PVD (Physical Vapor Deposition) coating analysis control method and related device

Abstract

The invention discloses a PVD (Physical Vapor Deposition) coating analysis control method and a related device, which are used for stabilizing the chromatic aberration of a product. The method comprises the following steps: acquiring color difference values of a first preset number of heats before a current heat; obtaining a current reference point corresponding to the current heat according to the number of the heat within the color difference specification range in the color difference values of the first preset number of heat; if the current reference point is larger than the preset reference value, color layer data and interference layer data corresponding to the current heat are obtained; determining a recovery moment based on the color layer data and the interference layer data; and parameters of the current heat are adjusted according to the rising moment. In the embodiment of the invention, according to the association between the reference point and the recovery moment, the coating duration is intelligently regulated and controlled, the color difference of the product is accurately corrected, and the stability of the product quality is effectively improved; and the optimal process parameters are intelligently recommended according to the recovery moment, so that human intervention is greatly reduced, the color difference fluctuation is minimized, and the production efficiency and the intelligent management level are remarkably improved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, specifically to a PVD coating analysis and control method and related apparatus. Background Technology

[0002] When producing large quantities of products in batches using PVD coating, the typical process is as follows: During the initial setup phase, two sets of twin target structures are used. By pre-setting the reaction gas ratio and coating time, the basic process parameters are calibrated. When entering the furnace inter-furnace production stage, the operator needs to manually adjust the reaction gas ratio and coating time parameters based on the color difference test results of the previous furnace product; after the product color difference reaches the preset specification standard, the process switches to the normal machine adjustment procedure. During the normal commissioning phase, it is still necessary to continuously refer to the color difference data of the previous batch and manually determine whether to adjust the reaction gas ratio and coating time.

[0003] However, the above process has significant drawbacks in practical applications: even after the product color difference meets the specifications, the key parameters of the coating still need to be judged and adjusted manually. This process is greatly affected by the experience level of the operators, which leads to insufficient stability of the coating process and insufficient consistency of product color difference. The product color difference fluctuates greatly between different batches, making it difficult to meet the high-precision production requirements. Summary of the Invention

[0004] In view of this, this application provides a PVD coating analysis and control method and related apparatus to facilitate the stabilization of product color difference.

[0005] In a first aspect, embodiments of this application provide a PVD coating analysis and control method, the method comprising: Obtain the color difference value of the first preset number of furnaces preceding the current furnace batch; Based on the number of furnaces whose color difference values ​​fall within the color difference specification range from the first preset number of furnaces, the current reference point corresponding to the current furnace is obtained. If the current reference point is greater than the preset reference value, then the color layer data and interference layer data corresponding to the current furnace are obtained; The recovery time is determined based on the color layer data and the interference layer data; Adjust the parameters of the current furnace batch according to the recovery time.

[0006] In this embodiment, the coating time is intelligently adjusted based on the correlation between the reference point and the recovery time, which accurately corrects the product color difference and effectively improves the stability of product quality. Furthermore, the optimal process parameters are intelligently recommended based on the recovery time, which greatly reduces human intervention, minimizes color difference fluctuations, and significantly improves production efficiency and intelligent management level.

[0007] In some possible embodiments, obtaining the current reference point corresponding to the current furnace based on the number of furnaces whose color difference values ​​fall within the color difference specification range from the first preset number of furnaces includes: If the number of furnaces within the color difference specification range of the first preset number of furnaces is greater than or equal to the preset number of furnaces, then the recovery time corresponding to the furnace within the color difference specification range is obtained. The average value of the color difference at the recovery time corresponding to the batches within the specified color difference specification range is used as the current reference point; or... If the number of furnaces within the color difference specification range among the first preset number of furnaces is less than the preset number of furnaces, then the preset reference value is used as the current reference point.

[0008] In some possible embodiments, obtaining the color layer data and interference layer data corresponding to the current furnace batch includes: Collect online data for the current furnace cycle; The color layer data and the interference layer data are filtered from the online data.

[0009] In some possible embodiments, determining the rise time based on the color layer data and the interference layer data includes: The moving average voltage corresponding to each calculation moment within the first duration is determined based on the color layer data and the interference layer data; the first duration is the duration between the start time of the current furnace and the current time. The minimum voltage time corresponding to each calculation time is determined based on the moving average voltage at each calculation time. The recovery time is determined based on the minimum voltage time corresponding to each calculation time.

[0010] In some possible embodiments, determining the moving average voltage corresponding to each calculation time point within the first time duration based on the color layer data and the interference layer data includes: The first process is executed for each calculation time to obtain the moving average voltage corresponding to each calculation time. The first process includes: The calculation duration is obtained based on the calculation time and the preset time window length; Obtain the voltage value at each calculation moment within the calculation duration; The average voltage value at each calculation moment within the calculation period is taken as the moving average voltage corresponding to that calculation moment.

[0011] In some possible embodiments, determining the minimum voltage time corresponding to each calculation time based on the moving average voltage at each calculation time includes: For each calculation time, the second process is executed to obtain the minimum voltage time corresponding to each calculation time; The second process includes: Obtain the moving average voltage corresponding to each calculation time between the start time of the current furnace and the calculation time; The moving average voltages are sorted to obtain a moving average voltage sequence; A second preset number of target moving average voltages are selected sequentially from the moving average voltage sequence; Calculate the average value of the second preset number of target moving average voltages at the calculation time; The mean value is taken as the minimum voltage time corresponding to the calculation time.

