters based on the items to be optimized and the targets to be optimized, obtain optimized substrate process control parameters, and perform process adaptive control on the substrate production line based on the optimized substrate process control parameters.

[0009] The third aspect of the present application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the digital twin-based substrate process adaptive control method provided in the present application.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0011] The present application obtains substrate strength requirements, substrate structure level requirements and substrate material cost requirements, performs multi-dimensional target screening combination in the polymer candidate library, and determines the substrate polymer combination candidate set; the substrate polymer combination candidate set and the glass substrate are respectively input into the digital twin platform for twin testing of the combination interlayer bonding mode and the combination thickness distribution, and the target substrate polymer combination, as well as the combination interlayer bonding mode and the combination thickness distribution, wherein the digital twin platform includes a virtual test scenario library, and the virtual test scenario library is used to build test scenarios for different combination interlayer bonding modes and combination thickness distributions of substrate polymer combinations in different application environments; substrate production line The digital twin platform performs actual production and simulated production based on the target substrate polymer combination and glass substrate, as well as the combination interlayer bonding mode and the combination thickness distribution, and uses a real-time data acquisition module to monitor the production process in real time, obtain a real-time monitoring data set, compare the real-time monitoring data set with the simulation data set of the simulated production, and determine the items to be optimized and the targets to be optimized; obtain the real-time substrate process control parameters of the substrate production line, optimize the real-time substrate process control parameters based on the items to be optimized and the targets to be optimized, obtain the optimized substrate process control parameters, and perform process adaptive control on the substrate production line based on the optimized substrate process control parameters. The present invention solves the technical problem of the lack of a precise dynamic response mechanism for process control in the manufacturing process of optical glass lens substrates in the prior art. By monitoring the production process data in real time and comparing it with the simulation data, the process control parameters are optimized, so as to improve the real-time and precision of the process parameter adjustment and achieve the technical effect of dynamic closed-loop control of the substrate process. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0013] Figure 1 A schematic flow chart of a substrate process adaptive control method based on digital twins provided in an embodiment of the present application;

[0014] Figure 2 Schematic diagram of the structure of the digital twin-based substrate process adaptive control system provided in an embodiment of the present application.

[0015] Explanation of the accompanying drawings: demand acquisition module 11, twin testing module 12, monitoring and comparison module 13, control module 14. DETAILED DESCRIPTION

[0016] This application provides a substrate process adaptive control method, system and medium based on digital twins to solve the technical problem of the existing technology in the manufacturing process of optical glass lens substrates, which is the lack of a precise dynamic response mechanism for process control. By real-time monitoring of production process data and comparing it with simulation data, the process control parameters are optimized to achieve the technical effect of improving the real-time and accuracy of process parameter adjustment and realizing dynamic closed-loop control of the substrate process.

[0017] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0018] It should be noted that any variations of the terms "include" and "have" 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 clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0019] Example 1, as Figure 1 As shown, the present application provides a substrate process adaptive control method based on digital twins, the method comprising:

[0020] Step S100: Obtain substrate strength requirements, substrate structure level requirements, and substrate material cost requirements, perform multi-dimensional target screening combination in the polymer candidate library, and determine a candidate set of substrate polymer combinations.

[0021] In an embodiment of the present application, first obtain the pre-set substrate strength requirements, substrate structure level requirements and substrate material cost requirements. Next, perform a multi-dimensional target screening combination in the polymer candidate library. Specifically, first retrieve the historical substrate polymer combination set in the polymer candidate library, and obtain the corresponding historical substrate strength requirement set, historical substrate structure level requirement set and substrate material cost requirement set. Then, map these historical requirement sets to a three-dimensional coordinate system, generate a historical call coordinate point set, and construct a historical call space based on these coordinate points. By setting a multi-dimensional requirement tolerance threshold as a constraint for multi-dimensional target screening, the substrate strength requirement, substrate structure level requirement and substrate material cost requirement are input into the historical call space, and multi-dimensional target screening is performed on the historical call coordinate point set. Finally, based on the screening results, select the historical substrate polymer combination that meets the requirements and add it to the substrate polymer combination candidate set.

