A detection device and method for accurately controlling current output

Abstract

The present invention discloses a detection device and method for accurately regulating current output, which relates to the technical field of current output regulation, and includes the following steps: obtaining the data of the pattern to be engraved from the control system of the lithography equipment; preprocessing the obtained pattern data to improve the pattern quality; parsing the pattern file, extracting the geometric information and hierarchical structure of the pattern, analyzing the geometric information and hierarchical structure, and then evaluating the complexity of the obtained pattern based on the analyzed geometric information and hierarchical structure. The present invention uses machine learning to generate pattern complexity coefficients through pattern data preprocessing and complexity evaluation, and adopts different current regulation strategies after classification. General patterns use standard frequencies to ensure energy efficiency and equipment life; highly complex patterns dynamically adjust the frequency through an intelligent system to meet high-precision requirements. This method improves lithography efficiency and accuracy, reduces energy consumption and equipment wear, and improves production reliability and economy.

Description

Technical Field

[0001] The present invention relates to the technical field of current output regulation, and in particular to a detection device and method for accurately regulating current output. Background Art

[0002] The detection equipment for precise control of current output refers to a device that can accurately control and adjust the current output, and has detection and monitoring functions. The main function of this device is to output a specific current value according to the needs in various application scenarios, and can detect in real time whether the output current meets the expectations, so as to ensure the stability and safety of the system. It is usually used in fields that require high-precision current control, such as laboratory research, industrial automation, electronic equipment testing, and medical instruments.

[0003] Such devices usually contain sophisticated electronic components and complex control systems that can finely adjust the current and monitor the actual output during the adjustment process. Through sensors and feedback loops, the device can detect small changes in the output current and make immediate corrections. This feedback control mechanism ensures the stability and accuracy of the current and avoids potential risks caused by current fluctuations. For example, in the semiconductor manufacturing process, small current fluctuations may cause product defects, and in medical equipment, unstable current may cause harm to patients. Therefore, detection equipment that accurately controls current output plays a vital role in these fields, ensuring the reliability and safety of system operation.

[0004] In the semiconductor manufacturing process, the current needs to be regulated at high frequency to ensure the quality of the product and the accuracy of the manufacturing process. Among them, photolithography is a key step in semiconductor manufacturing, which is used to transfer circuit patterns to silicon wafers. During the photolithography process, the exposure equipment needs to use a high-precision light source, usually a laser. In order to produce a stable laser beam, the current needs to be regulated at a high frequency to maintain the output stability and consistency of the laser. If the current fluctuates, it will cause the beam to be unstable, which in turn affects the clarity and accuracy of the photolithography pattern, and may ultimately cause the circuit performance to fail to meet the standards.

[0005] The prior art has the following deficiencies:

[0006] When regulating current during the lithography process, the prior art usually uses high-frequency current regulation frequencies to meet the needs of lithography, and these high-frequency current regulation frequencies are usually fixed. Although they can meet many conventional needs, when the lithography pattern is relatively simple, the continuous use of high-frequency current regulation will lead to excessive energy consumption of the equipment, waste a lot of energy, and increase unnecessary production costs. Secondly, the long-term use of fixed high-frequency current regulation will aggravate the wear of the lithography equipment and reduce the life and reliability of the equipment.

[0007] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0008] The purpose of the present invention is to provide a detection device and method for accurately controlling current output. The pattern data is preprocessed and the complexity is evaluated, and a pattern complexity coefficient is generated using a machine learning model. The patterns are classified according to the complexity coefficient, and corresponding current control strategies are adopted for patterns of different complexities. General patterns adopt a standard current control frequency to ensure energy efficiency and equipment life, while high-complexity patterns dynamically adjust the current control frequency through an intelligent system to meet the requirements of high-precision lithography. This method not only improves the efficiency and accuracy of the lithography process, reduces energy consumption and equipment wear, but also improves the reliability and economy of the overall production to solve the problems in the above-mentioned background technology.