[0012] In some possible embodiments, determining the recovery time based on the minimum voltage time corresponding to each calculation time includes: If the minimum voltage time corresponding to each calculation time determined within the second time period is the same, then the minimum voltage time is taken as the recovery time.

[0013] In some possible embodiments, adjusting the parameters of the current furnace batch according to the recovery time includes: The difference between the current benchmark point and the recovery time is taken as the target difference. The compensation ratio is obtained based on the target difference and the preset piecewise function; The control duration is obtained based on the difference between the compensation ratio and the target value; The control duration for the current furnace cycle is adjusted based on the control duration.

[0014] In some possible embodiments, the method further includes: Collect online data for the current furnace cycle; The online data is subjected to feature extraction processing to obtain the first feature value corresponding to the current furnace batch; The first feature value is subjected to parametric gridding to obtain multiple target feature value groups; The multiple target feature value groups are respectively input into the pre-trained color difference prediction model to obtain the color difference prediction value corresponding to each target feature value group output by the color difference prediction model. For each feature value group, the following steps are performed: A comprehensive color difference is obtained based on the color difference value of the furnace and the predicted color difference; The feature value group corresponding to the color difference prediction value with the smallest overall color difference is taken as the target feature value group; Adjust the parameters of the current furnace batch according to the target feature value set.

[0015] Secondly, embodiments of this application also provide a PVD coating analysis and control device, the device comprising: a memory and a processor, the memory being used to store instructions, the instructions stored in the memory being executed by the processor to implement the method described in any one of the first aspects.

[0016] Thirdly, embodiments of this application provide a system including a PVD device and the PVD coating analysis and control device described in the second aspect, wherein the PVD device receives adjusted parameters of the current batch or instructions to adjust the parameters of the current batch sent by the PVD coating control device.

[0017] Fourthly, embodiments of this application provide an electronic device, including a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, the electronic device is triggered to execute the method described in any one of the first aspects.

[0018] Fifthly, embodiments of this application provide a computer-readable storage medium including a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to perform the method described in any one of the first aspects. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A schematic diagram of a PVD equipment for a PVD coating analysis and control method provided in this application embodiment; Figure 2 This application provides an overall flowchart of a PVD coating analysis and control method. Figure 3 A flowchart illustrating the process of obtaining the current reference point corresponding to the current batch in a PVD coating analysis and control method provided in this application embodiment; Figure 4 A schematic diagram illustrating the process of determining the recovery time based on color layer data and interference layer data in a PVD coating analysis and control method provided in this application embodiment; Figure 5 This is a schematic diagram of the first process of a PVD coating analysis and control method provided in an embodiment of this application; Figure 6 A schematic diagram of the time window for a PVD coating analysis and control method provided in this application embodiment; Figure 7 This is a schematic diagram of the second process of a PVD coating analysis and control method provided in an embodiment of this application; Figure 8 A flowchart illustrating a PVD coating analysis and control method provided in this application, which adjusts the parameters of the current furnace based on the recovery time; Figure 9 A schematic diagram illustrating the automatic adjustment of PVD equipment parameters in a PVD coating analysis and control method provided in this application embodiment; Figure 10 This is a schematic diagram of the training process for a color difference prediction model of a PVD coating analysis and control method provided in an embodiment of this application. Detailed Implementation

[0021] To better understand the technical solution of this application, the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0022] It should be understood that the described embodiments are merely some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0023] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. The singular forms “a,” “the,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0024] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0025] In the parameter debugging and production process of the coating process, two sets of twin target structures are used in the initial machine adjustment stage. The basic process parameters are calibrated by pre-setting the reaction gas ratio and coating time.

[0026] When entering the inter-furnace production stage, operators need to manually adjust the reaction gas ratio and coating time parameters based on the color difference test results of the previous batch of products; after the product color difference reaches the preset specification standard, the process switches to the normal machine adjustment procedure.

[0027] During the normal commissioning phase, it is still necessary to continuously refer to the color difference data of the previous batch and manually determine whether to adjust the reaction gas ratio and coating time.

[0028] However, the above process has significant drawbacks in practical applications: even after the product color difference meets the specifications, the key parameters of the coating still need to be judged and adjusted manually. This process is greatly affected by the experience level of the operators, which leads to insufficient stability of the coating process and insufficient consistency of product color difference. The product color difference fluctuates greatly between different batches, making it difficult to meet the high-precision production requirements.

[0029] To address the aforementioned problems, this application provides a Physical Vapor Deposition (PVD) coating analysis and control method and related apparatus to solve these issues. The inventive concept of this application can be summarized as follows: obtaining the color difference values ​​of a first preset number of previous batches; based on the number of batches whose color difference values ​​fall within the color difference specification range from the first preset number of batches, obtaining the current reference point corresponding to the current batch; if the current reference point is greater than a preset reference value, obtaining the color layer data and interference layer data corresponding to the current batch; determining the recovery time based on the color layer data and interference layer data; and adjusting the parameters of the current batch according to the recovery time.

[0030] In this embodiment, the coating time is intelligently adjusted based on the correlation between the reference point and the recovery time, which accurately corrects the product color difference and effectively improves the stability of product quality. Furthermore, the optimal process parameters are intelligently recommended based on the recovery time, which greatly reduces human intervention, minimizes color difference fluctuations, and significantly improves production efficiency and intelligent management level.

