High vacuum furnace temperature control method and system based on partitioned thermal field reconstruction

Abstract

The application discloses a high-vacuum furnace temperature control method and system based on partitioned thermal field reconstruction, and relates to the technical field of furnace temperature feedback control. The method comprises the following steps: dividing a coating processing area of a photovoltaic cell in a high-vacuum furnace according to a temperature difference distribution, and determining a plurality of processing sub-areas; performing metal deposition rate analysis, and determining a standard deposition rate curve; based on the temperature control parameter threshold values of a plurality of heating elements in the high-vacuum furnace, taking the approximation of the standard deposition rate curve as the target, performing temperature control parameter optimization according to the plurality of processing sub-areas, and generating an optimal temperature control parameter scheme; controlling the high-vacuum furnace to coat the cell to be processed according to the optimal temperature control parameter scheme, and performing feedback regulation according to a preset PID control mechanism. The technical problems that the temperature control of the high-vacuum furnace is uneven during the coating process, and the processing precision is low and the coating quality is unstable in the prior art are solved, and the technical effects that the coating quality and the processing precision are improved through multi-point temperature control optimization are achieved.

Description

Technical Field

[0001] This invention relates to the field of furnace temperature feedback control technology, specifically to a high-vacuum furnace temperature control method and system based on partitioned thermal field reconstruction. Background Technology

[0002] The coating process of photovoltaic cells is one of the key steps in manufacturing high-efficiency solar cells, and the high-vacuum furnace, as the core equipment in the coating process, directly affects the quality and efficiency of the cells. The temperature control system within a high-vacuum furnace typically controls the temperature distribution within the furnace through heating elements, thereby influencing the metal deposition rate, film uniformity, and film quality during the coating process. However, existing high-vacuum furnaces often suffer from uneven temperature control in practical use, especially in managing the temperature of multiple heating elements and processing sub-areas. Due to uneven temperature distribution within the furnace, some areas exhibit significant temperature deviations, leading to fluctuations in the metal deposition rate during the coating process, which in turn affects the performance and quality of the photovoltaic cells. Low temperature control precision not only reduces processing accuracy but can also result in uneven film layers on the cells, affecting their photoelectric conversion efficiency. Existing high-vacuum furnace temperature control technologies mainly rely on single-point temperature control, making it difficult to achieve precise and comprehensive temperature regulation throughout the entire furnace space. Summary of the Invention

[0003] This application provides a high vacuum furnace temperature control method and system based on partitioned thermal field reconstruction, which solves the technical problem in the prior art that the temperature control of the high vacuum furnace is uneven during the coating process, resulting in low processing accuracy and unstable coating quality.

[0004] The first aspect of this application provides a high-vacuum furnace temperature control method based on partitioned thermal field reconstruction, the method comprising: The coating processing area of ​​photovoltaic cells in the high-vacuum furnace is divided according to the temperature difference distribution to determine multiple processing sub-regions. Based on the properties of the cells to be processed, the properties of the coating metal, the preset evaporation temperature, and the preset vacuum level, a metal deposition rate analysis is performed to determine a standard deposition rate curve. Based on the temperature control parameter thresholds of multiple heating elements in the high-vacuum furnace, with the goal of approximating the standard deposition rate curve, the temperature control parameters are optimized according to the multiple processing sub-regions to generate an optimal temperature control parameter scheme. The high-vacuum furnace is then controlled to coat the cells to be processed according to the optimal temperature control parameter scheme, and feedback regulation is performed according to a preset PID control mechanism.

[0005] A second aspect of this application provides a high-vacuum furnace temperature control system based on partitioned thermal field reconstruction, the system comprising: The system comprises the following modules: a region division module for dividing the photovoltaic cell coating processing area within the high-vacuum furnace according to temperature difference distribution, thus determining multiple processing sub-regions; an analysis module for analyzing the metal deposition rate based on the properties of the cells to be processed, the metal properties of the coating, the preset evaporation temperature, and the preset vacuum level, thus determining a standard deposition rate curve; an optimization module for optimizing the temperature control parameters based on the temperature control parameter thresholds of multiple heating elements within the high-vacuum furnace, aiming to approximate the standard deposition rate curve, and generating an optimal temperature control parameter scheme; and a control module for controlling the high-vacuum furnace to coat the cells to be processed according to the optimal temperature control parameter scheme, and for feedback regulation based on a preset PID control mechanism.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: First, the coating processing area of ​​photovoltaic cells in the high-vacuum furnace is divided according to the temperature difference distribution, determining multiple processing sub-regions. Next, based on the properties of the cells to be processed, the metal properties of the coating, the preset evaporation temperature, and the preset vacuum level, a metal deposition rate analysis is performed to determine a standard deposition rate curve. Then, based on the temperature control parameter thresholds of multiple heating elements in the high-vacuum furnace, with the goal of approximating the standard deposition rate curve, the temperature control parameters are optimized for multiple processing sub-regions to generate an optimal temperature control parameter scheme. Finally, the high-vacuum furnace is controlled according to the optimal temperature control parameter scheme to coat the cells to be processed, and feedback regulation is performed according to a preset PID control mechanism. This solves the technical problem in existing technologies where uneven temperature control during the coating process in high-vacuum furnaces leads to low processing accuracy and unstable coating quality, achieving the technical effect of improving coating quality and processing accuracy through multi-point temperature control optimization. Attached Figure Description

[0007] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0008] Figure 1 A schematic diagram of the high vacuum furnace temperature control method based on partitioned thermal field reconstruction provided in this application embodiment; Figure 2 A schematic diagram of the high vacuum furnace temperature control system based on partitioned thermal field reconstruction provided in this application embodiment.

