A method for processing a thin film capacitor and a thin film capacitor

Abstract

The application provides a processing method of a thin film capacitor and the thin film capacitor, and belongs to the technical field of thin film capacitor processing. The processing method comprises the following steps: step 1, training a processing scheme formulation model based on historical processing data of the thin film capacitor; step 2, generating a processing scheme according to required processing characteristics of the thin film capacitor and the processing scheme formulation model; and step 3, processing test based on the processing scheme. According to the processing method of the thin film capacitor and the thin film capacitor, the processing scheme formulation model is trained according to historical processing data of the thin film capacitor, which is labeled with historical material selection, historical thin film processing mode, historical capacitor structure and actual capacitor parameters of the thin film capacitor processed in history; the processing scheme is determined by using the processing scheme formulation model and required capacitor parameters of the thin film capacitor, and processing and test are carried out based on the processing scheme; and if the test is passed, subsequent batch application is carried out, thereby improving the suitability of the processing scheme formulation of the thin film capacitor.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of thin film capacitor processing, in particular to a thin film capacitor processing method and thin film capacitor. BACKGROUND

[0002] Thin film capacitors are widely used in power electronics, new energy, automotive electronics, industrial control, consumer electronics and other fields due to their advantages such as good high-frequency characteristics, high voltage resistance, long service life and strong self-healing property. For example, in a DC / DC circuit system, thin film capacitors are needed as EMC filters to reduce electromagnetic interference. For example, in an electric drive control circuit, thin film capacitors are needed for direct current support. Therefore, there is a great demand for thin film capacitors.

[0003] However, different thin film capacitors have different processing schemes. For example, thin film capacitors with low voltage resistance (<100V) require thin film (1-3μm) and single-sided metallization, while thin film capacitors with high voltage resistance (>1000V) require thick film (10-20μm), double-sided edge design and epoxy encapsulation. Inappropriate processing schemes can directly result in thin film capacitors that do not meet the processing requirements.

[0004] Therefore, there is an urgent need for a thin film capacitor processing method and thin film capacitor to at least solve the above problems. SUMMARY

[0005] One of the objects of the present application is to provide a thin film capacitor processing method and thin film capacitor. A processing scheme development model is trained according to historical processing data of thin film capacitors, which includes historical material selection, historical thin film processing method, historical capacitor structure and historical processing of actual capacitor parameters of thin film capacitors. The processing scheme development model and the required capacitor parameters of the thin film capacitor are used to determine the processing scheme, and the processing is carried out based on the processing scheme and tested. If the test is passed, subsequent batch application is carried out, and the suitability of the thin film capacitor processing scheme development is improved.

[0006] The thin film capacitor processing method provided by the embodiment of the present application comprises:

[0007] Step 1: training a processing scheme development model based on historical processing data of thin film capacitors;

[0008] Step 2: generating a processing scheme according to the required processing characteristics and the processing scheme development model of the thin film capacitor, wherein the required processing characteristics are the required capacitor parameters of the thin film capacitor;

[0009] Step 3: processing and testing based on the processing scheme.

[0010] Preferably, the historical processing data indicates the historical material selection, historical film processing method, historical capacitor structure, and actual capacitor parameters of the historically processed film capacitors.

[0011] Preferably, step 3: conduct processing tests based on the processing plan, including:

[0012] Standardized inspections are conducted during the processing and testing process.

[0013] The method for processing a thin-film capacitor provided in this embodiment of the invention further includes:

[0014] Step 4: During the processing and testing process, the stability of the thin film is controlled based on machine vision technology.

[0015] Preferably, thin film formation stability control is performed based on machine vision technology, including:

[0016] The first film-forming segment extruded by the extruder is captured by a pre-set industrial camera.

[0017] Extract the first defect feature of the first film-forming segment;

[0018] Based on the first defect characteristic, construct a target film-forming segment that does not meet the film-forming stability standard;

[0019] Defect features are matched between the simulated film-forming segment in the target film-forming segment and the future film-forming segment of the first film-forming segment. When the simulated film-forming segment is in the front The simulated defect features are matched with the defect features detected in real time during the future film formation stage, based on the simulated film formation stage. Based on simulated defect characteristics, an extruder adjustment strategy is determined; among them... This represents the total number of simulated defect features in the simulated film-forming segment;

[0020] Based on the extruder adjustment strategy, the extruder is controlled to adjust the extrusion strategy.