[0022] Furthermore, in the method provided in the embodiment of the application, the substrate strength requirement, substrate structure level requirement, and substrate material cost requirement are obtained, and a multi-dimensional target screening combination is performed in the polymer candidate library to determine a candidate set of substrate polymer combinations, which also includes:

[0023] Retrieve the historical substrate polymer combination set and the corresponding historical substrate strength requirement set, historical substrate structure level requirement set and substrate material cost requirement set from the polymer candidate library; map the historical substrate strength requirement set, historical substrate structure level requirement set and substrate material cost requirement set into a three-dimensional coordinate system to obtain a historical call coordinate point set, use the historical substrate polymer combination set to identify the historical call coordinate point set, and construct a historical call space, wherein the historical call space includes a historical call coordinate point set; with a multi-dimensional requirement tolerance threshold as a constraint, input the substrate strength requirement, substrate structure level requirement and substrate material cost requirement into the historical call space to perform multi-dimensional target screening on the historical call coordinate point set to determine the screened historical call coordinate point set; add the historical substrate polymer combination corresponding to the screened historical call coordinate point set into the substrate polymer combination candidate set.

[0024] Furthermore, the method provided in the application embodiment also includes:

[0025] The multi-dimensional demand tolerance threshold includes a substrate strength demand tolerance threshold, a substrate structure level demand tolerance threshold, and a substrate material cost demand tolerance threshold.

[0026] In this embodiment, a set of historical substrate polymer combinations in a polymer candidate library is first retrieved, and the corresponding historical substrate strength requirement set, historical substrate structure level requirement set, and substrate material cost requirement set are obtained. The polymer candidate library is a database containing multiple polymers and their combinations, recording the performance parameters, applicability conditions, and cost of each polymer.

[0027] Next, the historical substrate strength requirement set, the historical substrate structure level requirement set, and the historical substrate material cost requirement set are mapped into a three-dimensional coordinate system, with each data point representing a different requirement condition. In this way, all historical requirement data is converted into points in the coordinate system, forming a historical call coordinate point set.

[0028] The historical base polymer combination set is then used to identify the historical call coordinate point set. By associating each historical demand coordinate point with the corresponding base polymer combination, an identification relationship is formed to determine which base polymer combinations correspond to which historical demand conditions. Based on this, a spatial model reflecting the distribution of historical data is constructed, namely the historical call space. The historical call space reflects the distribution of base polymer combinations under past historical demand conditions.

[0029] Then, by setting multi-dimensional demand tolerance thresholds, namely, substrate strength demand tolerance thresholds, substrate structure level demand tolerance thresholds, and substrate material cost demand tolerance thresholds, the historical call coordinate point set is screened. In this process, the substrate strength demand, substrate structure level demand, and substrate material cost demand are first input into the historical call space to obtain the target coordinate point. Then, in combination with the target coordinate point, based on the set substrate strength demand tolerance threshold, substrate structure level demand tolerance threshold, and substrate material cost demand tolerance threshold, the historical call coordinate point set is preliminarily screened to obtain the initial screening historical call coordinate point set. Subsequently, the initial screening historical call coordinate point set is processed by the mean shift method to determine the target screening historical call coordinate point. Finally, based on the target screening historical call coordinate point, the preliminary screening set is advanced screened to ultimately determine the screening historical call coordinate point set that meets the requirements.

[0030] Finally, the historical substrate polymer combinations corresponding to the filtered historical call coordinate point sets are added to the candidate set of substrate polymer combinations. The candidate set of substrate polymer combinations is a set of polymer combinations that have been screened to meet the current substrate requirements (strength, structural hierarchy, and cost), providing alternative solutions for subsequent testing and process optimization.

[0031] Furthermore, in the method provided in the embodiment of the application, the substrate strength requirement, substrate structure level requirement, and substrate material cost requirement are input into the historical call space to perform multi-dimensional target screening on the historical call coordinate point set, and determining the screened historical call coordinate point set further includes:

[0032] The substrate strength requirement, substrate structure level requirement and substrate material cost requirement are input into the historical call space to obtain a target coordinate point; in combination with the target coordinate point, based on the substrate strength requirement tolerance threshold, the substrate structure level requirement tolerance threshold and the substrate material cost requirement tolerance threshold, the historical call coordinate point set is subjected to multi-dimensional target screening to determine an initial screening historical call coordinate point set; mean shift is performed on the initial screening historical call coordinate point set to determine a target screening historical call coordinate point; based on the target screening historical call coordinate point, the initial screening historical call coordinate point set is subjected to advanced screening to determine the screening historical call coordinate point set.

[0033] In an embodiment of the present application, the substrate strength requirement, substrate structure level requirement and substrate material cost requirement are first input into the aforementioned historical call space, that is, the substrate strength requirement, substrate structure level requirement and substrate material cost requirement are mapped into a three-dimensional coordinate system to obtain the corresponding target coordinate point.