[0009] In order to achieve the above object, the present invention provides the following technical solution: a detection method for accurately controlling current output, comprising the following steps:

[0010] Acquire the pattern data to be currently engraved from the control system of the lithography equipment;

[0011] Preprocess the acquired pattern data to improve the pattern quality;

[0012] Parsing the pattern file, extracting geometric information and hierarchical structure of the pattern, analyzing the geometric information and hierarchical structure, and then evaluating the complexity of the acquired pattern based on the analyzed geometric information and hierarchical structure;

[0013] According to the results of complexity analysis, the patterns are classified into highly complex patterns and general patterns according to their complexity;

[0014] For lithography processes classified as general patterns, control is performed based on the standard current control frequency. For lithography processes classified as highly complex patterns, the actual current control frequency is adjusted through an intelligent system based on the standard current control frequency.

[0015] Preferably, the pattern file is parsed and the geometric information of the pattern is extracted, and the specific steps are as follows:

[0016] Loading pattern files from storage media;

[0017] Parse the format of the loaded pattern file and use the corresponding parser to read the file header, metadata and pattern data;

[0018] After the file format is parsed, each layer and object in the file is traversed, the geometric information is converted into a standard data structure, and the geometric information of the pattern is extracted;

[0019] After extracting the geometric information, the extracted information is verified to detect and correct potential errors and inconsistencies.

[0020] Preferably, the pattern file is parsed to extract the hierarchical structure of the pattern, and the specific steps are as follows:

[0021] Parse file header information to extract relevant information by parsing the first few lines or header area of ​​the file to ensure a comprehensive understanding of the file structure;

[0022] According to the number and structure information of layers obtained from the file header information, the data of each layer is parsed layer by layer, and the specific content is parsed by traversing each layer data block in the file and converting it into a data structure suitable for processing;

[0023] In the process of parsing the layer data, the geometric information in each layer is extracted, and the complete pattern representation of each layer is constructed by extracting the geometric information;

[0024] The parsed layers and their geometric information are organized according to the hierarchical structure of the design file to form a complete hierarchical model.

[0025] Preferably, the geometric information and hierarchical structure of the pattern are extracted, and after analyzing the geometric information and the hierarchical structure, a geometric complexity index and a hierarchical complexity index are generated respectively. Based on the analyzed geometric complexity index and hierarchical complexity index, a pattern complexity coefficient is generated through a machine learning model, and the complexity of the acquired pattern is evaluated through the pattern complexity coefficient.

[0026] Preferably, the pattern complexity coefficient generated after analyzing the pattern to be engraved is compared and analyzed with a preset pattern complexity coefficient reference threshold, and the pattern is divided according to complexity, and the division process is as follows:

[0027] If the pattern complexity coefficient is greater than or equal to the pattern complexity coefficient reference threshold, the pattern is classified as a high-complexity pattern according to its complexity;

[0028] If the pattern complexity coefficient is less than the pattern complexity coefficient reference threshold, the pattern is classified as a general pattern according to the complexity.

[0029] Preferably, for a photolithography process classified as a highly complex pattern, the specific steps of adjusting the actual current control frequency by the intelligent system based on the standard current control frequency are as follows:

[0030] Get the preset standard current control frequency f from the system sta and the complexity coefficient of the current pattern Pattern ξ At the same time, obtain the pattern complexity coefficient reference threshold Cref , providing a basis for subsequent regulatory processes;

[0031] According to the pattern complexity coefficient Pattern ξ and the pattern complexity coefficient reference threshold C ref The difference is used to calculate the adjustment coefficient δ, which is used to determine the correction amount of the actual current control frequency. The calculation expression is:

[0032] According to the adjustment coefficient δ, the initial adjustment frequency is calculated. The calculation expression is: f adj =f sta ·δ, where f adj Indicates the initial adjustment frequency;

[0033] Combine the standard current control frequency with the initial adjustment frequency to calculate the actual current control frequency. The calculation expression is: In the formula, f actual Indicates the actual current regulation frequency.

[0034] Preferably, after analyzing the geometric information, the specific steps of generating the geometric complexity index are as follows:

[0035] Extract geometric information from pattern data and establish a geometric element set based on the basic information of geometric elements, where each element g i The geometric information includes vertex coordinates (x i，j ,y i，j ), side length L i and area A i , where (x ij ,y ij ) represents the coordinates of vertex j, n i Represents the geometric element g i The number of vertices, L i is the geometric element g i The area and side length of i is the geometric element g i area;

[0036] For each geometric element g i Calculate its complexity index C i The complexity index is calculated based on the number of vertices, edge lengths, and areas of geometric elements. The calculation expression is:

[0037] Based on the complexity index C i Calculate the geometric complexity index, the calculation expression is: In the formula, C complexityrepresents the geometric complexity index, and N represents the total number of geometric elements.