[0009] Figure labeling: Region division module 11, analysis module 12, optimization module 13, control module 14. Detailed Implementation

[0010] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0011] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. 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 explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0012] Example 1, as Figure 1 As shown, this application provides a high-vacuum furnace temperature control method based on partitioned thermal field reconstruction, wherein the method includes: The coating processing area of ​​photovoltaic cells in the high vacuum furnace is divided according to the temperature difference distribution, and multiple processing sub-areas are determined.

[0013] In this embodiment, a temperature sensor is installed inside a high-vacuum furnace to monitor the coating processing area of ​​the photovoltaic cell in real time, collecting temperature data of this area and generating a temperature distribution map. The temperature distribution map can reflect the temperature differences at different locations on the surface of the photovoltaic cell, revealing the heating effect of each area and the uneven distribution of heat. Based on the temperature difference data, the coating processing area of ​​the photovoltaic cell is divided, and each area with a similar temperature distribution is grouped into a sub-region, resulting in multiple processing sub-regions.

[0014] Furthermore, the coating processing area of ​​photovoltaic cells in the high-vacuum furnace is divided according to the temperature difference distribution, and multiple processing sub-regions are determined, including: Based on the historical processing logs of the high vacuum furnace, multiple historical infrared temperature distribution maps of the coating processing area at the same node are collected. The coating processing area is divided into grids according to a preset grid step size to obtain a processing area grid map. Based on a preset temperature difference threshold and the processing area grid map, temperature parameters are extracted from the multiple historical infrared temperature distribution maps to determine multiple sample sub-region sets. If the temperature difference between adjacent grid areas is less than or equal to the preset temperature difference threshold, they are clustered into the same sub-region. Based on the multiple sample sub-region sets, the frequency proportion of grids within the same sub-region is calculated. If the frequency proportion is greater than a preset proportion threshold, they are clustered into the same processing sub-region, thus determining multiple processing sub-regions.

[0015] Preferably, multiple historical infrared temperature distribution maps at the same node (such as the same coating stage and the same heating power setting) are collected from the historical processing log of the high vacuum furnace. These historical infrared temperature distribution maps reflect the real-time temperature status of different locations in the furnace during the coating process. The coating processing area is divided into grids according to a preset grid step size, generating a processing area grid map and assigning a unique coordinate identifier to each grid unit. Based on a preset temperature difference threshold and the processing area grid map, the temperature value of the corresponding grid unit is extracted from each historical infrared temperature distribution map to form a temperature matrix. The temperature difference between adjacent grid units is calculated. If the absolute value of the temperature difference between adjacent grid areas is less than or equal to the preset temperature difference threshold, they are temporarily clustered into the same sub-region, thereby generating multiple sample sub-region sets. All sample sub-region sets are traversed, and the frequency of each grid unit being clustered into the same sub-region is counted. If a grid unit is clustered into the same sub-region in historical data exceeding a preset percentage threshold (such as 80%), it is permanently assigned to that sub-region, and grid units that meet the frequency threshold are merged to form the final multiple processing sub-regions.

[0016] Based on the properties of the battery cells to be processed, the properties of the coated metal, the preset evaporation temperature, and the preset vacuum level, the metal deposition rate is analyzed to determine the standard deposition rate curve.

[0017] The properties of the solar cells to be processed include their material type, surface finish, size, and shape. The properties of the metal to be coated include its physical properties, such as evaporation temperature, thermal conductivity, and melting point. Preset evaporation temperature and preset vacuum level are used as process parameters. By setting appropriate evaporation temperature and vacuum level, the evaporation and deposition process of the metal in the high-vacuum furnace is controlled. Too low an evaporation temperature may lead to insufficient metal deposition, while too high a temperature may lead to excessive evaporation or loss of metal. The vacuum level, on the other hand, affects the control of the metal evaporation rate and the deposition process.

[0018] Based on the properties of the battery cells to be processed, the properties of the coated metal, the preset evaporation temperature, and the preset vacuum level, metal deposition rate analysis is performed to determine a standard deposition rate curve, which reflects the trend of metal deposition rate under given process conditions.