[0034] Then, combined with the target coordinate points, the historical call coordinate point set is subjected to multi-dimensional target screening based on the substrate strength requirement tolerance threshold, substrate structure level requirement tolerance threshold, and substrate material cost requirement tolerance threshold. This process ensures that historical data points that meet the current demand conditions are screened out by setting the tolerance range of each demand parameter. For example, if the substrate strength requirement tolerance threshold is ±10MPa, it means that the substrate strength is allowed to fluctuate within an error range of 10MPa. The substrate strength, substrate structure level, and substrate material cost of each historical data point are compared with the current demand, and points that meet the requirements are screened out to form the initial screening historical call coordinate point set.

[0035] The initial screening history call coordinate point set is then mean shifted. In this process, each data point is adjusted according to the density within its neighborhood, with the goal of finding the dense center of the area where the data point is located. Specifically, the neighborhood density of each data point is first calculated. The more data points in the neighborhood, the higher the density. The data point will then drift toward the area with dense neighborhoods, that is, it will be moved to an area with higher density. It is not that the closer to the target coordinate point, the better, but considering that the data points in dense areas have stronger representativeness and stability, they are used as screening centers. Through this method, we can eventually obtain target screening history call coordinate points with higher density and better representativeness on the basis of meeting current demand conditions.

[0036] Finally, the initial screening history call coordinate point set is advanced screened based on the target screening history call coordinate point, that is, by calculating the distance between each point in the initial screening history call coordinate point set and the target screening history call coordinate point, the first m sites closest to the target point are selected, and these closest points are added to the screening history call coordinate point set.

[0037] Furthermore, the method provided in the application embodiment also includes:

[0038] The initial screening history call coordinate point whose distance to the target screening history call coordinate point in the initial screening history call coordinate point set is located in the first m places is added to the screening history call coordinate point set.

[0039] In the embodiment of the present application, the Euclidean distance between each point in the initial screening history call coordinate point set and the target screening history call coordinate point is first calculated, i.e., a distance metric is used to quantify the difference between each point and the target point. Then, all points in the initial screening history call coordinate point set are sorted by their distance to the target screening history call coordinate point, and the first m closest points are selected in ascending order. These first m closest initial screening history call coordinate points are added to the screening history call coordinate point set.

[0040] Step S200: Input the candidate set of substrate polymer combinations and the glass substrate into the digital twin platform respectively to perform twin tests on the combination interlayer bonding mode and the combination thickness distribution to determine the target substrate polymer combination, as well as the combination interlayer bonding mode and the combination thickness distribution, wherein the digital twin platform includes a virtual test scenario library, and the virtual test scenario library is used to build test scenarios for different combination interlayer bonding modes and combination thickness distributions of substrate polymer combinations in different application environments.

[0041] In this embodiment, the substrate-polymer combinations and glass substrates from the candidate set of substrate-polymer combinations are first input into the substrate overlay module of the digital twin platform to simulate the interlayer bonding method and thickness distribution of the combinations. This simulation results in a candidate set of substrate-polymer combinations.

[0042] The digital twin platform then accesses the virtual test scenario library and places candidate simulation sets of substrate-polymer combinations into the platform for scenario-based testing. The virtual test scenario library constructs test scenarios for different interlayer bonding methods and thickness distributions of substrate-polymer combinations to simulate different application environments, such as the impact of external factors such as temperature changes and pressure fluctuations on substrate performance. The test results are then transmitted through scenario-based testing to the combination evaluation module, which evaluates the test results and outputs a scored set of candidate simulation sets of substrate-polymer combinations.

[0043] Finally, the substrate polymer combination with the highest score is obtained from the candidate simulation score set of substrate polymer combinations, and the substrate polymer combination is used as the target substrate polymer combination, and its corresponding combination interlayer bonding mode and combination thickness distribution are determined.

[0044] Furthermore, in the method provided in the application embodiment, the candidate set of substrate polymer combinations and the glass substrate are respectively input into the digital twin platform for twin testing of the combination interlayer bonding mode and the combination thickness distribution to determine the target substrate polymer combination, the combination interlayer bonding mode and the combination thickness distribution. The digital twin platform includes a virtual test scenario library, which is used to build test scenarios for different combination interlayer bonding modes and combination thickness distributions of substrate polymer combinations in different application environments, and also includes:

[0045] The substrate polymer combination and the glass substrate in the substrate polymer combination candidate set are respectively input into the substrate superposition module of the digital twin platform to simulate the combination interlayer bonding mode and the combination thickness distribution to obtain the substrate polymer combination candidate simulation set; the virtual test scenario library is called to perform scenario-based testing on the substrate polymer combination candidate simulation set in the digital twin platform, and the scenario-based test results are transmitted to the combination evaluation module to output the substrate polymer combination candidate simulation score set; the substrate polymer combination, combination interlayer bonding mode and combination thickness distribution corresponding to the maximum value in the substrate polymer combination candidate simulation score set are obtained.