[0031] For ease of understanding, the PVD coating analysis and control method and related apparatus provided in this application will be described in detail below with reference to the accompanying drawings: First, the PVD device in the embodiments of this application will be described, such as... Figure 1The diagram shows the structure of a PVD (Polymer Dioxide) device. A PVD device is a machine that uses electrodes to transform gases (including but not limited to argon (Ar), nitrogen (N2), and acetylene (C2H2)) into plasma, which then impacts a target material (including but not limited to chromium (Cr), silicon (Si), and titanium (Ti)) to sputter metals or compounds onto a metal product. A PVD device includes at least: a target power supply, a target rotary motor, a thermometer, a heating controller, a compressive strength gauge, a capacitance gauge, a vacuum pump controller, a gas flow controller, a cooling water controller, a frequency converter, a product rotary motor, a product power supply, and a network module; wherein: The target power supply is used to control the intermediate frequency of the twin target using an AC power supply. The target rotation motor is used as a fixed-speed motor to control the rotation of the target; The thermometer is used to measure furnace body temperature, target material temperature, etc.; it facilitates the network module to collect the target material temperature and furnace body temperature based on the thermometer's measurement value, so as to calculate the recovery time of the current furnace cycle based on the online data collected by the network module. The heating controller is used for: the heating rod controller inside the furnace. When the temperature of the workpiece is low when it enters the furnace, the output power of the heating controller will be automatically adjusted and the parameter changes during the process will be recorded to ensure that the furnace temperature reaches the set value quickly. The pressure gauge is used to: measure the pressure of the furnace body when it is not powered on; and compare the pressure when it is not powered on with the initial vacuum threshold set for the furnace to determine the pre-vacuuming effect of the vacuum pump. Capacitive pressure gauges are used to measure the pressure of a furnace when it is powered on. Excessive pressure fluctuations inside the furnace can lead to uneven film deposition. The pressure value measured by the capacitive pressure gauge can be used to dynamically control the air pump. The air pump controller includes: large coarse pump, small coarse pump, and 3 fine pumps, used to: reduce the pressure in the furnace; if the humidity of the workpiece is high when it enters the furnace, there will be more water vapor in the furnace, so the pre-evacuation time of the air pump can be optimized according to the humidity of the workpiece when it enters the furnace; during the workpiece coating process, the air pump can be dynamically adjusted to reduce the pressure fluctuation in the furnace. The gas flow controller is used to control the intake of argon, nitrogen, and acetylene respectively, so as to adjust the acetylene value corresponding to the color layer and the acetylene value corresponding to the interference layer. The cooling water controller is used to: control the temperature of the cooling water flowing through the 6 sets of targets and the furnace body; ensure that the furnace body temperature is stable within the set value range, and avoid affecting the quality of the film layer due to overheating of the furnace body; Inverters are used to control the speed of the rotating motor of a product; if the speed is too fast or too slow, it will cause uneven film deposition. The inverter is adjusted to make the motor run within the optimal speed range. The product rotary motor is used to rotate the product, including servo motors or stepper motors; it operates within the optimal speed range under the control of the frequency converter to ensure the uniformity of the film layer. The product power supply is used to: bring the product to the same potential, and has a settable duty cycle; The network module includes a communication system that is electrically connected to the target rotary motor, target power supply, thermometer, heating control, compatibility gauge, capacitance gauge, air pump controller, gas flow controller, cooling water controller, frequency converter, product rotary motor, and product power supply. Specifically, it can be connected via analog control and digital communication. The communication system is used to collect online data and send it to the PVD coating analysis and control device. Online data includes, but is not limited to: target temperature, workpiece temperature and humidity upon entering the furnace, acetylene value corresponding to the color layer, acetylene value corresponding to the interference layer, coating time corresponding to the interference layer, furnace setpoint, material quantity, and furnace pressure.

[0032] The aforementioned controllers can be integrated into a PLC (Programmable Logic Controller) system, and the PLC system can also integrate some of the aforementioned controllers according to actual needs.

[0033] like Figure 2 The diagram shown is an overall flowchart of a PVD coating analysis and control method provided in an embodiment of this application. The PVD coating analysis and control method of this application is applied to a PVD coating analysis and control device, wherein the PVD coating analysis and control method includes: In step 201: Obtain the color difference value of the first preset number of furnaces before the current furnace.

[0034] In this embodiment of the application, before the PVD equipment is about to perform PVD on the products in the current furnace, the PVD coating analysis and control device obtains and analyzes the color difference values ​​of the first preset number of furnaces before the current furnace. Since different reference points are set according to the number of furnaces with different color difference values ​​within the color difference specification range, it is necessary to obtain the color difference values ​​of the first preset number of furnaces when determining the current reference point of the current furnace.

[0035] The first preset quantity can be set to any value between 2 and 1000, or even a value exceeding 1000. Optionally, the first preset quantity can be set to 5.

[0036] It should be noted that the specific value of the first preset quantity given above is only one embodiment and is not intended to limit the specific value of the first preset quantity. In specific implementation, the value of the first preset quantity can be set according to the requirements.

[0037] In step 202: Based on the number of furnaces whose color difference values ​​fall within the color difference specification range from the first preset number of furnaces, the current reference point corresponding to the current furnace is obtained.

[0038] In this embodiment of the application, if the color difference value of a large number of furnaces in the first preset number of furnaces falls within the color difference specification range, it indicates that the recovery time corresponding to the multiple furnaces is of reference value; otherwise, it is not of reference value.