[0021] Preferably, based on the first defect characteristic, a target film-forming segment that does not meet the film-forming stability standard is constructed, including:

[0022] Locate the grid position corresponding to the first detected first defect feature;

[0023] The first film-forming segment following the grid column where the grid position is located is taken as the second film-forming segment;

[0024] A simulated film-forming segment is constructed following the second film-forming segment. During construction, the target film-forming segment, which is composed of the second film-forming segment and the simulated film-forming segment, is constrained to not meet the film-forming stability standard.

[0025] Preferably, according to the simulated film-forming segment, the first Based on simulated defect characteristics, extruder adjustment strategies are determined, including:

[0026] According to the Based on simulated defect characteristics, the first sub-strategy is determined;

[0027] Determine the first sub-policy pair The global impact of each simulated defect feature, based on the global impact... The simulated defect features determine the second sub-strategy;

[0028] The extruder adjustment strategy is to execute the first sub-strategy first, followed by the second sub-strategy.

[0029] Preferably, the first sub-strategy pair is determined. The global impact of each simulated defect feature includes:

[0030] Based on the extruder control experiment records, the extruder control model was trained, and the extruder control model and the extruder simulation model communicated and connected.

[0031] Configure the extruder simulation model based on the current extruder parameters, and input the first sub-strategy into the extruder control model;

[0032] Based on the third film-forming stage simulated by the extruder simulation model after the first sub-strategy is applied, the second defect feature is extracted.

[0033] contrast I simulated defect features and second defect features to determine the global impact.

[0034] Preferred, comparison I simulated defect features and second defect features to determine the global impact, including:

[0035] According to the simulated detection order of the second defect feature and Each simulated defect feature is compared one by one to obtain the comparison results;

[0036] If before If the similarity of all comparison results is less than or equal to the preset similarity threshold, then the determination of global influence is terminated, and the film formation stability control is re-based on machine vision technology.

[0037] Otherwise, when all After the simulated defect features are compared, the comparison results are integrated to obtain the global impact.

[0038] The present invention provides a thin-film capacitor, which is processed using the above-described thin-film capacitor processing method.

[0039] An embodiment of the present invention provides a processing system for a thin-film capacitor, comprising:

[0040] a model training module configured to train a processing scheme formulation model based on historical processing data of the film capacitor, the historical processing data being annotated with historical material selection, historical film processing mode, historical capacitor parameter, and historical processing of the film capacitor;

[0041] a processing scheme generation module configured to generate a processing scheme according to the required capacitor parameter of the film capacitor and the processing scheme formulation model, the required capacitor parameter being a required capacitor parameter of the film capacitor;

[0042] a processing test module configured to perform a processing test based on the processing scheme.

[0043] The present application has the following beneficial effects:

[0044] The present application trains a processing scheme formulation model based on historical processing data of the film capacitor, the historical processing data being annotated with historical material selection, historical film processing mode, historical capacitor parameter, and historical processing of the film capacitor; utilizes the processing scheme formulation model and the required capacitor parameter of the film capacitor to determine a processing scheme, and performs processing and testing based on the processing scheme, and performs subsequent batch application after the testing is passed, thereby improving the suitability of the processing scheme formulation of the film capacitor.

[0045] Other features and advantages of the present application will be set forth in the following description, and in part will be apparent from the description, or can be learned by practice of the present application. The objects and other advantages of the present application will be realized and attained by the structure particularly pointed out in the written description and claims thereof.

[0046] The technical solutions of the present application will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0047] The accompanying drawings are included to provide a further understanding of the present application, and constitute a part of the specification, illustrate the present application, and explain the technical solutions of the present application, and do not constitute a limitation on the present application. In the drawings:

[0048] Figure 1 FIG. 1 is a schematic diagram of a processing method of a film capacitor according to an embodiment of the present application;

[0049] Figure 2 FIG. 2 is a schematic diagram of a processing system of a film capacitor according to an embodiment of the present application. DETAILED DESCRIPTION

[0050] The preferred embodiments of the present application will be described below with reference to the accompanying drawings, and it should be understood that the preferred embodiments described herein are only used to explain and illustrate the present application, and do not constitute a limitation on the present application.