[0019] Furthermore, based on the properties of the battery cells to be processed, the properties of the coated metal, the preset evaporation temperature, and the preset vacuum level, a metal deposition rate analysis is performed to determine a standard deposition rate curve, including: Constrained by the property characteristics of the battery cell to be processed, the property information of the coated metal, the preset evaporation temperature, and the preset vacuum degree, historical coating processing logs are retrieved to collect a set of sample metal deposition rate curves and a set of sample coating quality coefficients. The sample coating quality coefficients are determined by evaluating historical coating results. The set of sample coating quality coefficients is then filtered according to a preset quality coefficient threshold to obtain a set of qualified coating quality coefficients, which is then mapped to determine a set of qualified metal deposition rate curves. A standard deposition rate curve is determined based on the set of qualified metal deposition rate curves.

[0020] Specifically, the historical coating processing logs of the high-vacuum furnace are thoroughly searched, using constraints such as the properties of the battery cells to be processed (including but not limited to material type, size specifications, surface roughness, etc.), the properties of the coating metal (such as metal type, chemical purity, physical morphology, etc.), the preset evaporation temperature range, and the preset vacuum requirements. By filtering historical processing records that meet the above constraints, a set of sample metal deposition rate curves generated under the same or highly similar process environments is collected, along with the corresponding set of sample coating quality coefficients. The sample coating quality coefficient is a quantitative value determined by comprehensively evaluating key indicators such as film uniformity, metal adhesion, thickness accuracy, and surface defect rate of historical coating results. Subsequently, the set of sample coating quality coefficients is rigorously screened according to a preset quality coefficient threshold (e.g., a quality coefficient ≥ 0.9 represents excellent coating quality), eliminating substandard data points and retaining only the set of qualified coating quality coefficients. Based on this, data mapping technology is used to extract the metal deposition rate curves corresponding to the qualified coating quality coefficients, forming a set of qualified metal deposition rate curves. Finally, an in-depth analysis of the qualified metal deposition rate curve set was conducted. Methods such as statistical averaging, curve fitting, or machine learning were used to extract the common features and trends of each curve, eliminate abnormal fluctuations caused by accidental factors, and thus determine a standard deposition rate curve that can represent the optimal coating quality.

[0021] Furthermore, the qualified metal deposition rate curve set is mapped and distributed on a two-dimensional coordinate system, and the curves are smoothly fitted from early to late time to obtain the standard deposition rate curve.

[0022] Preferably, the selected set of qualified metal deposition rate curves is visualized by mapping these curves onto a two-dimensional coordinate system with time as the horizontal axis and deposition rate as the vertical axis. These curves are then sorted in chronological order to ensure that the influence of the time series on the deposition rate is fully considered during the fitting process. Subsequently, curve smoothing fitting techniques (such as polynomial fitting, spline interpolation, or moving average) are used to process the sorted curve set to eliminate abnormal data points caused by accidental factors such as equipment fluctuations and operational errors, while preserving and enhancing the common trends and periodic characteristics of the curves. Through smoothing fitting, a continuous and smooth standard deposition rate curve can be generated.

[0023] Based on the temperature control parameter thresholds of multiple heating elements in the high vacuum furnace, with the goal of approximating the standard deposition rate curve, the temperature control parameters are optimized according to the multiple processing sub-regions to generate the optimal temperature control parameter scheme.

[0024] The temperature control parameter thresholds for heating elements in a high-vacuum furnace include maximum heating power, temperature control range, and temperature response speed. Specifically, intelligent optimization algorithms (such as genetic algorithms, particle swarm optimization algorithms, or simulated annealing algorithms) are used to optimize temperature control parameters in parallel across multiple processing sub-regions. During the optimization process, the matching degree between the actual deposition rate and the standard deposition rate curve in each sub-region is used as the optimization objective function, while the temperature control parameter thresholds of the heating elements are considered as constraints to ensure that the generated temperature control parameters meet equipment requirements and effectively improve coating quality. The algorithm iteratively searches and continuously adjusts parameters such as the power distribution and heating time of the heating elements in each sub-region, gradually approaching the standard deposition rate curve until the preset optimization accuracy or number of iterations is reached. Finally, the generated optimal temperature control parameter scheme will specify the specific temperature control parameters of the heating elements in each processing sub-region, including but not limited to heating power and heating time, ensuring that each sub-region can accurately reproduce the standard deposition rate curve during the coating process, thereby achieving a significant improvement in key indicators such as coating thickness, uniformity, and adhesion.

[0025] Furthermore, based on the temperature control parameter thresholds of multiple heating elements within the high-vacuum furnace, and with the goal of approximating the standard deposition rate curve, the temperature control parameters are optimized according to the multiple processing sub-regions to generate an optimal temperature control parameter scheme, including: Multiple temperature control parameter thresholds for multiple heating elements are obtained, wherein the temperature control parameter is the heating power; multiple temperature control time nodes are configured according to a preset coating cycle, and parameters are randomly configured within the multiple temperature control parameter thresholds to generate a first temperature control parameter scheme, wherein the temperature control parameter scheme includes multiple heating power sequences; with the goal of approximating the standard deposition rate curve, the temperature control parameters are optimized according to the multiple processing sub-regions and the first temperature control parameter scheme to generate an optimal temperature control parameter scheme.