[0039] In other embodiments, the current reference point corresponding to the current furnace can be obtained based on the number of furnaces whose color difference values ​​are outside the color difference specification range from the first preset number of furnaces. Specifically, the current reference point corresponding to the current furnace can also be obtained by analyzing the number of furnaces whose color difference values ​​are outside the color difference specification range from the first preset number of furnaces. For ease of understanding, this application uses "the number of furnaces whose color difference values ​​are within the color difference specification range from the first preset number of furnaces" for analysis. Based on this, in some possible embodiments, the current reference point corresponding to the current furnace is obtained based on the number of furnaces whose color difference values ​​are within the color difference specification range from the first preset number of furnaces. Specifically, this can be implemented as follows: Figure 3 The steps shown are as follows: In step 301: Determine whether the number of furnaces with color difference values ​​within the color difference specification range of the first preset number of furnaces is greater than or equal to the preset number of furnaces. If it is greater than or equal to the preset number, proceed to step 302; otherwise, proceed to step 304. In this embodiment, if the color difference values ​​of furnaces with color difference values ​​greater than or equal to the preset number of furnaces are within the color difference specification range, it indicates that a large number of furnaces have color difference values ​​within the color difference specification range. Therefore, the recovery time corresponding to the furnaces with color difference values ​​within the color difference specification range is more reliable, and the reference point of the current furnace can be determined based on the recovery time of these furnaces.

[0040] In step 302: Obtain the recovery time corresponding to the furnace batch with color difference value within the color difference specification range.

[0041] In this embodiment, the recovery time refers to the point in time when the temperature of the coating chamber begins to rise after the cooling phase. This is determined by a continuous positive change in the temperature monitoring data of the chamber (or target material, substrate). The recovery time for each batch can be obtained through data analysis software in the PVD equipment.

[0042] In step 303: the average value of the color difference value at the recovery time corresponding to the furnace batch within the color difference specification range is used as the current reference point.

[0043] In this embodiment of the application, when a large number of furnace batches have color difference values ​​falling within the color difference specification range, the average value of the recovery time corresponding to the furnace batches with multiple color difference values ​​within the color difference specification range can reflect the overall level of these furnace batches. Moreover, the average value has stability and strong representativeness, so the average value is used as the current benchmark.

[0044] In some possible embodiments, the color difference specification range can be any percentage within the range of 0.1%, 5%, 10%, 20%, 30%, and 40% of the color difference specification. Optionally, the color difference specification range can be within 30% of the color difference specification. Specifically, the color difference specification range can be as shown in Formula 1, where: , (Formula 1) in, For color difference specifications, This is the upper limit of the deviation. This is the lower limit of the deviation.

[0045] It should be noted that the specific values ​​of color difference specifications, upper limit of deviation, and lower limit of deviation can be determined by technicians based on the workpiece being processed. This application does not limit the specific values ​​of color difference specifications, upper limit of deviation, and lower limit of deviation.

[0046] In step 304: the preset reference value is used as the current reference point.

[0047] In this embodiment of the application, when the color difference value of a small number of furnaces falls within the color difference specification range, it indicates that the recovery time corresponding to those furnaces is unreliable. Therefore, a preset reference value can be used as the current reference point.

[0048] For example: the first preset quantity is 5, the preset batch quantity is 3, and the color difference values ​​of the most recent 5 batches are obtained. It is determined that the color difference values ​​of 3 of these batches are within the color difference specification range. The batch quantity with color difference values ​​within the specification range is 3, which is equal to the preset batch quantity of 3. Therefore, the average of the recovery times of the 3 batches with color difference values ​​within the specification range can be used as the current reference point. Assume that the recovery times of the most recent 5 batches are { , , , , }, of which furnace number { If the color difference value falls within the color difference specification range, then the current reference point for the current batch is... for .

[0049] For example: the first preset quantity is 5, the preset furnace batch quantity is 3, and the preset baseline value is 0. The color difference values ​​of the most recent 5 furnace batches are obtained, and it is determined that the color difference values ​​of 2 of these furnace batches are within the color difference specification range. The furnace batch with a color difference value within the specification range is 2, which is less than the preset furnace batch quantity of 3. Therefore, the preset baseline value is used as the current baseline point for the current furnace batch. It is 0.

[0050] In step 203: If the current reference point is greater than the preset reference value, then obtain the color layer data and interference layer data corresponding to the current furnace.

[0051] In this embodiment, when the current reference point is greater than the preset reference value, it indicates that the coating time needs to be adjusted to correct the product color difference. When the current reference point is equal to the preset reference value, there is no need to continue with the subsequent steps.

[0052] In some possible embodiments, obtaining the color layer data and interference layer data corresponding to the current furnace batch can be specifically implemented as follows: collecting online data of the current furnace batch; and filtering out the color layer data and interference layer data from the online data.

[0053] In this embodiment, the PVD equipment can use a PLC system to record the data of the coating process in real time and upload it to a database (databases include but are not limited to: DB, Kafka, Redis, etc.); then the PVD coating analysis and control device reads the data uploaded by the PLC system from the Kafka streaming data in real time.

[0054] In this embodiment of the application, it is necessary to determine the recovery time based on the color layer data and the interference layer data. Therefore, it is necessary to determine the color layer data and the interference layer data from each online data.