[0051] The present application provides a processing method of a film capacitor, as shown in FIG. 1, which comprises the following steps: Figure 1

[0052] ​Step 1: training a processing scheme formulation model based on historical processing data of the film capacitor;

[0053] The historical processing data is marked with historical material selection, historical film processing mode, historical capacitor structure, and actual capacitor parameters of the historical processed film capacitor.

[0054] In the training of the processing scheme formulation model, the historical material selection, historical film processing mode, and historical capacitor structure are used as the input of the neural network model, and the actual capacitor parameters of the historical processed film capacitor are used as the output of the neural network model for training.

[0055] Step 2: generating a processing scheme according to the required processing characteristics of the film capacitor and the processing scheme formulation model;

[0056] The required processing characteristics are the required capacitor parameters of the film capacitor.

[0057] Step 3: processing test based on the processing scheme;

[0058] During the processing test, a specification inspection is performed to check whether the processing test process conforms to the processing specification corresponding to the processing scheme.

[0059] The working principle and beneficial effects of the above technical solution are as follows:

[0060] The processing scheme formulation model is trained according to the historical processing data marked with historical material selection, historical film processing mode, historical capacitor structure, and actual capacitor parameters of the historical processed film capacitor.

[0061] The embodiment of the application provides a film capacitor processing method, which further comprises:

[0062] Step 4: during the processing test, film forming stability control is performed based on machine vision technology;

[0063] The film forming stability control based on machine vision technology comprises:

[0064] The first film forming section extruded by the extruder is photographed based on a preset industrial camera;

[0065] The preset industrial camera is an industrial CCD camera arranged at an extrusion port of the extruder, and the first film forming section is a film section that has been formed in a current film manufacturing task and is photographed by the industrial camera.

[0066] The first defect feature of the first film forming section is extracted.

[0067] The first film forming section is divided into grids according to a square grid with a side length of 1 cm as a grid division standard, the grid lines are parallel or perpendicular to the film edge lines, and the image of the grid area is input into the film defect detection model to obtain a defect detection result, such as scratches, cracks, bubbles, fisheye, burrs, plating layer shedding, and impurities, and the corresponding relationship between the defect detection result and the grid is the defect feature. The film defect detection model is learned by a large number of artificially labeled film images;

[0068] According to the first defect feature, a target film forming section that does not meet the film forming stability standard is constructed.

[0069] According to the first defect feature, a target film forming section that does not meet the film forming stability standard is constructed.

[0070] The grid position corresponding to the first detected first defect feature is located.

[0071] The first film forming section after the grid position is taken as the second film forming section.

[0072] An analog film forming section that continues the film forming of the second film forming section is constructed, and when constructing, the target film forming section composed of the second film forming section and the analog film forming section does not meet the film forming stability standard.

[0073] The film forming stability standard is an artificially set standard, and is specifically set according to the film process requirements, such as the pinhole density on the plating layer surface being less than or equal to 10 / cm².

[0074] The grid column is perpendicular to the film forming extrusion direction.

[0075] The analog film forming section is a simulated future film forming section that continues the second film forming section, and is realized based on three-dimensional reconstruction technology.

[0076] The analog film forming section and the future film forming section of the first film forming section in the target film forming section are matched in defect features. When the first simulated defect feature in the analog film forming section and the real-time detected defect feature of the future film forming section are matched, the first simulated defect feature of the analog film forming section is used to determine the adjustment strategy of the extruder; wherein, is the total number of simulated defect features in the analog film forming section.

[0077] Specifically, when performing defect feature matching between the simulated film-forming segment in the target film-forming segment and the future film-forming segment of the first film-forming segment, each simulated defect feature in the simulated film-forming segment is matched with the corresponding defect feature of the future film-forming segment. "Corresponding" means that the simulated grid position corresponds to the real grid position in the future film-forming segment. During matching, it is determined whether the defect similarity between the simulated defect corresponding to the grid position and the defect of the future film-forming segment is greater than a preset similarity threshold (e.g., 0.9).

[0078] Among them, the former In the context of simulated defect features, "front" refers to the position of the corresponding mesh element that is relatively forward.