[0046] In an embodiment of the present application, the substrate polymer combination and the glass substrate in the substrate polymer combination candidate set are first input into the substrate superposition module of the digital twin platform respectively. The substrate superposition module is used to simulate the combination interlayer bonding mode and the combination thickness distribution. Its purpose is to superimpose different polymer combinations and glass substrates to form a multilayer composite material structure model. In this process, the digital twin platform simulates the properties, thickness and interlayer bonding mode of each layer of material, such as pressing, heat treatment and gluing. In this way, a candidate simulation set of substrate polymer combinations is obtained.

[0047] Next, the digital twin platform calls the virtual test scenario library to perform scenario-based testing on the candidate simulation set of substrate polymer combinations. The virtual test scenario library is a database containing a variety of test environments. These scenarios simulate different working conditions and external environments, such as temperature fluctuations, humidity changes, etc. The digital twin platform simulates environmental factors by calling the virtual test scenario library and combining it with the aforementioned multi-layer composite material structure model to evaluate the performance of these combinations in various working environments. The purpose of scenario-based testing is to verify the performance of different substrate combinations in different application environments, including the impact of factors such as temperature resistance and moisture resistance on interlayer bonding strength and overall stability. After the test is completed, the scenario-based test results are generated.

[0048] The scenario-based test results are then transmitted to the combination evaluation module. The combination evaluation module uses a pre-built scoring model to comprehensively evaluate the scenario-based test results. The scoring model is built based on a neural network, and the training process uses historical scenario-based test results and corresponding scores. During training, the neural network gradually adjusts the network weights by inputting historical test data and its corresponding scores to minimize the error between the predicted score and the expert identification score. The network is optimized by the back-propagation algorithm, so that the model can learn the scoring rules under different test environments, and then automatically generate scores based on the scenario-based test results. By transmitting the scenario-based test results to the scoring model in the combination evaluation module, a candidate simulation score set for the substrate polymer combination is obtained.

[0049] Finally, the substrate polymer combination with the highest score is obtained from the candidate simulation score set of substrate polymer combinations as the target substrate polymer combination, and its corresponding combination interlayer bonding mode and combination thickness distribution are determined.

[0050] Step S300: The substrate production line and the digital twin platform respectively perform actual production and simulated production based on the target substrate polymer combination and the glass substrate, as well as the combination interlayer bonding mode and the combination thickness distribution, and utilize the real-time data acquisition module to monitor the production process in real time, obtain a real-time monitoring data set, and compare the real-time monitoring data set with the simulation data set of simulated production to determine the items and targets to be optimized.

[0051] In an embodiment of the present application, the substrate production line and the digital twin platform perform actual production and simulated production respectively based on the determined target substrate polymer combination, glass substrate, combined interlayer bonding method and combined thickness distribution. The substrate production line combines the substrate polymer combination with the glass substrate according to the preset production process, and adopts the determined interlayer bonding method (such as pressing, heat treatment, gluing, etc.) and thickness distribution to carry out actual production. At the same time, the digital twin platform performs virtual simulation of the same production process to simulate the performance of the target substrate polymer combination and glass substrate under the set process conditions.

[0052] During the actual production process, the real-time data acquisition module will monitor each link of the production in real time and obtain a real-time monitoring data set, which includes key process parameters such as temperature, pressure, time, and pressurization conditions.

[0053] Next, the real-time monitoring data set is compared with a simulated data set from a simulated production run. The simulated data set, generated by the digital twin platform, reflects the production process under ideal conditions. By comparing these two sets of data, discrepancies between actual production and expectations can be identified, such as uneven interlayer bonding or insufficient heat treatment.

[0054] Through this comparison, the system automatically identifies the items to be optimized and the targets to be optimized. Items to be optimized refer to parameters such as the heat treatment process and pressurization conditions that do not meet expectations. The targets to be optimized are to optimize the production process by adjusting these process parameters so that the final substrate product performance meets the expected standards.

[0055] Step S400: Acquire real-time substrate process control parameters of the substrate production line, optimize the real-time substrate process control parameters based on the items to be optimized and the targets to be optimized, obtain optimized substrate process control parameters, and perform process adaptive control on the substrate production line based on the optimized substrate process control parameters.

[0056] In the embodiments of the present application, the real-time substrate process control parameters of the substrate production line are first obtained. Key data from each step of the production process, such as temperature, pressure, heating time, and pressurization conditions, are collected through a real-time monitoring system. Next, based on the previously determined items to be optimized and the targets to be optimized, the real-time substrate process control parameters are optimized. This step utilizes a pre-built parameter optimizer to optimize the real-time substrate process control parameters for the items to be optimized and the targets to be optimized, thereby obtaining the optimized substrate process control parameters.