[0038] Preferably, after analyzing the hierarchical structure, the specific steps of generating the hierarchical complexity index are as follows:

[0039] For each layer, calculate the layer complexity, the calculation expression is: In the formula, C L Indicates the complexity of the layer, represents the complexity weight of the kth geometric element, Indicates the resolution requirement of the kth geometric element, N E Indicates the number of geometric elements in each layer;

[0040] Calculate the interaction complexity between layers. The interaction complexity between layers is expressed as the sum of the interaction effects between all adjacent layers. The calculation expression is: In the formula, I L Indicates the complexity of interaction between layers, I p，q represents the interaction between the pth layer and the qth layer, f(I p，q ) is a function that represents the interaction effect, N L Indicates the number of layers, that is, the total number of layers in the pattern file;

[0041] The complexity of all layers and the complexity of interactions between layers are combined to calculate the hierarchy complexity index. The calculation expression is: In the formula, I complexity represents the hierarchy complexity index, Indicates the layer complexity of the p-th layer.

[0042] A detection device for accurately controlling current output, comprising a pattern data acquisition module, a pattern data preprocessing module, a pattern analysis and complexity evaluation module, a pattern complexity classification module and a current control module;

[0043] A pattern data acquisition module, which acquires the current pattern data to be engraved from the control system of the lithography equipment;

[0044] A pattern data preprocessing module preprocesses the acquired pattern data to improve the pattern quality;

[0045] A pattern analysis and complexity evaluation module analyzes the pattern file, extracts the geometric information and hierarchical structure of the pattern, analyzes the geometric information and hierarchical structure, and then evaluates the complexity of the acquired pattern based on the analyzed geometric information and hierarchical structure;

[0046] The pattern complexity classification module classifies the patterns according to the results of complexity analysis, and divides the patterns into highly complex patterns and general patterns according to their complexity;

[0047] The current control module controls the photolithography process classified as a general pattern based on the standard current control frequency. For the photolithography process classified as a highly complex pattern, the actual current control frequency is adjusted through an intelligent system based on the standard current control frequency.

[0048] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0049] The present invention preprocesses the pattern data and evaluates the complexity, uses a machine learning model to generate a pattern complexity coefficient, classifies the patterns according to the complexity coefficient, and adopts corresponding current control strategies for patterns of different complexities. General patterns use a standard current control frequency to ensure energy efficiency and equipment life, while high-complexity patterns dynamically adjust the current control frequency through an intelligent system to meet the needs of high-precision lithography. This method not only improves the efficiency and accuracy of the lithography process, reduces energy consumption and equipment wear, but also improves the reliability and economy of the overall production. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0051] Figure 1 The present invention is a method flow chart of a detection method for accurately controlling current output.

[0052] Figure 2 This is a module schematic diagram of a detection device for accurately controlling current output according to the present invention. DETAILED DESCRIPTION

[0053] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0054] The present invention provides Figure 1 A detection method for accurately controlling current output is shown, comprising the following steps:

[0055] Acquire the pattern data to be currently engraved from the control system of the lithography equipment;

[0056] The design system contains all the circuit patterns that need to be photolithographic in the semiconductor manufacturing process. Through the interface with the design system, the current batch of patterns to be engraved are automatically obtained. This step ensures that the acquired pattern data is the latest and contains all the detailed information required for photolithography.

[0057] Preprocess the acquired pattern data to improve the pattern quality;

[0058] Preprocessing the acquired pattern data is a crucial step in the lithography process, which can improve the pattern quality and ensure the smooth progress of subsequent steps. Preprocessing includes the following aspects, each of which has its specific role:

[0059] 1. Data format conversion: Convert the pattern design file (such as GDSII file) into a format that can be recognized and processed by the lithography equipment. Different lithography equipment may use different data formats. In order to ensure compatibility and correct processing, the pattern data needs to be converted from the design software format to a format that the equipment can recognize. This step ensures the integrity and accuracy of the data and avoids pattern errors caused by format incompatibility.