[0026] First, based on the equipment manual and historical experimental data, the temperature control parameter thresholds for the heating power of multiple heating elements in the high vacuum furnace are obtained, which are the minimum and maximum power of each heating element. Then, according to the coating process requirements, the entire coating cycle is divided into multiple temperature control time nodes (such as heating up and holding periods). Within the temperature control parameter threshold range of each heating element, an initial heating power value is generated for each time node using a random sampling method, forming multiple sets of first temperature control parameter schemes containing heating power sequences. Next, a coupled simulation model of the temperature field and metal deposition rate inside the high-vacuum furnace is constructed. Each set of first temperature control parameter schemes is input into the model to calculate the real-time temperature field and corresponding metal deposition rate curve of each processing sub-region during the coating cycle. The approximation degree of each scheme to the target curve is quantified by defining an objective function (such as the root mean square error (RMSE) between the actual deposition rate curve and the standard deposition rate curve). Then, an intelligent optimization algorithm (such as a genetic algorithm) is used to iteratively optimize the heating power sequence within the temperature control parameter threshold range with the objective function minimization as the goal. New temperature control parameter schemes are generated and the objective function value is re-evaluated until the convergence condition is met (such as reaching the maximum number of iterations or the objective function value is less than the preset threshold). Finally, the scheme with the smallest objective function value is selected as the optimal temperature control parameter scheme.

[0027] Furthermore, with the goal of approximating the standard deposition rate curve, the temperature control parameters are optimized based on the multiple processing sub-regions and the first temperature control parameter scheme to generate the optimal temperature control parameter scheme, including: Based on the first temperature control parameter scheme, the metal deposition rate is predicted for each of the multiple processing sub-regions to obtain a first set of predicted deposition rate curves. With the goal of approximating the standard deposition rate curve, the fitness of the first scheme is evaluated and determined based on the first set of predicted deposition rate curves. Using the multiple temperature control parameter thresholds as the optimization space, the temperature control parameter scheme is iteratively selected and its fitness is evaluated until a preset number of iterations is reached. The temperature control parameter scheme with the maximum fitness is then set as the optimal temperature control parameter scheme.

[0028] Based on the first temperature control parameter scheme (which includes the heating power sequence of each heating element at different temperature control time points), a coupled simulation model of the temperature field and metal deposition rate in a high-vacuum furnace is used to dynamically predict the metal deposition rate for multiple pre-divided processing sub-regions. By simulating the temperature changes and metal evaporation process in each sub-region during the coating cycle, a first set of predicted deposition rate curves corresponding to the first temperature control parameter scheme is obtained. Each curve reflects the trend of deposition rate change over time in a specific sub-region under a particular scheme. Subsequently, using the standard deposition rate curve as a benchmark, an objective function is defined to quantify the approximation degree between the first set of predicted deposition rate curves and the standard curve. For example, the root mean square error (RMSE) or correlation coefficient is used to calculate the fitness of the first scheme. The higher the fitness value, the closer the predicted deposition rate curve is to the standard curve, and the better the consistency and stability of the coating quality are guaranteed. Next, the temperature control parameter thresholds (such as the heating power range) of multiple heating elements are used as the optimization space, and intelligent optimization algorithms (such as genetic algorithms, particle swarm optimization algorithms, etc.) are used to iteratively select the temperature control parameter scheme. In each iteration, the algorithm generates new temperature control parameter schemes within the optimization space based on the current fitness evaluation results, and re-predicts deposition rate and re-evaluates fitness. Through continuous iterative optimization, it gradually approaches the optimal solution. When the number of iterations reaches a preset value or the fitness improvement approaches saturation, the temperature control parameter scheme with the maximum fitness is output and set as the optimal temperature control parameter scheme.

[0029] Furthermore, based on the first temperature control parameter scheme, the metal deposition rate is predicted for each of the multiple processing sub-regions to obtain a first set of predicted deposition rate curves, including: Based on historical processing logs, a mapping relationship between temperature control parameters and processing sub-region temperatures is established; based on the mapping relationship, multiple region temperature sequences of multiple processing sub-regions are obtained according to the first temperature control parameter scheme; a deposition rate prediction plugin is pre-trained; using the deposition rate prediction plugin, metal deposition rate is predicted according to the multiple region temperature sequences, and a first predicted deposition rate curve set is output.

[0030] Preferably, based on a large amount of measured data from historical processing logs, a precise mapping relationship is established between temperature control parameters (such as heating power) and the temperature of processing sub-regions. According to the pre-generated first temperature control parameter scheme (including the heating power settings of each heating element at different time points), the established mapping relationship is used to convert the temperature control parameters into regional temperature sequences of multiple processing sub-regions. The temperature sequence of each sub-region describes in detail the temperature change of that region over time during the coating cycle, providing a key input for deposition rate prediction. The multiple regional temperature sequences obtained by mapping are input into a pre-trained deposition rate prediction plugin to perform dynamic prediction of the metal deposition rate. The deposition rate prediction plugin calculates the corresponding deposition rate value in real time according to the temperature change of each sub-region and integrates them into a first set of predicted deposition rate curves.