[0057] Finally, the process adaptive control of the substrate production line is performed based on the optimized substrate process control parameters. In this step, the control systems of the production equipment (such as the temperature control system, heating device, and pressure control device) are automatically adjusted to adjust the process parameters in real time during the production process to ensure that production is always maintained at the optimal state.

[0058] Furthermore, in the method provided in the embodiment of the application, real-time substrate process control parameters of the substrate production line are obtained, and the real-time substrate process control parameters are optimized based on the items to be optimized and the targets to be optimized to obtain optimized substrate process control parameters, further comprising:

[0059] A parameter optimizer is pre-built; and the parameter optimizer is used to optimize the real-time substrate process control parameters for the items to be optimized and the targets to be optimized, thereby obtaining optimized substrate process control parameters.

[0060] In an embodiment of the present application, a parameter optimizer is first pre-built. To build this parameter optimizer, a linear regression method is used and training is performed based on historical production data. These historical data include the relationship between real-time process control parameters (such as temperature, pressure, heating time, etc.) and the items to be optimized that appear in the production process, as well as the degree of deviation of these items to be optimized relative to the process target (i.e., the target to be optimized). By using linear regression, the optimizer can understand how these process control parameters affect the items to be optimized (such as problems such as uneven interlayer bonding and insufficient heat treatment), thereby providing a mathematical model for subsequent optimization. This model enables the optimizer to predict the impact of process control parameters on the error between the item to be optimized and the target to be optimized through training, and provides a basis for the optimization process.

[0061] The parameter optimizer then optimizes the target. The target might be a deviation or issue encountered during the production process, such as uneven heat treatment or insufficient interlayer bonding strength. The target is a pre-defined ideal process performance indicator, such as bonding strength reaching a set threshold or temperature distribution uniformity within the standard error limit. The optimizer uses a gradient descent method to adjust the real-time substrate process control parameters. The gradient descent method calculates the error between the current parameters and the target and the target, and then adjusts the process control parameters to minimize this error. For example, if heat treatment is insufficient, the gradient descent method adjusts the temperature setting or heating time to gradually reduce the error until the optimal state is reached. Finally, through the optimization steps, the optimized substrate process control parameters are obtained. These parameters are adjusted to maximize the improvement of the target, bringing it closer to the performance boundary set by the target. For example, if the goal is to reduce uneven interlayer bonding, the optimized parameters will adjust the heating time or temperature distribution to ensure that the process meets the expected requirements.

[0062] Furthermore, the method provided in the application embodiment also includes:

[0063] Obtain a preset control feedback window; collect feedback monitoring data of the substrate production line in the preset control feedback window, and perform feedback verification based on the obtained feedback monitoring data set. If the feedback verification fails, generate a control alarm message.

[0064] In the embodiment of the present application, a preset control feedback window is first obtained. The preset control feedback window is pre-set by a technical expert and is a fixed time period.

[0065] Next, within the preset control feedback window, feedback monitoring data collection is performed. This process uses a real-time monitoring system to collect the results of the optimized items after the optimized substrate process control parameters are executed. In other words, the optimized process parameters (such as heating temperature, heating time, pressurization conditions, etc.) are applied to the production process, and data related to these optimized processes is collected through sensors and data acquisition equipment within the feedback window. This data forms the feedback monitoring data set.

[0066] Feedback verification is then performed based on the collected feedback monitoring data. The purpose of this feedback verification is to check whether the optimization parameters (such as interlayer bonding strength and heat treatment uniformity) have improved after the optimized process parameters have been implemented. If the collected data indicates that the optimization parameters have not improved, the verification fails, indicating that the current optimization has not achieved the expected results.

[0067] Finally, if the feedback verification fails, a control alarm message is generated to prompt the operator that the current process parameters cannot effectively solve the problem (for example, uneven interlayer bonding) and that the relevant process parameters need to be adjusted. The alarm helps the operator to discover and correct the problem in a timely manner.