[0055] To determine whether a layer in online data is a color layer, the following four conditions must be met simultaneously. If a layer of data meets all four conditions, then that layer can be considered color layer data. Time range limitation: Data prior to the first preset time in the formation process of this layer; the first preset time can be any value between 30 and 60 seconds, and optionally, the preset time can be 50 seconds; Acetylene threshold: If the monitored acetylene value is greater than a preset value before a preset time, the preset value can be any value between 7 and 15, and optionally, the preset value is 10. Layer time setting requirements: The overall preset process time of this layer is greater than the second preset time. The first preset time can be any value between 7000 and 10000 seconds. Optionally, the preset time can be 9000 seconds. Acetylene value diversity requirement: Within the first 50 seconds, the number of different acetylene values ​​detected should not be less than 10.

[0056] Based on the confirmed color layer data, the next adjacent film layer is taken as the interference layer. For example, if the film layers of a workpiece are, in sequence, the bottom layer, the color layer, the top layer, and the surface layer, and the second layer has been determined to be the color layer, then the third layer can be directly identified as the interference layer without the need for additional parameter determination.

[0057] In step 204: the recovery time is determined based on the color layer data and the interference layer data.

[0058] In some possible embodiments, step 204 can be specifically implemented as follows: Figure 4 The process shown is as follows: In step 401: the moving average voltage corresponding to each calculation moment within the first time period is determined based on the color layer data and the interference layer data; the first time period is the duration between the start time of the current furnace and the current time.

[0059] In this embodiment, the moving average voltage is the voltage value obtained by averaging voltage data over a continuous period of time using a sliding window method. The moving average voltage can reflect and smooth out instantaneous voltage fluctuations, characterizing the overall trend of voltage change over a certain time scale. The calculation time is the moment within the first time period at which the moving average voltage needs to be calculated, determined according to a pre-set period.

[0060] In some possible embodiments, the moving average voltage corresponding to each calculation moment within the first time period is determined based on the color layer data and the interference layer data in step 401 above. Specifically, this can be implemented by performing the following steps for each calculation moment: Figure 5 The first process shown yields the moving average voltage at each calculation time point, where: In step 501: the calculation duration is obtained based on the calculation time and the preset time window length.

[0061] In one embodiment of the application, the preset time window length can be set to 900 seconds, and the calculation time is the execution time. Figure 5 The time shown in the first process is recorded as the calculation time. The calculation time is .

[0062] It should be noted that the specific value of the preset time window length is only for illustrative purposes and is not intended to limit the length of the preset time window. In actual implementation, you can set the specific value of the preset time window length yourself.

[0063] In step 502: Obtain the voltage value at each calculation moment within the calculation duration.

[0064] In the embodiments of this application, a voltage acquisition device connected to the target can be used to obtain the voltage value at each calculation moment, or the monitoring system corresponding to the PVD device can be used to obtain the voltage value at each calculation moment. This application does not limit the method of obtaining the voltage value, and the method can be selected by the user in specific implementation.

[0065] In step 503: the average voltage value at each calculation time within the calculation period is used as the moving average voltage at the corresponding calculation time.

[0066] For example: Figure 6 As shown, the calculation time is The preset time window length is 900 seconds, and the calculation time is... The computation time contains 900 computation moments, which means it is executed once per second. Figure 5 The process shown calculates the time. The corresponding moving average voltage can be determined using formula 2, where: , (Formula 2) in: For calculating time The corresponding moving average voltage, For the first The voltage value of the second target.

[0067] In step 402: the minimum voltage time corresponding to each calculation time is determined based on the moving average voltage corresponding to each calculation time.

[0068] In this embodiment of the application, the minimum voltage time corresponding to each calculation time is the average of the moving average voltage of a second preset number of calculation times with the minimum moving average voltage between the start time of the current furnace and the calculation time.

[0069] In some possible embodiments, in step 402 above, the minimum voltage time corresponding to each calculation time is determined based on the moving average voltage corresponding to each calculation time. Specifically, a second process can be performed for each calculation time to obtain the minimum voltage time corresponding to each calculation time; the second process can be implemented as follows: Figure 7 The process shown is as follows: In step 701: Obtain the moving average voltage corresponding to each calculation time between the start time of the current furnace and the calculation time.

[0070] In this embodiment of the application, determining the moving average voltage corresponding to each calculation time can be specifically implemented as described above. Figure 5 The steps shown will not be repeated here.

[0071] In step 702: the moving average voltages are sorted to obtain a moving average voltage sequence.

[0072] In this embodiment of the application, for each calculation time For the start time and calculation time of the current furnace cycle The moving average voltages are sorted according to their respective values, and the resulting sequence is the moving average voltage sequence, as shown in Formula 3: , (Formula 3) in, For moving average voltage sequence To arrange the time intervals from smallest to largest The moving average voltage at each calculation time point is sorted. For calculating time The corresponding moving average voltage.

[0073] In step 703: Select a second preset number of target moving average voltages sequentially from the moving average voltage sequence.

[0074] In this embodiment of the application, a second preset number of moving average voltages are selected from the moving average voltage sequence in ascending order as the target moving average voltage.

[0075] For example, if the second preset quantity is 200, then the second preset quantity of target moving average voltages is as shown in Formula 4: , (Formula 4) in, The calculation times are the corresponding to the second preset number of target moving average voltages. The second preset number of target moving average voltages is 200, where 200 is the second preset number.

[0076] In step 704: Calculate the average value of the calculation time corresponding to the second preset number of target moving average voltages.

[0077] In step 705: the mean value is used as the minimum voltage time corresponding to the calculation time.