[0079] Among them, corresponding matching refers to the grid positions where the defects are located being matched and the defect similarity being greater than a preset similarity threshold;

[0080] Among them, according to the simulated film-forming segment, the first The extruder adjustment strategy is determined based on the simulated defect characteristics: First, the first sub-strategy is determined, which is based on the... The first simulated defect feature determined the overcoming of the first The extruder control actions and extruder control time for each simulated defect characteristic are specifically determined based on the [number]th [details]. The grid distance of the simulated defect feature to the starting grid of the simulation determines the extruder control time. The extruder control actions are determined by identifying the defect types corresponding to the simulated defect characteristics; then, a second sub-strategy is determined, which is to overcome the influence of the first sub-strategy. The extruder control actions and timing for each simulated defect characteristic, after being affected by the first sub-strategy. The simulated defect features are obtained from the extruder control experiment records. After determining the new simulated defect features, the extruder control time and extruder control actions are determined for each new simulated defect feature. The two sub-strategies are then integrated to obtain the second sub-strategy. The extruder adjustment strategy is to execute the first sub-strategy first and then the second sub-strategy.

[0081] Based on the extruder adjustment strategy, the extruder is controlled to adjust the extrusion strategy.

[0082] The working principle and beneficial effects of the above technical solution are as follows:

[0083] The stability control of thin film formation is a crucial step in the processing of thin film capacitors, as it directly affects the performance, reliability, and lifespan of the capacitors. Therefore, machine vision technology can be introduced to control the stability of thin film formation in real time.

[0084] Specifically, the first film-forming section extruded by the extruder is captured in real time by an industrial CCD camera. The first film-forming section is then divided into grids, and the grid images are input into the film defect detection model to obtain defect detection results. The defect detection results and grid positions are then correlated to obtain the first defect features.

[0085] Next, based on the first defect feature, a target film-forming segment that does not meet the film-forming stability standard is simulated. During the simulation, the grid position corresponding to the first detected first defect feature is located, and the first film-forming segment after the grid column where the grid position is located is taken as the second film-forming segment. Then, a simulated film-forming segment is constructed after the second film-forming segment. The target film-forming segment composed of the second film-forming segment and the simulated film-forming segment is constrained to not meet the film-forming stability standard. The target film-forming segment is the possible situation where the first film-forming segment does not meet the film-forming stability standard, such as: the longitudinal linear scratches on the film surface have a width greater than 20 μm and a length greater than 5 mm, or the density of pinholes distributed on the coating surface is greater than 10 / cm².

[0086] Finally, defect feature matching is performed between the simulated film-forming segment and the future film-forming segment. When the simulated film-forming segment... The matching between simulated defect features and real-time detected defect features in the future film formation stage indicates that the film formation stability control of this stage is about to fail the film formation stability control standard. First, based on the... The simulated defect features determine how to overcome the first... The extruder control actions and timing are controlled in real time based on simulated defect characteristics, ensuring accurate matching for situations that do not meet film stability standards, and enabling real-time control of film stability.

[0087] Since the first sub-strategy only performed the first... Targeted control of simulated defect characteristics has its limitations in addressing the previous issues. The problem is that some simulated defect features cannot be well suppressed. Therefore, after executing the first sub-strategy, it is necessary to determine how to overcome the impact of the first sub-strategy. The extruder control actions and timing that simulate defect characteristics are defined as the second sub-strategy. The extruder adjustment strategy is to execute the first sub-strategy first, followed by the second sub-strategy, and then adjust the extrusion strategy based on this strategy.

[0088] This invention accurately identifies unstable situations, greatly reducing control resources; it performs immediate policy response, determines the impact of immediate policy response, and integrates the application of the final policy, improving the accuracy of intervention control and effectively achieving stability control in the thin film formation process, making it more intelligent.

[0089] In one embodiment, the first sub-policy pair is determined. The global impact of each simulated defect feature includes:

[0090] According to the extruder control experiment record, the extruder control model is trained, the extruder control model and the extruder simulation model are communicated and connected;

[0091] Wherein, the extruder control experiment record is a process record of electric control of the extruder, including the equipment parameters of the extruder, the corresponding control instructions and the response parameters;

[0092] Wherein, the extruder control model is a mathematical model constructed by machine learning (such as neural network, reinforcement learning) of the extruder control experiment record, used for generating control instructions (such as adjusting temperature, rotating speed) according to real-time state;

[0093] According to the current extruder parameter, the extruder simulation model is configured, and the first sub-strategy is input into the extruder control model;

[0094] According to the third film forming section simulated by the extruder simulation model after the application of the first sub-strategy, the second defect feature is extracted;

[0095] Wherein, the extraction principle of the second defect feature is the same as that of the first defect feature;

[0096] Comparison The simulation defect feature and the second defect feature are compared to determine the global influence.