[0060] 2. Data denoising: Remove noise and irrelevant information from pattern data to improve data quality and reliability. Pattern data may introduce noise during the design and transmission process, which will affect the accuracy of lithography. Through denoising, these irrelevant information can be filtered out and the core features of the pattern can be retained, thereby improving the clarity and accuracy of the pattern.

[0061] 3. Geometric correction: Perform geometric correction on the pattern to correct the geometric distortion introduced during the design and transmission process. During the design and transmission process, the pattern may have geometric distortion, such as stretching, compression or rotation. Geometric correction corrects these distortions through precise algorithms to ensure that the size and shape of the pattern meet the design requirements, thereby ensuring the accuracy and consistency of lithography.

[0062] 4. Resolution enhancement: Improve the resolution of the pattern, and enhance the details and clarity of the pattern. The photolithography process has very high requirements for the resolution of the pattern, especially for complex and fine patterns. Through resolution enhancement technology, the ability to express the details of the pattern can be improved, making the tiny features clearer, thereby improving the photolithography effect and the quality of the final product.

[0063] 5. Pattern optimization: Optimize the pattern to make it more suitable for the lithography process and reduce the defects that may occur during the lithography process. Pattern optimization includes operations such as adjusting the edges of the pattern, filling small holes, and smoothing curves. These optimization steps can reduce edge roughness, fractures, and other defects that may occur during the lithography process, ensuring the quality and reliability of the lithography results.

[0064] 6. Size calibration: The size of the pattern is calibrated to ensure that the pattern size meets the process requirements of the lithography equipment. Different lithography equipment and processes have strict requirements on the pattern size. Through size calibration, the size of the pattern can be adjusted to meet the process specifications of the equipment, ensuring the accuracy and consistency of the lithography process.

[0065] 7. Layer separation: Separate the multi-layer pattern data to facilitate layer-by-layer lithography operations. In the semiconductor manufacturing process, lithography usually involves multiple layers, each layer corresponding to a different process step. Layer separation breaks down these layer data to ensure that each layer can be properly processed and aligned during the lithography process, avoiding interference and misalignment between layers.

[0066] 8. Error detection and repair: Detect errors in pattern data and repair them to ensure the integrity and accuracy of the data. Pattern data may introduce errors during the design, transmission and conversion process, such as broken lines, overlaps and missing lines. Through error detection and repair, these errors can be found and corrected to ensure the integrity and accuracy of pattern data, thereby improving the quality of lithography.

[0067] 9. Alignment mark generation: Generate alignment marks to assist lithography equipment in precise alignment during multi-layer lithography. Alignment marks are important auxiliary tools in the lithography process and are used to accurately align multi-layer patterns. Generate alignment marks and add them to the pattern data to ensure that each layer of pattern can be accurately aligned during the lithography process, improving the accuracy and consistency of lithography.

[0068] 10. Data compression and storage: compress and store the pre-processed pattern data to facilitate subsequent processing and transmission. The lithography pattern data is usually very large, and compression technology can reduce the amount of data for easy storage and transmission. At the same time, the pre-processed data is stored to ensure that it can be quickly called and processed when needed, improving the efficiency and flexibility of the lithography process.

[0069] Through the above preprocessing steps, the quality and applicability of pattern data can be greatly improved, ensuring the accuracy and efficiency of the lithography process, and ultimately improving the overall quality and efficiency of semiconductor manufacturing.

[0070] Parsing the pattern file, extracting geometric information and hierarchical structure of the pattern, analyzing the geometric information and hierarchical structure, and then evaluating the complexity of the acquired pattern based on the analyzed geometric information and hierarchical structure;

[0071] Parse the pattern file and extract the geometric information of the pattern. The specific steps are as follows:

[0072] Loading pattern files from storage media;

[0073] Pattern files are usually in a format generated by design software, such as GDSII or OASIS. When loading a file, you need to ensure that the file is complete and not damaged. In order to improve processing efficiency, you can first load the file into memory, which can speed up subsequent parsing and processing.