[0031] Accuracy and efficiency of predictions are ensured by training a deposition rate prediction plugin. This plugin, based on machine learning algorithms (such as neural networks and support vector machines), is trained using historical temperature sequences and actual deposition rate data to learn the complex nonlinear relationship between temperature changes and deposition rate.

[0032] Furthermore, the pre-trained deposition rate prediction plugin includes: Based on historical coating processing logs, and constrained by the property characteristics of the battery cell to be processed, the property information of the coating metal, the preset evaporation temperature, and the preset vacuum degree, a sample area temperature sequence set and a sample metal deposition rate set are collected. Using the sample area temperature sequence set and the sample metal deposition rate set, a generative adversarial network is trained until the network converges, resulting in a deposition rate prediction plugin.

[0033] In the pre-training process of the deposition rate prediction plugin, the first step is to filter historical records from the historical coating processing logs that match the current cell's properties (such as material type and size specifications), coating metal properties (such as metal type and purity), preset evaporation temperature range, and preset vacuum requirements. Based on these constraints, corresponding sample area temperature sequence sets and sample metal deposition rate sets are collected. The sample area temperature sequence set records detailed temperature changes in each processing sub-region during the coating cycle, while the sample metal deposition rate set reflects the actual measured results of metal deposition rates under these temperature conditions. Next, a Generative Adversarial Network (GAN) is used as the core training framework. A GAN consists of a generator and a discriminator. The generator is responsible for generating a predicted deposition rate curve based on the input temperature sequence, while the discriminator distinguishes the generator's predicted curve from the actual deposition rate data. During training, the generator and discriminator continuously optimize their performance through adversarial learning: the generator strives to improve prediction accuracy to deceive the discriminator, while the discriminator enhances its recognition capabilities to drive the generator's improvement. Through iterative training with a large amount of sample data, when the predicted deposition rate curve generated by the generator can consistently fool the discriminator, and the discriminator cannot effectively distinguish between real and generated data, it indicates that the GAN has reached convergence. At this point, the trained generator can be used as a deposition rate prediction plugin, capable of accurately predicting the metal deposition rate based on the input temperature sequence.

[0034] Furthermore, with the goal of approximating the standard deposition rate curve, the fitness of the first scheme is evaluated and determined based on the first set of predicted deposition rate curves, including: Using the standard deposition rate curve as a benchmark, the deposition rate deviation at the same time point is calculated for the first predicted deposition rate curve set, and the mean is calculated to obtain the first rate deviation sequence; the rate fluctuation analysis at the same time point is performed on the first predicted deposition rate curve set, and the first fluctuation coefficient sequence is output, wherein the fluctuation coefficient is the ratio of the rate standard deviation to the rate mean at the same time point; multiple stage weights are configured according to the importance of the thin film deposition stage, and the first rate deviation sequence and the first fluctuation coefficient sequence are weighted and fused to output the first scheme fitness, wherein the scheme fitness is negatively correlated with the rate deviation and the fluctuation coefficient.

[0035] In evaluating the suitability of the first temperature control parameter scheme, the deposition rate deviation at the same time point is calculated for each curve in the first predicted deposition rate curve set, using a pre-determined standard deposition rate curve as a benchmark. Specifically, the difference between the predicted deposition rate and the standard deposition rate at each time point is calculated, and the absolute values ​​of these differences are averaged to obtain the first rate deviation sequence. This sequence directly reflects the degree of deviation between the predicted deposition rate and the standard curve under each scheme; the smaller the rate deviation, the closer the scheme is to the ideal state. Subsequently, rate fluctuation analysis is performed on the first predicted deposition rate curve set at the same time point. By calculating the ratio of the standard deviation to the mean of the predicted deposition rate at each time point (i.e., the fluctuation coefficient), the first fluctuation coefficient sequence is obtained. The fluctuation coefficient measures the stability of the deposition rate over time; the smaller the coefficient value, the smoother the rate change and the more stable the coating process. Based on the impact of rate deviation and fluctuation on coating quality, multiple stage weights are configured according to the importance of different stages of thin film deposition. For example, in the initial and final stages of coating, due to the higher requirements for thin film adhesion formation and surface quality, greater weights can be assigned. Subsequently, the first rate deviation sequence and the first fluctuation coefficient sequence are weighted and fused to calculate the fitness of the first scheme. The fitness of the scheme is negatively correlated with the rate deviation and the fluctuation coefficient. That is, the smaller the deviation and fluctuation, the higher the fitness value, indicating that the temperature control parameter scheme is better.

[0036] The high vacuum furnace is controlled to coat the battery cells to be processed according to the optimal temperature control parameter scheme, and feedback regulation is performed according to the preset PID control mechanism.

[0037] During the coating process in the high vacuum furnace, the temperature distribution of each processing sub-region is set and adjusted according to the optimal temperature control parameter scheme to ensure that the temperature distribution is as close as possible to the standard deposition rate curve and to achieve the best metal deposition rate.