[0078] In this embodiment of the application, step 704 can be implemented as formula 5, wherein: , (Formula 5) in, For calculating time The corresponding minimum voltage moment, The calculation times are the corresponding to the second preset number of target moving average voltages. To calculate the average value of the moving average voltage of the second preset number of targets at the corresponding calculation times.

[0079] In step 403: the recovery time is determined based on the minimum voltage time corresponding to each calculation time.

[0080] In this embodiment of the application, if the minimum voltage time corresponding to each calculation time determined within the second time period is the same, then the minimum voltage time is taken as the recovery time.

[0081] For example, if the second duration is 300 seconds, then the minimum voltage time corresponding to each calculation time determined within 300 seconds is the same, and the same minimum voltage time within 300 seconds is taken as the recovery time.

[0082] In step 205: Adjust the parameters of the current furnace according to the recovery time.

[0083] In this embodiment of the application, the recovery control duration corresponding to the current furnace is determined based on the recovery time, and then the current furnace is adjusted based on the determined recovery control duration.

[0084] In some possible embodiments, the parameters of the current furnace are adjusted according to the recovery time, specifically as follows: Figure 8 The process shown is as follows: In step 801: the difference between the current benchmark point and the recovery time is taken as the target difference.

[0085] In this embodiment of the application, Formula 6 can be used to determine the target difference, wherein: , (Formula 6) in, For the target difference, For the time of recovery, This is the current benchmark point.

[0086] In step 802: the compensation ratio is obtained based on the target difference and the preset piecewise function.

[0087] In this embodiment of the application, in order to make the obtained control duration more accurate, different compensation ratios are set for different target differences. Specifically, the compensation ratio can be determined using Formula 7, where: , (Formula 7) in, For the compensation ratio, The target difference.

[0088] In step 803: the control duration is obtained based on the difference between the compensation ratio and the target.

[0089] In this embodiment of the application, the control duration can be obtained using Formula 8, wherein: , (Formula 8) in, For the duration of the control, For the compensation ratio, The target difference.

[0090] In step 804: Adjust the control duration of the current furnace cycle according to the control duration.

[0091] In this embodiment of the application, after obtaining the return control duration Subsequently, the PVD coating analysis and control device feeds back the control time to the PVD equipment so that the control time can be written into the PLC system to control the coating end time and correct color difference.

[0092] In some possible embodiments, after approximately 30-60 heats, the target material may age or experience performance degradation due to continuous use (e.g., increased target voltage fluctuations, unstable coating effects, and increased color difference deviations). Target replacement and equipment maintenance are necessary to ensure production line stability and product quality. To guarantee color difference stability, the above-mentioned procedures can be performed after target replacement or equipment maintenance. Figure 2 The steps shown correct the return time of the PVD equipment to ensure the stability of the product color.

[0093] In other possible embodiments, PVD coating technology requires multiple precision processes within a sealed furnace, including evacuation, pressure holding, etching, color layer coating, and interference layer coating. The process parameters for each process traditionally rely on manual experience for pre-setting. Due to the subjectivity and limitations of manual experience, existing processes struggle to achieve precise parameter control over complex coating flows, easily leading to color difference fluctuations and failing to meet the high-precision, high-stability requirements of industrial production. Even in the intelligent feedback control stage, the technical solution can rely on real-time target voltage parameters to calculate the reference point and dynamically predict the recovery point time to correct the coating duration, achieving preliminary precise control of color difference. However, core process parameters such as the acetylene value of the color layer, the acetylene value of the interference layer, and the coating time of the interference layer still fall within the scope of manual experience setting, becoming a key bottleneck restricting further optimization of the production process. Based on this, this application also provides a PVD coating analysis and control method to achieve automatic adjustment of PVD equipment parameters, such as... Figure 9 As shown, where: In step 901: Collect online data for the current furnace cycle.

[0094] In this embodiment of the application, considering that the temperature and humidity of the workpiece when it enters the furnace will affect the color difference of the current furnace, the online data to be collected must include at least the temperature and humidity of the workpiece when it enters the furnace. In order to recommend the parameters of the color layer and the interference layer, it is also necessary to collect the parameters of the color layer and the interference layer. The online data to be collected must also include the acetylene value corresponding to the color layer, the acetylene value corresponding to the interference layer, and the coating time corresponding to the interference layer. In order to further improve the accuracy of the determined parameters, the online data to be collected may also include parameters such as furnace set value and material quantity.

[0095] In step 902: Feature extraction processing is performed on the online data to obtain the first feature value corresponding to the current furnace batch.

[0096] In this embodiment of the application, in order to further improve accuracy, the difference between the color layer and interference layer parameters between the current furnace batch and the previous furnace batch can be determined, and the difference is used together with other parameters as the first feature value; that is, the first feature value includes: the temperature and humidity of the workpiece when it enters the furnace, the difference between the acetylene value corresponding to the color layer and the acetylene value corresponding to the color layer in the previous furnace batch, the difference between the acetylene value corresponding to the interference layer and the acetylene value in the previous furnace batch, the difference between the coating time corresponding to the interference layer and the coating time corresponding to the interference layer in the previous furnace batch, the furnace batch setting value, and the material quantity.

[0097] In step 903: the first feature value is subjected to parametric gridding to obtain multiple target feature value groups.

[0098] In this embodiment of the application, the first feature value is denoted as: Parametric meshing is performed on some or all of the first feature values ​​to obtain multiple target feature value groups.