[0068] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects:

[0069] The present application obtains substrate strength requirements, substrate structure level requirements and substrate material cost requirements, performs multi-dimensional target screening combination in the polymer candidate library, and determines the substrate polymer combination candidate set; the substrate polymer combination candidate set and the glass substrate are respectively input into the digital twin platform for twin testing of the combination interlayer bonding mode and the combination thickness distribution, and the target substrate polymer combination, as well as the combination interlayer bonding mode and the combination thickness distribution, wherein the digital twin platform includes a virtual test scenario library, and the virtual test scenario library is used to build test scenarios for different combination interlayer bonding modes and combination thickness distributions of substrate polymer combinations in different application environments; substrate production line The digital twin platform performs actual production and simulated production based on the target substrate polymer combination and glass substrate, as well as the combination interlayer bonding mode and the combination thickness distribution, and uses a real-time data acquisition module to monitor the production process in real time, obtain a real-time monitoring data set, compare the real-time monitoring data set with the simulation data set of the simulated production, and determine the items to be optimized and the targets to be optimized; obtain the real-time substrate process control parameters of the substrate production line, optimize the real-time substrate process control parameters based on the items to be optimized and the targets to be optimized, obtain the optimized substrate process control parameters, and perform process adaptive control on the substrate production line based on the optimized substrate process control parameters. The present invention solves the technical problem of the lack of a precise dynamic response mechanism for process control in the manufacturing process of optical glass lens substrates in the prior art. By monitoring the production process data in real time and comparing it with the simulation data, the process control parameters are optimized, so as to improve the real-time and precision of the process parameter adjustment and achieve the technical effect of dynamic closed-loop control of the substrate process.

[0070] Embodiment 2 is based on the same inventive concept as the substrate process adaptive control method based on digital twin in the above embodiment. Figure 2 As shown, the present application provides a substrate process adaptive control system based on digital twins. The system and method embodiments in the present application are based on the same inventive concept. The system includes:

[0071] The demand acquisition module 11 is used to obtain the substrate strength requirements, substrate structure level requirements and substrate material cost requirements, perform multi-dimensional target screening combination in the polymer candidate library, and determine the substrate polymer combination candidate set; the twin test module 12 is used to input the substrate polymer combination candidate set and the glass substrate into the digital twin platform respectively to perform twin tests on the combination interlayer bonding mode and combination thickness distribution, and determine the target substrate polymer combination, as well as the combination interlayer bonding mode and combination thickness distribution, wherein the digital twin platform includes a virtual test scenario library, and the virtual test scenario library is used to build test scenarios for different combination interlayer bonding modes and combination thickness distributions of substrate polymer combinations in different application environments; the monitoring comparison module Block 13 is used for the substrate production line and the digital twin platform to perform actual production and simulated production based on the target substrate polymer combination and the glass substrate, as well as the combined interlayer bonding mode and the combined thickness distribution, respectively, and use the real-time data acquisition module to monitor the production process in real time, obtain a real-time monitoring data set, compare the real-time monitoring data set with the simulation data set of the simulated production, and determine the items to be optimized and the targets to be optimized; the control module 14 is used to obtain the real-time substrate process control parameters of the substrate production line, optimize the real-time substrate process control parameters based on the items to be optimized and the targets to be optimized, obtain the optimized substrate process control parameters, and perform process adaptive control on the substrate production line based on the optimized substrate process control parameters.

[0072] Furthermore, the system is also used to implement the following functions:

[0073] Retrieve the historical substrate polymer combination set and the corresponding historical substrate strength requirement set, historical substrate structure level requirement set and substrate material cost requirement set from the polymer candidate library; map the historical substrate strength requirement set, historical substrate structure level requirement set and substrate material cost requirement set into a three-dimensional coordinate system to obtain a historical call coordinate point set, use the historical substrate polymer combination set to identify the historical call coordinate point set, and construct a historical call space, wherein the historical call space includes a historical call coordinate point set; with a multi-dimensional requirement tolerance threshold as a constraint, input the substrate strength requirement, substrate structure level requirement and substrate material cost requirement into the historical call space to perform multi-dimensional target screening on the historical call coordinate point set to determine the screened historical call coordinate point set; add the historical substrate polymer combination corresponding to the screened historical call coordinate point set into the substrate polymer combination candidate set.

[0074] Furthermore, the system is also used to implement the following functions:

[0075] The multi-dimensional demand tolerance threshold includes a substrate strength demand tolerance threshold, a substrate structure level demand tolerance threshold, and a substrate material cost demand tolerance threshold.

[0076] Furthermore, the system is also used to implement the following functions:

[0077] The substrate strength requirement, substrate structure level requirement and substrate material cost requirement are input into the historical call space to obtain a target coordinate point; in combination with the target coordinate point, based on the substrate strength requirement tolerance threshold, the substrate structure level requirement tolerance threshold and the substrate material cost requirement tolerance threshold, the historical call coordinate point set is subjected to multi-dimensional target screening to determine an initial screening historical call coordinate point set; mean shift is performed on the initial screening historical call coordinate point set to determine a target screening historical call coordinate point; based on the target screening historical call coordinate point, the initial screening historical call coordinate point set is subjected to advanced screening to determine the screening historical call coordinate point set.