[0038] During the coating process, an R-type thermocouple is used to construct a temperature monitoring array to monitor the temperature of each processing sub-area in the high vacuum furnace in real time. The temperature monitoring array collects temperature data of each area in real time and uses this temperature data as the basis for feedback control. Dynamic adjustment is performed through a preset PID (proportional-integral-derivative) control mechanism. The proportional control adjusts the heating power proportionally according to the deviation between the real-time data and the set value to quickly reduce the deviation. The integral control performs integral processing on the deviation to eliminate static error. The derivative control predicts the trend of deviation change and adjusts the power in advance to suppress the expansion of deviation. The three work together to ensure that the coating process is always close to the optimal state.

[0039] If the deviation exceeds a preset threshold, the PID control system will issue an adjustment command to promptly adjust the temperature control parameters. For example, if the temperature in a certain processing sub-area is too high or too low, the control system will automatically adjust the power output of the heating element in that area to quickly correct the deviation and ensure temperature uniformity and deposition rate consistency during the coating process.

[0040] In summary, the embodiments of this application have at least the following technical effects: First, the coating processing area of ​​photovoltaic cells in the high-vacuum furnace is divided according to the temperature difference distribution, determining multiple processing sub-regions. Next, based on the properties of the cells to be processed, the metal properties of the coating, the preset evaporation temperature, and the preset vacuum level, a metal deposition rate analysis is performed to determine a standard deposition rate curve. Then, based on the temperature control parameter thresholds of multiple heating elements in the high-vacuum furnace, with the goal of approximating the standard deposition rate curve, the temperature control parameters are optimized for multiple processing sub-regions to generate an optimal temperature control parameter scheme. Finally, the high-vacuum furnace is controlled according to the optimal temperature control parameter scheme to coat the cells to be processed, and feedback regulation is performed according to a preset PID control mechanism. This solves the technical problem in existing technologies where uneven temperature control during the coating process in high-vacuum furnaces leads to low processing accuracy and unstable coating quality, achieving the technical effect of improving coating quality and processing accuracy through multi-point temperature control optimization.

[0041] Example 2, based on the same inventive concept as the high vacuum furnace temperature control method based on partitioned thermal field reconstruction in the previous examples, such as... Figure 2 As shown, this application provides a high-vacuum furnace temperature control system based on partitioned thermal field reconstruction, wherein the system includes: The region division module 11 is used to divide the coating processing area of ​​photovoltaic cells in the high vacuum furnace according to the temperature difference distribution, and determine multiple processing sub-regions; the analysis module 12 is used to analyze the metal deposition rate based on the property characteristics of the cells to be processed, the property information of the coating metal, the preset evaporation temperature and the preset vacuum degree, and determine the standard deposition rate curve; the optimization module 13 is used to optimize the temperature control parameters based on the temperature control parameter thresholds of multiple heating elements in the high vacuum furnace, with the goal of approximating the standard deposition rate curve, according to the multiple processing sub-regions, and generate the optimal temperature control parameter scheme; the control module 14 is used to control the high vacuum furnace to coat the cells to be processed according to the optimal temperature control parameter scheme, and to perform feedback regulation according to the preset PID control mechanism.

[0042] Furthermore, the region division module 11 is used to perform the following method: Based on the historical processing logs of the high vacuum furnace, multiple historical infrared temperature distribution maps of the coating processing area at the same node are collected. The coating processing area is divided into grids according to a preset grid step size to obtain a processing area grid map. Based on a preset temperature difference threshold and the processing area grid map, temperature parameters are extracted from the multiple historical infrared temperature distribution maps to determine multiple sample sub-region sets. If the temperature difference between adjacent grid areas is less than or equal to the preset temperature difference threshold, they are clustered into the same sub-region. Based on the multiple sample sub-region sets, the frequency proportion of grids within the same sub-region is calculated. If the frequency proportion is greater than a preset proportion threshold, they are clustered into the same processing sub-region, thus determining multiple processing sub-regions.

[0043] Furthermore, the analysis module 12 is used to perform the following methods: Constrained by the property characteristics of the battery cell to be processed, the property information of the coated metal, the preset evaporation temperature, and the preset vacuum degree, historical coating processing logs are retrieved to collect a set of sample metal deposition rate curves and a set of sample coating quality coefficients. The sample coating quality coefficients are determined by evaluating historical coating results. The set of sample coating quality coefficients is then filtered according to a preset quality coefficient threshold to obtain a set of qualified coating quality coefficients, which is then mapped to determine a set of qualified metal deposition rate curves. A standard deposition rate curve is determined based on the set of qualified metal deposition rate curves.

[0044] Furthermore, the analysis module 12 is used to perform the following methods: The qualified metal deposition rate curve set is mapped onto a two-dimensional coordinate system, and the curves are smoothly fitted from early to late time to obtain the standard deposition rate curve.

[0045] Furthermore, the optimization module 13 is used to perform the following method: Multiple temperature control parameter thresholds for multiple heating elements are obtained, wherein the temperature control parameter is the heating power; multiple temperature control time nodes are configured according to a preset coating cycle, and parameters are randomly configured within the multiple temperature control parameter thresholds to generate a first temperature control parameter scheme, wherein the temperature control parameter scheme includes multiple heating power sequences; with the goal of approximating the standard deposition rate curve, the temperature control parameters are optimized according to the multiple processing sub-regions and the first temperature control parameter scheme to generate an optimal temperature control parameter scheme.