[0099] For example: first eigenvalue Among them The difference in acetylene values ​​between the color layers. The difference in acetylene values ​​between the interference layers. The time difference of the interference layer coating is used to determine the first characteristic value. , , After parametric mesh processing, the resulting eigenvalue sets are as follows: ; ; ; After performing parametric mesh processing on each first eigenvalue, the resulting set of multiple target eigenvalues ​​is denoted as: ,in, For the first of multiple sets of target feature values Group target feature values, For the first of multiple sets of target feature values Group corresponding The value of ( The first in the corresponding parameter grid indivual), For the first of multiple sets of target feature values Group corresponding The value of ( The first in the corresponding parameter grid indivual), For the first of multiple sets of target feature values Group corresponding The value of ( The first in the corresponding parameter grid indivual).

[0100] In step 904: multiple target feature value groups are input into the pre-trained color difference prediction model to obtain the color difference prediction value corresponding to each target feature value group output by the color difference prediction model.

[0101] In this embodiment of the application, multiple target feature value groups are input into a pre-trained color difference prediction model. The color difference prediction model outputs a corresponding color difference prediction value for each group of target feature values, thus obtaining multiple groups of color difference prediction values.

[0102] For example: ,in For the first of multiple sets of target feature values Group target feature values, The number of target feature groups, for The corresponding color difference prediction value.

[0103] In step 905: For each feature value group, the color difference prediction value is executed: the comprehensive color difference is obtained based on the color difference value of the furnace and the color difference prediction value.

[0104] In this embodiment of the application, the color difference value of the previous batch is obtained, and the color difference prediction value corresponding to each feature value group obtained by superimposing the color difference value of the previous batch is recorded as the comprehensive color difference corresponding to each feature value group.

[0105] For example: the color difference value after standardization in the previous batch was... , No. The color difference prediction value corresponding to the group target feature value is Then the first The comprehensive color difference corresponding to the group target feature values ​​is: .

[0106] In step 906: the feature value group corresponding to the color difference prediction value with the smallest overall color difference is taken as the target feature value group.

[0107] In this embodiment, to ensure that the selected parameters are optimal, the feature value group corresponding to the color difference prediction value with the minimum overall color rubbing and sweating is taken as the target feature value group. That is, it can be obtained by formula 9 (in formula 9, only the feature values ​​are considered). , , (Taking the eigenvalue set obtained after parametric mesh processing as an example) to determine the target eigenvalue.

[0108] , (Formula 9) in, This refers to the color difference value after the previous round of standardization. For the first Color difference prediction value corresponding to the group target feature value. For the first of multiple sets of target feature values Group corresponding The value of ( The first in the corresponding parameter grid indivual), For the first of multiple sets of target feature values Group corresponding The value of ( The first in the corresponding parameter grid indivual), For the first of multiple sets of target feature values Group corresponding The value of ( The first in the corresponding parameter grid indivual) In step 907: Adjust the parameters of the current furnace according to the target feature value group.

[0109] In this embodiment of the application, after obtaining the target feature value group, the parameters corresponding to the current furnace are adjusted according to the parameters of the previous furnace and the target feature values.

[0110] For example: Determine the target feature value set as , The acetylene values ​​of the color layer, interference layer, and coating time of the interference layer in the previous batch were as follows: , , In the target feature set The acetylene value for the color layer. The value of acetylene in the interference layer. Given the interference layer coating time, the acetylene value of the color layer corresponding to the current furnace batch is... + The acetylene value of the interference layer corresponding to the current furnace batch is The current furnace batch corresponds to an interference layer coating time of 10 minutes. .

[0111] In some possible embodiments, the color difference prediction model in step 904 above can be an XGBoost model; the goal of the XGBoost model is to optimize the predicted values ​​output by the model through iterative optimization. As close as possible to the true value The following describes the training process of the XGBoost model for color difference prediction. Figure 10 As shown, where: In step 1001: Obtain the online data corresponding to multiple furnace cycles.

[0112] The specific implementation method of this step is the same as that of step 901 above, and will not be repeated here.

[0113] In step 1002: Feature extraction processing is performed on the online data corresponding to each furnace batch to obtain the first feature value corresponding to each furnace batch.

[0114] The specific implementation method of this step is the same as that of step 902 above, and will not be repeated here.

[0115] In step 1003: the first feature value corresponding to each furnace is subjected to parameter gridding processing to obtain multiple target feature value groups corresponding to each furnace.

[0116] The specific implementation method of this step is the same as that of step 903 above, and will not be repeated here.

[0117] In step 1004: training samples are constructed based on multiple target feature value groups corresponding to each furnace, and a training sample set is obtained based on the training samples corresponding to each furnace.

[0118] In the embodiments of this application, each training sample includes multiple target feature value groups corresponding to the furnace batch.

[0119] In step 1005: the initial XGBoost model is trained using the training sample set, and the converged XGBoost model is used as the color difference prediction model.

[0120] In this embodiment, the converged XGBoost model is used as the color difference prediction model, thereby realizing the automated recommendation of parameters in the PVD coating process.

[0121] Based on the same inventive concept, this application also provides a PVD coating analysis and control device, the device comprising: a memory and a processor, the memory being used to store instructions, the instructions stored in the memory being executed by the processor to achieve the above. Figures 2-10 Any one of the methods described.

[0122] Based on the same inventive concept, this application also provides a system including a PVD equipment and the above-mentioned PVD coating analysis and control device, wherein the PVD equipment receives adjusted parameters of the current batch or instructions to adjust the parameters of the current batch sent by the PVD coating control device.