[0078] Furthermore, the system is also used to implement the following functions:

[0079] The initial screening history call coordinate point whose distance to the target screening history call coordinate point in the initial screening history call coordinate point set is located in the first m places is added to the screening history call coordinate point set.

[0080] Furthermore, the system is also used to implement the following functions:

[0081] The substrate polymer combination and the glass substrate in the substrate polymer combination candidate set are respectively input into the substrate superposition module of the digital twin platform to simulate the combination interlayer bonding mode and the combination thickness distribution to obtain the substrate polymer combination candidate simulation set; the virtual test scenario library is called to perform scenario-based testing on the substrate polymer combination candidate simulation set in the digital twin platform, and the scenario-based test results are transmitted to the combination evaluation module to output the substrate polymer combination candidate simulation score set; the substrate polymer combination, combination interlayer bonding mode and combination thickness distribution corresponding to the maximum value in the substrate polymer combination candidate simulation score set are obtained.

[0082] Furthermore, the system is also used to implement the following functions:

[0083] A parameter optimizer is pre-built; and the parameter optimizer is used to optimize the real-time substrate process control parameters for the items to be optimized and the targets to be optimized, thereby obtaining optimized substrate process control parameters.

[0084] Furthermore, the system is also used to implement the following functions:

[0085] Obtain a preset control feedback window; collect feedback monitoring data of the substrate production line in the preset control feedback window, and perform feedback verification based on the obtained feedback monitoring data set. If the feedback verification fails, generate a control alarm message.

[0086] Example three, based on the same inventive concept as the substrate process adaptive control method based on digital twins in the previous example, this application also provides a computer-readable storage medium, on which a computer program is stored, and the computer program implements the steps of any one of the methods described in the above example one when executed.

[0087] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0088] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0089] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A substrate process adaptive control method based on digital twins, characterized in that: The method comprises: Obtain substrate strength requirements, substrate structure level requirements, and substrate material cost requirements, perform multi-dimensional target screening and combination in the polymer candidate library, and determine a candidate set of substrate polymer combinations; The candidate set of substrate polymer combinations and the glass substrate are respectively input into the digital twin platform for twin testing of the combination interlayer bonding mode and the combination thickness distribution, to determine the target substrate polymer combination, as well as the combination interlayer bonding mode and the combination thickness distribution, wherein the digital twin platform includes a virtual test scenario library, and the virtual test scenario library is used to build test scenarios for different combination interlayer bonding modes and combination thickness distributions of substrate polymer combinations in different application environments; The substrate production line and the digital twin platform respectively perform actual production and simulated production based on the target substrate polymer combination and glass substrate, as well as the combination interlayer bonding mode and the combination thickness distribution, and utilize a real-time data acquisition module to monitor the production process in real time, obtain a real-time monitoring data set, and compare the real-time monitoring data set with the simulated data set of the simulated production to determine the items and targets to be optimized; Acquire real-time substrate process control parameters of the substrate production line, optimize the real-time substrate process control parameters based on the items to be optimized and the targets to be optimized, obtain optimized substrate process control parameters, and perform process adaptive control on the substrate production line based on the optimized substrate process control parameters.

2. The substrate process adaptive control method based on digital twin according to claim 1, characterized in that: Obtain substrate strength requirements, substrate structure level requirements, and substrate material cost requirements, perform multi-dimensional target screening combinations in the polymer candidate library, and determine the candidate set of substrate polymer combinations, including: Retrieving a historical substrate polymer combination set and a corresponding historical substrate strength requirement set, a historical substrate structure level requirement set, and a substrate material cost requirement set from the polymer candidate library; Mapping a historical substrate strength requirement set, a historical substrate structure level requirement set, and a substrate material cost requirement set into a three-dimensional coordinate system to obtain a historical call coordinate point set, and using the historical substrate polymer combination set to identify the historical call coordinate point set to construct a historical call space, wherein the historical call space includes the historical call coordinate point set; Taking the multi-dimensional requirement tolerance threshold as a constraint, the substrate strength requirement, substrate structure level requirement, and substrate material cost requirement are input into the historical call space to perform multi-dimensional target screening on the historical call coordinate point set to determine the screened historical call coordinate point set; The historical substrate polymer combination corresponding to the screening history call coordinate point set is added to the substrate polymer combination candidate set.

3. The substrate process adaptive control method based on digital twin according to claim 2, characterized in that: The multi-dimensional demand tolerance threshold includes a substrate strength demand tolerance threshold, a substrate structure level demand tolerance threshold, and a substrate material cost demand tolerance threshold.