[0046] Furthermore, the optimization module 13 is used to perform the following method: Based on the first temperature control parameter scheme, the metal deposition rate is predicted for each of the multiple processing sub-regions to obtain a first set of predicted deposition rate curves. With the goal of approximating the standard deposition rate curve, the fitness of the first scheme is evaluated and determined based on the first set of predicted deposition rate curves. Using the multiple temperature control parameter thresholds as the optimization space, the temperature control parameter scheme is iteratively selected and its fitness is evaluated until a preset number of iterations is reached. The temperature control parameter scheme with the maximum fitness is then set as the optimal temperature control parameter scheme.

[0047] Furthermore, the optimization module 13 is used to perform the following method: Based on historical processing logs, a mapping relationship between temperature control parameters and processing sub-region temperatures is established; based on the mapping relationship, multiple region temperature sequences of multiple processing sub-regions are obtained according to the first temperature control parameter scheme; a deposition rate prediction plugin is pre-trained; using the deposition rate prediction plugin, metal deposition rate is predicted according to the multiple region temperature sequences, and a first predicted deposition rate curve set is output.

[0048] Furthermore, the optimization module 13 is used to perform the following method: Based on historical coating processing logs, and constrained by the property characteristics of the battery cell to be processed, the property information of the coating metal, the preset evaporation temperature, and the preset vacuum degree, a sample area temperature sequence set and a sample metal deposition rate set are collected. Using the sample area temperature sequence set and the sample metal deposition rate set, a generative adversarial network is trained until the network converges, resulting in a deposition rate prediction plugin.

[0049] Furthermore, the optimization module 13 is used to perform the following method: Using the standard deposition rate curve as a benchmark, the deposition rate deviation at the same time point is calculated for the first predicted deposition rate curve set, and the mean is calculated to obtain the first rate deviation sequence; the rate fluctuation analysis at the same time point is performed on the first predicted deposition rate curve set, and the first fluctuation coefficient sequence is output, wherein the fluctuation coefficient is the ratio of the rate standard deviation to the rate mean at the same time point; multiple stage weights are configured according to the importance of the thin film deposition stage, and the first rate deviation sequence and the first fluctuation coefficient sequence are weighted and fused to output the first scheme fitness, wherein the scheme fitness is negatively correlated with the rate deviation and the fluctuation coefficient.

[0050] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0051] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0052] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A high-vacuum furnace temperature control method based on partitioned thermal field reconstruction, characterized in that, The method includes: The coating processing area of ​​photovoltaic cells in the high vacuum furnace is divided according to the temperature difference distribution, and multiple processing sub-areas are determined. Based on the property characteristics of the battery cell to be processed, the property information of the coated metal, the preset evaporation temperature and the preset vacuum degree, the metal deposition rate is analyzed to determine the standard deposition rate curve. Based on the temperature control parameter thresholds of multiple heating elements in the high vacuum furnace, with the goal of approximating the standard deposition rate curve, the temperature control parameters are optimized according to the multiple processing sub-regions to generate the optimal temperature control parameter scheme. The high vacuum furnace is controlled to coat the battery cells to be processed according to the optimal temperature control parameter scheme, and feedback regulation is performed according to the preset PID control mechanism. Based on the temperature control parameter thresholds of multiple heating elements within the high-vacuum furnace, and aiming to approximate the standard deposition rate curve, the temperature control parameters are optimized according to the multiple processing sub-regions to generate an optimal temperature control parameter scheme, including: Obtain multiple temperature control parameter thresholds for multiple heating elements, where the temperature control parameter is the heating power; Multiple temperature control time nodes are configured according to a preset coating cycle, and parameters are randomly configured within the multiple temperature control parameter thresholds to generate a first temperature control parameter scheme, wherein the temperature control parameter scheme includes multiple heating power sequences. With the goal of approximating the standard deposition rate curve, temperature control parameters are optimized based on the multiple processing sub-regions and the first temperature control parameter scheme to generate an optimal temperature control parameter scheme; this optimization, based on the multiple processing sub-regions and the first temperature control parameter scheme, includes: Based on the first temperature control parameter scheme, the metal deposition rate is predicted for each of the multiple processing sub-regions to obtain a first set of predicted deposition rate curves; With the goal of approximating the standard deposition rate curve, the fitness of the first scheme is evaluated and determined based on the first set of predicted deposition rate curves. Using the multiple temperature control parameter thresholds as the optimization space, iterative selection and fitness evaluation of temperature control parameter schemes are performed until a preset number of iterations is reached. The temperature control parameter scheme with the maximum fitness is then set as the optimal temperature control parameter scheme.