[0123] In a specific implementation, the present invention also provides a computer storage medium, wherein the computer storage medium may store a program, which, when executed, may include some or all of the steps of the various embodiments of the PVD coating analysis and control method provided by the present invention. The storage medium may be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0124] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present invention.

[0125] The same or similar parts between the various embodiments in this specification can be referred to mutually. In particular, the device embodiments and terminal embodiments are basically similar to the method embodiments, so the description is relatively simple, and the relevant parts can be referred to the description in the method embodiments.

Claims

1. A method for analyzing and controlling PVD coatings, characterized in that, The method includes: Obtain the color difference value of the first preset number of furnaces preceding the current furnace batch; Based on the number of furnaces whose color difference values ​​fall within the color difference specification range from the first preset number of furnaces, the current reference point corresponding to the current furnace is obtained. If the current reference point is greater than the preset reference value, then the color layer data and interference layer data corresponding to the current furnace are obtained; The recovery time is determined based on the color layer data and the interference layer data; Adjust the parameters of the current furnace batch according to the recovery time.

2. The method according to claim 1, characterized in that, The step of obtaining the current reference point corresponding to the current furnace based on the number of furnaces whose color difference values ​​fall within the color difference specification range from the first preset number of furnaces includes: If the number of furnaces within the color difference specification range of the first preset number of furnaces is greater than or equal to the preset number of furnaces, then the recovery time corresponding to the furnace within the color difference specification range is obtained. The average value of the color difference at the recovery time corresponding to the batches within the specified color difference specification range is used as the current reference point; or... If the number of furnaces within the color difference specification range among the first preset number of furnaces is less than the preset number of furnaces, then the preset reference value is used as the current reference point.

3. The method according to claim 1, characterized in that, The acquisition of the color layer data and interference layer data corresponding to the current furnace batch includes: Collect online data for the current furnace cycle; The color layer data and the interference layer data are filtered from the online data.

4. The method according to claim 1, characterized in that, The determination of the rise time based on the color layer data and the interference layer data includes: The moving average voltage corresponding to each calculation moment within the first duration is determined based on the color layer data and the interference layer data; the first duration is the duration between the start time of the current furnace and the current time. The minimum voltage time corresponding to each calculation time is determined based on the moving average voltage at each calculation time. The recovery time is determined based on the minimum voltage time corresponding to each calculation time.

5. The method according to claim 4, characterized in that, The step of determining the moving average voltage corresponding to each calculation time point within the first time period based on the color layer data and the interference layer data includes: The first process is executed for each calculation time to obtain the moving average voltage corresponding to each calculation time. The first process includes: The calculation duration is obtained based on the calculation time and the preset time window length; Obtain the voltage value at each calculation moment within the calculation duration; The average voltage value at each calculation moment within the calculation period is taken as the moving average voltage corresponding to that calculation moment.

6. The method according to claim 4, characterized in that, The step of determining the minimum voltage time corresponding to each calculation time based on the moving average voltage at each calculation time includes: For each calculation time, the second process is executed to obtain the minimum voltage time corresponding to each calculation time; The second process includes: Obtain the moving average voltage corresponding to each calculation time between the start time of the current furnace and the calculation time; The moving average voltages are sorted to obtain a moving average voltage sequence; A second preset number of target moving average voltages are selected sequentially from the moving average voltage sequence; Calculate the average value of the second preset number of target moving average voltages at the calculation time; The mean value is taken as the minimum voltage time corresponding to the calculation time.

7. The method according to claim 4, characterized in that, The step of determining the recovery time based on the minimum voltage time corresponding to each calculation time includes: If the minimum voltage time corresponding to each calculation time determined within the second time period is the same, then the minimum voltage time is taken as the recovery time.

8. The method according to claim 1, characterized in that, The step of adjusting the parameters of the current furnace batch according to the recovery time includes: The difference between the current benchmark point and the recovery time is taken as the target difference. The compensation ratio is obtained based on the target difference and the preset piecewise function; The control duration is obtained based on the difference between the compensation ratio and the target value; The control duration for the current furnace cycle is adjusted based on the control duration.

9. The method according to claim 1, characterized in that, The method further includes: Collect online data for the current furnace cycle; The online data is subjected to feature extraction processing to obtain the first feature value corresponding to the current furnace batch; The first feature value is subjected to parametric gridding to obtain multiple target feature value groups; The multiple target feature value groups are respectively input into the pre-trained color difference prediction model to obtain the color difference prediction value corresponding to each target feature value group output by the color difference prediction model. For each feature value group, the following steps are performed: A comprehensive color difference is obtained based on the color difference value of the furnace and the predicted color difference; The feature value group corresponding to the color difference prediction value with the smallest overall color difference is taken as the target feature value group; Adjust the parameters of the current furnace batch according to the target feature value set.

10. A PVD coating analysis and control device, characterized in that, The apparatus includes a memory and a processor, the memory storing instructions which are executed by the processor to implement the method as claimed in any one of claims 1-9.

11. A system, characterized in that, The device includes a PVD equipment and the PVD coating analysis and control device as described in claim 10, wherein the PVD equipment receives adjusted parameters of the current batch or instructions to adjust the parameters of the current batch sent by the PVD coating control device.

12. An electronic device, characterized in that, It includes a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, the electronic device is triggered to perform the method of any one of claims 1-9.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform the method according to any one of claims 1-9.

Images