[0097] Wherein, when comparing The simulation defect feature and the second defect feature to determine the global influence, the second defect feature is matched with The simulation defect feature according to the time sequence of simulation detection to determine the global influence.

[0098] The working principle and beneficial effects of the above technical scheme are as follows:

[0099] The extruder control model is trained by introducing the extruder control experiment record, and the extruder control model and the extruder simulation model are communicated and connected, the first sub-strategy is responded by the extruder control model and visualized in the extruder simulation model. The third film forming section simulated by the extruder simulation model after the application of the first sub-strategy is obtained, and the second defect feature is extracted; comparison The simulation defect feature and the second defect feature are compared to determine the global influence, and the determination accuracy of the global influence is improved.

[0100] In one embodiment, the simulation defect feature and the second defect feature are compared to determine the global influence, including: According to the simulation detection sequence of the second defect feature and

[0101] The simulation defect feature is matched one by one to obtain the comparison result;

[0102] ​The comparison results include: the feature similarity of the defect features compared one by one, such as the similarity of the relative positions of the defects and the similarity of the types of defects.

[0103] If before If the similarity of all comparison results is less than or equal to the preset similarity threshold, then the determination of global influence is terminated, and the film formation stability control is re-based on machine vision technology.

[0104] The preset result similarity threshold is set manually in advance;

[0105] Otherwise, when all After the simulated defect features are compared, the comparison results are integrated to obtain the global impact.

[0106] The working principle and beneficial effects of the above technical solution are as follows:

[0107] After real-time control is implemented, two scenarios arise due to changes in the extruder control parameters: First, the change in extruder control parameters leads to a significant adjustment of defects caused by the parameters before the change; second, the change in extruder control parameters leads to a minor adjustment of defects caused by the parameters before the change. This is based on the global impact. The second sub-strategy is determined based on the simulated defect features because... The simulated defect features are known; therefore, the second case can be pre-retrieved when matching the second sub-strategy. The library of overcoming strategies associated with each simulated defect feature allows for more efficient acquisition of the second sub-strategy when the impact is minor. However, if the adjustment is significant, the aforementioned film formation stability control can be continued directly based on the newly determined simulated defect features, improving the timeliness and rationality of the extruder adjustment strategy determination.

[0108] This invention provides a thin-film capacitor, which is processed using the thin-film capacitor processing method described in the above embodiments.

[0109] This invention provides a processing system for thin-film capacitors, such as... Figure 2 As shown, it includes:

[0110] Model training module 1 is used to train a model for formulating processing schemes based on historical processing data of thin film capacitors;

[0111] The processing scheme generation module 2 is used to formulate a model and generate a processing scheme based on the required processing characteristics and processing scheme of the film capacitor. The required processing characteristics are the required capacitor parameters of the film capacitor.

[0112] The processing test module 3 is used to perform processing tests based on the processing scheme.

[0113] The processing system for the thin-film capacitor also includes:

[0114] A thin film formation stability control module is used to control the stability of thin film formation during processing and testing based on machine vision technology, including:

[0115] The first film-forming segment extruded by the extruder is captured by a pre-set industrial camera.

[0116] Extract the first defect feature of the first film-forming segment;

[0117] Based on the first defect characteristic, construct a target film-forming segment that does not meet the film-forming stability standard;

[0118] Defect features are matched between the simulated film-forming segment in the target film-forming segment and the future film-forming segment of the first film-forming segment. When the simulated film-forming segment is in the front The simulated defect features are matched with the defect features detected in real time during the future film formation stage, based on the simulated film formation stage. Based on simulated defect characteristics, an extruder adjustment strategy is determined; among them... This represents the total number of simulated defect features in the simulated film-forming segment;

[0119] Based on the extruder adjustment strategy, control the extruder to adjust the extrusion strategy;

[0120] Among them, based on the first defect characteristic, the target film-forming segment that does not meet the film-forming stability standard is constructed, including:

[0121] Locate the grid position corresponding to the first detected first defect feature;

[0122] The first film-forming segment following the grid column where the grid position is located is taken as the second film-forming segment;

[0123] A simulated film-forming segment is constructed following the second film-forming segment. During construction, the target film-forming segment composed of the second film-forming segment and the simulated film-forming segment is constrained to not meet the film-forming stability standard.