2. The high-vacuum furnace temperature control method based on partitioned thermal field reconstruction according to claim 1, characterized in that, The coating processing area of ​​photovoltaic cells in the high vacuum furnace is divided according to the temperature difference distribution, and multiple processing sub-regions are determined, including: Based on the historical processing logs of the high vacuum furnace, multiple historical infrared temperature distribution maps of the coating processing area at the same node were collected. The coating processing area is divided into grids according to a preset grid step size to obtain a grid map of the processing area; Based on the preset temperature difference threshold and the processing area grid map, temperature parameters are extracted from the multiple historical infrared temperature distribution maps to determine multiple sample sub-region sets. If the temperature difference between adjacent grid regions is less than or equal to the preset temperature difference threshold, they are clustered into the same sub-region. The samples are divided into sub-region sets, and the frequency ratio of grids in the same sub-region is counted. If the frequency ratio is greater than a preset ratio threshold, they are clustered into the same processing sub-region, and multiple processing sub-regions are determined.

3. The high-vacuum furnace temperature control method based on partitioned thermal field reconstruction according to claim 1, characterized in that, Based on the property characteristics of the battery cell to be processed, the property information of the coated metal, the preset evaporation temperature, and the preset vacuum degree, the metal deposition rate is analyzed to determine the standard deposition rate curve, including: Constrained by the property characteristics of the battery cell to be processed, the property information of the coated metal, the preset evaporation temperature and the preset vacuum degree, the historical coating processing log is retrieved, and a set of sample metal deposition rate curves and a set of sample coating quality coefficients are collected. The sample coating quality coefficients are determined by evaluating historical coating results. The sample coating quality coefficient set is screened according to a preset quality coefficient threshold to obtain a qualified coating quality coefficient set, and a qualified metal deposition rate curve set is determined by mapping. The standard deposition rate curve is determined based on the analysis of the qualified metal deposition rate curve set.

4. The high-vacuum furnace temperature control method based on partitioned thermal field reconstruction according to claim 3, characterized in that, The qualified metal deposition rate curve set is mapped onto a two-dimensional coordinate system, and the curves are smoothly fitted from early to late time to obtain the standard deposition rate curve.

5. The high-vacuum furnace temperature control method based on partitioned thermal field reconstruction according to claim 1, characterized in that, Based on the first temperature control parameter scheme, the metal deposition rate is predicted for each of the multiple processing sub-regions to obtain a first set of predicted deposition rate curves, including: Based on historical processing logs, establish a mapping relationship between temperature control parameters and the temperature of processing sub-regions; Based on the mapping relationship, multiple temperature sequences of multiple processing sub-regions are obtained according to the first temperature control parameter scheme. Pre-trained deposition rate prediction plugin; Using the aforementioned deposition rate prediction plugin, metal deposition rate is predicted based on the temperature sequences of the multiple regions, and a first set of predicted deposition rate curves is output.

6. The high-vacuum furnace temperature control method based on partitioned thermal field reconstruction according to claim 5, characterized in that, The pre-trained deposition rate prediction plugin includes: Based on historical coating processing logs, and constrained by the property characteristics of the battery cell to be processed, the coating metal property information, the preset evaporation temperature, and the preset vacuum degree, a set of temperature sequences and a set of metal deposition rates for the sample area are collected. Using the temperature sequence set and metal deposition rate set of the sample region, a generative adversarial network is trained until the network converges, resulting in a deposition rate prediction plugin.

7. The high-vacuum furnace temperature control method based on partitioned thermal field reconstruction according to claim 1, characterized in that, With the goal of approximating the standard deposition rate curve, the fitness of the first scheme is evaluated and determined based on the first set of predicted deposition rate curves, including: Using the standard deposition rate curve as a benchmark, the deposition rate deviation at the same time point is calculated for the first predicted deposition rate curve set, and the mean value is used to obtain the first rate deviation sequence. The rate fluctuation analysis at the same time point is performed on the first predicted deposition rate curve set respectively, and the first fluctuation coefficient sequence is output, wherein the fluctuation coefficient is the ratio of the rate standard deviation to the rate mean at the same time point. Multiple stage weights are configured according to the importance of the thin film deposition stage. The first rate deviation sequence and the first fluctuation coefficient sequence are weighted and fused to output the first scheme fitness, wherein the scheme fitness is negatively correlated with the rate deviation and fluctuation coefficient.

8. A high-vacuum furnace temperature control system based on partitioned thermal field reconstruction, characterized in that, The system is used to implement the high-vacuum furnace temperature control method based on partitioned thermal field reconstruction as described in any one of claims 1-7, the system comprising: The region division module is used to divide the coating processing area of ​​photovoltaic cells in the high vacuum furnace according to the temperature difference distribution, and to determine multiple processing sub-regions; The analysis module is used to analyze the metal deposition rate based on the property characteristics of the battery cell to be processed, the property information of the coated metal, the preset evaporation temperature and the preset vacuum degree, and to determine the standard deposition rate curve. The optimization module is used to optimize the temperature control parameters based on the temperature control parameter thresholds of multiple heating elements in the high vacuum furnace, with the goal of approximating the standard deposition rate curve, and generate the optimal temperature control parameter scheme according to the multiple processing sub-regions. The control module is used to control the high vacuum furnace to coat the battery cells to be processed according to the optimal temperature control parameter scheme, and to perform feedback regulation according to the preset PID control mechanism.

Images