[0074] Parse the format of the loaded pattern file and use the corresponding parser to read the file header, metadata and pattern data;

[0075] Pattern files of different formats have different internal structures and encoding methods. The parsing step requires the use of a corresponding parser to read the file header, metadata, and pattern data. The purpose of file format parsing is to convert the information in the pattern file into a usable data structure for subsequent processing. During the parsing process, attention should be paid to the file hierarchy and the definition of various objects.

[0076] After the file format is parsed, each layer and object in the file is traversed, the geometric information is converted into a standard data structure, and the geometric information of the pattern is extracted;

[0077] The geometric information includes the basic shape (such as rectangle, circle, line, etc.), size, position and relative relationship of the pattern. The process of extracting geometric information requires traversing each layer and object in the file and converting the geometric information into a standard data structure (such as points, lines, and surfaces in a two-dimensional coordinate system). This process ensures that the geometric information of the pattern is obtained completely and accurately.

[0078] After extracting the geometric information, verify the extracted information to find and correct potential errors and inconsistencies;

[0079] The verification step includes checking the integrity of geometric objects, no overlap, no breakpoints, etc. It is also necessary to verify whether the geometric information meets the design specifications and lithography process requirements. Through geometric information verification, potential errors and inconsistencies can be discovered and corrected to ensure the accuracy and reliability of pattern data.

[0080] Parse the pattern file and extract the hierarchical structure of the pattern. The specific steps are as follows:

[0081] Parse file header information to extract relevant information by parsing the first few lines or header area of ​​the file to ensure a comprehensive understanding of the file structure;

[0082] The header of a pattern file usually contains basic information about the file structure and content, including the size, unit, number of layers, etc. of the pattern. The purpose of parsing the file header information is to obtain these basic metadata to guide the subsequent data parsing process. This step extracts relevant information by parsing the first few lines of the file or a specific header area to ensure a comprehensive understanding of the file structure.

[0083] According to the number and structure information of layers obtained from the file header information, the data of each layer is parsed layer by layer, and the specific content is parsed by traversing each layer data block in the file and converting it into a data structure suitable for processing;

[0084] The data for each layer may include various geometric shapes, paths, and other design elements.

[0085] In the process of parsing the layer data, the geometric information in each layer is extracted, and the complete pattern representation of each layer is constructed by extracting the geometric information;

[0086] Geometric information includes the coordinates, dimensions, and other properties of lines, rectangles, polygons, and other elements. This step requires identifying and parsing the specific format of each geometric element and converting it into a standardized data format for subsequent processing and analysis.

[0087] Organize the parsed layers and their geometric information according to the hierarchical structure of the design file to form a complete hierarchical model;

[0088] This step involves arranging the geometric data of each layer according to the layer order and hierarchical relationship, ensuring the correct position and relationship of each layer in the final model. By building a hierarchical model, the hierarchy and mutual relationship of the entire pattern can be clearly represented, providing a basis for subsequent lithography processing and optimization.

[0089] The geometric information and hierarchical structure of the pattern are extracted. After analyzing the geometric information and hierarchical structure, a geometric complexity index and a hierarchical complexity index are generated respectively. Based on the analyzed geometric complexity index and hierarchical complexity index, a pattern complexity coefficient is generated through a machine learning model. The complexity of the acquired pattern is evaluated through the pattern complexity coefficient.

[0090] After analyzing the geometric information, the specific steps to generate the geometric complexity index are as follows:

[0091] Extract geometric information from pattern data and establish a geometric element set based on the basic information of geometric elements, where each element g i The geometric information includes vertex coordinates (x i，j ,y i，j ), side length L i and area A i , where (x ij ,y ij ) represents the coordinates of vertex j, n i Represents the geometric element g i The number of vertices, L i is the geometric element g i The area and side length ofi is the geometric element g i area;

[0092] Geometric information includes the specific coordinates and sizes of all lines, rectangles, polygons and other elements.

[0093] For each geometric element g i Calculate its complexity index C i The complexity index is calculated based on the number of vertices, edge lengths, and areas of geometric elements. The calculation expression is:

[0094] Based on the complexity index C i Calculate the geometric complexity index, the calculation expression is: In the formula, C complexity represents the geometric complexity index, and N represents the total number of geometric elements.