[0124] Among them, based on the first defect characteristic, based on the simulated film-forming segment's... Based on simulated defect characteristics, extruder adjustment strategies are determined, including:

[0125] According to the Based on simulated defect characteristics, the first sub-strategy is determined;

[0126] Determine the first sub-policy pair The global impact of each simulated defect feature, based on the global impact... The simulated defect features determine the second sub-strategy;

[0127] The extruder adjustment strategy is to execute the first sub-strategy first, followed by the second sub-strategy.

[0128] Obviously, many modifications and variations of the present application are possible in light of the above teachings. It is, therefore, to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A method for processing a thin-film capacitor, characterized in that, include: Step 1: Train a model for processing schemes based on historical processing data of thin-film capacitors; Step 2: Develop a model based on the required processing characteristics and processing scheme of the film capacitor, and generate a processing scheme. The required processing characteristics are the required capacitor parameters of the film capacitor. Step 3: Conduct processing tests based on the processing plan; Step 4: During the processing and testing process, the first film-forming segment extruded by the extruder is photographed using a preset industrial camera; Extract the first defect feature of the first film-forming segment; Locate the grid position corresponding to the first detected first defect feature; The first film-forming segment following the grid column where the grid position is located is taken as the second film-forming segment; A simulated film-forming segment is constructed following the second film-forming segment. During construction, the target film-forming segment, which is composed of the second film-forming segment and the simulated film-forming segment, is constrained to not meet the film-forming stability standard. The simulated film-forming segment is a simulated film-forming segment that will continue to form from the second film-forming segment in the future. Defect features are matched between the simulated film-forming segment in the target film-forming segment and the future film-forming segment of the first film-forming segment. When the simulated film-forming segment is in the front The simulated defect features are matched with the defect features detected in real time during the future film formation stage, based on the simulated film formation stage. Based on simulated defect characteristics, an extruder adjustment strategy is determined; among them... This represents the total number of simulated defect features in the simulated film-forming segment; Based on the extruder adjustment strategy, the extruder is controlled to adjust the extrusion strategy.

2. The method for processing a thin-film capacitor as described in claim 1, characterized in that, Historical processing data indicates the historical material selection, historical film processing methods, historical capacitor structure, and actual capacitor parameters of historically processed film capacitors.

3. The method for processing a thin-film capacitor as described in claim 1, characterized in that, Step 3: Conduct processing tests based on the processing plan, including: Standardized inspections are conducted during the processing and testing process.

4. The method for processing a thin-film capacitor as described in claim 1, characterized in that, According to the simulated film-forming segment Based on simulated defect characteristics, extruder adjustment strategies are determined, including: According to the Based on simulated defect characteristics, the first sub-strategy is determined; Determine the first sub-policy pair The global impact of each simulated defect feature, based on the global impact... The simulated defect features determine the second sub-strategy; The extruder adjustment strategy is to execute the first sub-strategy first, followed by the second sub-strategy.

5. The method for processing a thin-film capacitor as described in claim 4, characterized in that, Determine the first sub-policy pair The global impact of each simulated defect feature includes: Based on the extruder control experiment records, the extruder control model was trained, and the extruder control model and the extruder simulation model communicated and connected. Configure the extruder simulation model based on the current extruder parameters, and input the first sub-strategy into the extruder control model; Based on the third film-forming stage simulated by the extruder simulation model after the first sub-strategy is applied, the second defect feature is extracted. contrast I simulated defect features and second defect features to determine the global impact.

6. The method for processing a thin-film capacitor as described in claim 5, characterized in that, contrast I simulated defect features and second defect features to determine the global impact, including: According to the simulated detection order of the second defect feature and Each simulated defect feature is compared one by one to obtain the comparison results; If before If the similarity of all comparison results is less than or equal to the preset similarity threshold, then the determination of global influence is terminated, and the film formation stability control is re-based on machine vision technology. Otherwise, when all After the simulated defect features are compared, the comparison results are integrated to obtain the global impact.

7. A thin-film capacitor, characterized in that, The film capacitor is processed using the processing method described in any one of claims 1-6.

Images