4. The substrate process adaptive control method based on digital twin according to claim 3, characterized in that: Taking the multi-dimensional requirement tolerance threshold as a constraint, the substrate strength requirement, substrate structure level requirement, and substrate material cost requirement are input into the historical call space to perform multi-dimensional target screening on the historical call coordinate point set, and determining the screened historical call coordinate point set includes: Inputting the substrate strength requirement, substrate structure level requirement, and substrate material cost requirement into the history call space to obtain a target coordinate point; In combination with the target coordinate point, based on the substrate strength requirement tolerance threshold, the substrate structure level requirement tolerance threshold, and the substrate material cost requirement tolerance threshold, the historical call coordinate point set is subjected to multi-dimensional target screening to determine an initial screening historical call coordinate point set; Performing mean shift on the initial screening history call coordinate point set to determine the target screening history call coordinate point; The initial screening history call coordinate point set is advanced screened based on the target screening history call coordinate point to determine the screening history call coordinate point set.

5. The substrate process adaptive control method based on digital twin according to claim 4, characterized in that: The initial screening history call coordinate point whose distance to the target screening history call coordinate point in the initial screening history call coordinate point set is located in the first m places is added to the screening history call coordinate point set.

6. The substrate process adaptive control method based on digital twin according to claim 1, characterized in that: The candidate set of substrate polymer combinations and the glass substrate are respectively input into the digital twin platform for twin testing of the combination interlayer bonding mode and the combination thickness distribution to determine the target substrate polymer combination, the combination interlayer bonding mode and the combination thickness distribution. The digital twin platform includes a virtual test scenario library, which is used to build test scenarios for different combination interlayer bonding modes and combination thickness distributions of substrate polymer combinations in different application environments, including: Inputting the substrate polymer combination and the glass substrate in the substrate polymer combination candidate set into the substrate superposition module of the digital twin platform to simulate the combination interlayer bonding mode and combination thickness distribution, thereby obtaining a substrate polymer combination candidate simulation set; Calling the virtual test scenario library, performing scenario-based testing on the substrate polymer combination candidate simulation set in the digital twin platform, transmitting the scenario-based test results to the combination evaluation module, and outputting a substrate polymer combination candidate simulation score set; Obtain the substrate polymer combination, combination interlayer bonding mode and combination thickness distribution corresponding to the maximum value in the substrate polymer combination candidate simulation score set.

7. The substrate process adaptive control method based on digital twin according to claim 1, characterized in that: Acquiring real-time substrate process control parameters of the substrate production line, optimizing the real-time substrate process control parameters based on the items to be optimized and the targets to be optimized, and obtaining optimized substrate process control parameters, including: Pre-built parameter optimizer; The parameter optimizer is used to optimize the real-time substrate process control parameters for the items to be optimized and the targets to be optimized, so as to obtain optimized substrate process control parameters.

8. The substrate process adaptive control method based on digital twin according to claim 1, characterized in that: include: Get the preset control feedback window; Feedback monitoring data of the substrate production line is collected in the preset control feedback window, and feedback verification is performed based on the acquired feedback monitoring data set. If the feedback verification fails, a control alarm message is generated.

9. The substrate process adaptive control system based on digital twin is characterized by: The system is used to execute the substrate process adaptive control method based on digital twin according to any one of claims 1 to 8, and the system includes: A requirement acquisition module is used to obtain substrate strength requirements, substrate structure level requirements, and substrate material cost requirements, perform multi-dimensional target screening and combination in the polymer candidate library, and determine a candidate set of substrate polymer combinations; A twin test module, configured to input the candidate set of substrate polymer combinations and the glass substrate into a digital twin platform, respectively, to perform twin tests on the interlayer bonding mode and the combined thickness distribution, thereby determining a target substrate polymer combination, as well as the interlayer bonding mode and the combined thickness distribution. The digital twin platform includes a virtual test scenario library, which is configured to construct test scenarios for different interlayer bonding modes and combined thickness distributions of substrate polymer combinations under different application environments. A monitoring and comparison module is used for the substrate production line and the digital twin platform to respectively perform actual production and simulated production based on the target substrate polymer combination and glass substrate, as well as the combined interlayer bonding mode and the combined thickness distribution, and to use the real-time data acquisition module to monitor the production process in real time, obtain a real-time monitoring data set, compare the real-time monitoring data set with the simulated data set of the simulated production, and determine the items and targets to be optimized; A control module is used to obtain real-time substrate process control parameters of the substrate production line, optimize the real-time substrate process control parameters based on the items to be optimized and the targets to be optimized, obtain optimized substrate process control parameters, and perform process adaptive control on the substrate production line based on the optimized substrate process control parameters.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the digital twin-based substrate process adaptive control method as described in any one of claims 1 to 8 is implemented.

Images