[0095] It can be seen from the geometric complexity index that the larger the performance value of the geometric complexity index, the more complex the pattern to be engraved. Specifically, the geometric complexity index comprehensively considers factors such as the number of vertices, side lengths, areas, and vertex coordinates of geometric elements, and reflects the overall complexity of the pattern through these parameters. Therefore, when the geometric complexity index is higher, it means that the pattern has more vertices, larger side lengths, more complex shapes, and larger areas. Conversely, a lower geometric complexity index indicates that the pattern is relatively simple.

[0096] After analyzing the hierarchy, the specific steps to generate the hierarchy complexity index are as follows:

[0097] For each layer, calculate the layer complexity, the calculation expression is: In the formula, C L Indicates the complexity of the layer, represents the complexity weight of the kth geometric element, Indicates the resolution requirement of the kth geometric element, N E Indicates the number of geometric elements in each layer, including lines, rectangles, polygons, etc.;

[0098] The complexity weight is a value assigned to each type of geometric element to reflect its relative complexity. Different geometric elements have different processing difficulties during the lithography process. Elements such as lines, rectangles, polygons, and complex curves have different complexities. The higher the weight value, the higher the complexity of the element, which requires more processing resources and time. The complexity weight can be obtained through empirical data and experimental results. For example, the relative complexity of different geometric elements can be determined by analyzing their processing time, resource consumption, and error rate during the lithography process. Some reference values ​​are usually provided in industry standards and professional literature. In addition, these weights can be gradually adjusted and optimized through data statistics and analysis in actual production to reflect the situation in the real production environment.

[0099] Resolution requirements are a measure of the fineness and clarity required of geometric elements during the lithography process. It represents the level of accuracy that each geometric element needs to achieve during the lithography process. The higher the resolution, the richer the details of the lithography pattern, and the higher the requirements for equipment and processes. High resolution requirements are usually accompanied by stricter control and more complex processing processes. Resolution requirements can be determined by design specifications and process standards. Design engineers set resolution standards based on the performance requirements of the final product when designing patterns. In addition, the testing and verification process in actual production can also provide feedback information to help adjust and determine the appropriate resolution requirements. The resolution requirements for each geometric element can be finally determined by comparing with customer needs and industry standards, combined with the capabilities of the equipment and process characteristics.

[0100] Calculate the interaction complexity between layers. The interaction complexity between layers is expressed as the sum of the interaction effects between all adjacent layers. The calculation expression is: In the formula, I L Indicates the complexity of interaction between layers, I p，q represents the interaction between the pth layer and the qth layer, f(I p，q ) is a function that represents the interaction effect, N L Indicates the number of layers, that is, the total number of layers in the pattern file;

[0101] The interaction effect I between the pth layer and the qth layer p，q Refers to the degree to which the two layers of patterns interact and influence each other during the lithography process. This interaction may be reflected in many aspects, such as the geometry of the two layers overlapping or close in space, resulting in changes in energy distribution during lithography, or mutual influence caused by electromagnetic interference. The greater the interaction, the more precision and control are required during the lithography process to avoid pattern distortion or defects.

[0102] The interaction function f(I p，q) is a mathematical function used to quantify and represent the interaction between two layers. This function converts various factors of interaction (such as overlapping area, electromagnetic interference, etc.) into a numerical value for easy use in complexity calculation. Function f(I p，q ) can be defined according to specific needs and actual conditions, for example, it can be a ratio of overlapping area, a function of interference intensity or a combination of other related parameters, which is not specifically limited here.

[0103] The complexity of all layers and the complexity of interactions between layers are combined to calculate the hierarchy complexity index. The calculation expression is: In the formula, I complexity represents the hierarchy complexity index, Indicates the layer complexity of the p-th layer.

[0104] It can be seen from the hierarchical complexity index that the larger the performance value of the generated hierarchical complexity index, the more complex the pattern to be engraved. This means that the pattern has more geometric elements, higher resolution requirements, more significant interactions between layers, and requires more sophisticated and complex lithography processing. On the contrary, if the hierarchical complexity index is small, it means that the pattern to be engraved is relatively simple, and the accuracy and complexity required for processing are lower.

[0105] The machine learning model is not specifically limited here, and can achieve the geometric complexity index C complexity and the hierarchy complexity index I complexity Perform comprehensive analysis to generate pattern complexity coefficient Pattern ξ In order to realize the technical solution of the present invention, the present invention provides a specific implementation method;

[0106] Pattern complexity coefficientPattern ξ The generated calculation formula is: Pattern ξ =v 1 ×C complexity +v 2 ×I complexity , where v 1 、v 2 They are the geometric complexity index C complexity and the hierarchy complexity index I complexity The preset proportionality factor, and v 1 、v 2 Both are greater than 0.

[0107] It can be seen from the pattern complexity coefficient that the greater the expression value of the geometric complexity index generated after analyzing the geometric information of the pattern to be engraved, the greater the expression value of the hierarchical complexity index generated after analyzing the hierarchical structure of the pattern to be engraved. That is, the greater the expression value of the pattern complexity coefficient generated after analyzing the pattern to be engraved, the more complex the pattern to be engraved, and vice versa.

[0108] According to the results of complexity analysis, the patterns are classified into highly complex patterns and general patterns according to their complexity;

[0109] The pattern complexity coefficient generated after analyzing the pattern to be engraved is compared and analyzed with the preset pattern complexity coefficient reference threshold, and the pattern is divided according to complexity. The division process is as follows:

[0110] If the pattern complexity coefficient is greater than or equal to the pattern complexity coefficient reference threshold, the pattern is classified as a high-complexity pattern according to its complexity;

[0111] If the pattern complexity coefficient is less than the pattern complexity coefficient reference threshold, the pattern is classified as a general pattern according to the complexity.

[0112] For photolithography processes classified as general patterns, the current control frequency is regulated based on the standard current control frequency. For photolithography processes classified as highly complex patterns, the actual current control frequency is adjusted through the intelligent system based on the standard current control frequency.

[0113] For lithography processes classified as general patterns, the system uses a pre-set standard current control frequency for control and regulation. General patterns are relatively simple and require less lithography accuracy and complexity, so there is no need for complex current control strategies. Using a standard current control frequency can ensure the stability and efficiency of the lithography process while avoiding unnecessary waste of resources.

[0114] Standard current control frequency refers to a pre-set, fixed current regulation frequency used to control the current output of lithography equipment when processing general patterns. This frequency is formulated based on the performance of the equipment and general lithography requirements, aiming to provide a stable and efficient operating mode to meet the lithography requirements of general patterns.

[0115] The standard current control frequency does not change with the complexity and details of the pattern, but remains constant. This fixed frequency helps to simplify the control system design of the lithography process, reduce the adjustment burden of the equipment, and ensure that general lithography tasks can be completed with consistent quality and efficiency. By using a standard frequency, it is possible to reduce complex real-time adjustment processes, reduce system complexity and failure rate.

[0116] The setting criteria for the standard current control frequency are based on the capabilities of the lithography equipment, the process requirements of general patterns, and historical data analysis. The setting criteria include the evaluation of equipment performance parameters, the demand analysis of typical lithography tasks, and statistical analysis of data collected during production to find a most suitable fixed frequency.

[0117] In the process of setting the standard, factors such as the maximum and minimum operating frequency range of the equipment, the complexity distribution of the most common lithography patterns, and the current regulation effect in actual production are comprehensively considered. Through experiments and simulations, an optimal frequency value is found that can meet the needs of most lithography tasks and ensure stable operation of the equipment. This standard frequency needs to be strictly verified and optimized to ensure that it can provide the best performance and the least resource consumption in actual applications.

[0118] For the lithography process classified as a highly complex pattern, the specific steps for adjusting the actual current control frequency through the intelligent system based on the standard current control frequency are as follows:

[0119] Get the preset standard current control frequency f from the system sta and the complexity coefficient of the current pattern Pattern ξ At the same time, obtain the pattern complexity coefficient reference threshold C ref , providing a basis for subsequent regulatory processes;

[0120] According to the pattern complexity coefficient Pattern ξ and the pattern complexity coefficient reference threshold C ref The difference is used to calculate the adjustment coefficient δ, which is used to determine the correction amount of the actual current control frequency. The calculation expression is:

[0121] This formula reflects the relative change of the pattern complexity coefficient relative to the pattern complexity coefficient reference threshold. The squaring operation increases the adjustment amplitude in the case of large differences, thereby ensuring that high-complexity patterns receive more significant frequency adjustments.

[0122] According to the adjustment coefficient δ, the initial adjustment frequency is calculated. The calculation expression is: f adj =f sta ·δ, where f adj Indicates the initial adjustment frequency;

[0123] This formula ensures that the adjustment amplitude is proportional to the relative change in pattern complexity, so that complex patterns receive the necessary frequency adjustment to accommodate their complexity requirements.

[0124] Combine the standard current control frequency with the initial adjustment frequency to calculate the actual current control frequency. The calculation expression is: In the formula, factual Indicates the actual current regulation frequency;

[0125] This formula introduces the influence of the adjustment coefficient into the actual current control frequency, ensuring that highly complex patterns receive a moderate frequency increase based on the standard current control frequency to meet actual needs.

[0126] The present invention preprocesses the pattern data and evaluates the complexity, uses a machine learning model to generate a pattern complexity coefficient, classifies the patterns according to the complexity coefficient, and adopts corresponding current control strategies for patterns of different complexities. General patterns use a standard current control frequency to ensure energy efficiency and equipment life, while high-complexity patterns dynamically adjust the current control frequency through an intelligent system to meet the needs of high-precision lithography. This method not only improves the efficiency and accuracy of the lithography process, reduces energy consumption and equipment wear, but also improves the reliability and economy of the overall production.

[0127] The present invention provides Figure 2 A detection device for accurately controlling current output is shown, comprising a pattern data acquisition module, a pattern data preprocessing module, a pattern analysis and complexity evaluation module, a pattern complexity classification module and a current control module;

[0128] A pattern data acquisition module, which acquires the current pattern data to be engraved from the control system of the lithography equipment;

[0129] A pattern data preprocessing module preprocesses the acquired pattern data to improve the pattern quality;

[0130] A pattern analysis and complexity evaluation module analyzes the pattern file, extracts the geometric information and hierarchical structure of the pattern, analyzes the geometric information and hierarchical structure, and then evaluates the complexity of the acquired pattern based on the analyzed geometric information and hierarchical structure;

[0131] The pattern complexity classification module classifies the patterns according to the results of complexity analysis, and divides the patterns into highly complex patterns and general patterns according to their complexity;

[0132] The current control module controls the photolithography process classified as a general pattern based on the standard current control frequency, and adjusts the actual current control frequency through the intelligent system based on the standard current control frequency for the photolithography process classified as a highly complex pattern;

[0133] A detection method for accurately controlling current output provided in an embodiment of the present invention is realized by the above-mentioned detection device for accurately controlling current output. The specific method and process of the detection device for accurately controlling current output are detailed in the embodiment of the above-mentioned detection method for accurately controlling current output, which will not be repeated here.

[0134] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0135] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0136] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A detection method for accurately controlling current output, characterized in that: The following steps are involved: Acquire the pattern data to be currently engraved from the control system of the lithography equipment; Preprocess the acquired pattern data to improve the pattern quality; Parsing the pattern file, extracting geometric information and hierarchical structure of the pattern, analyzing the geometric information and hierarchical structure, and then evaluating the complexity of the acquired pattern based on the analyzed geometric information and hierarchical structure; According to the results of complexity analysis, the patterns are classified into highly complex patterns and general patterns according to their complexity; For lithography processes classified as general patterns, control is performed based on the standard current control frequency. For lithography processes classified as highly complex patterns, the actual current control frequency is adjusted through an intelligent system based on the standard current control frequency.

2. A detection method for accurately controlling current output according to claim 1, characterized in that: Parse the pattern file and extract the geometric information of the pattern. The specific steps are as follows: Loading pattern files from storage media; Parse the format of the loaded pattern file and use the corresponding parser to read the file header, metadata and pattern data; After the file format is parsed, each layer and object in the file is traversed, the geometric information is converted into a standard data structure, and the geometric information of the pattern is extracted; After extracting the geometric information, the extracted information is verified to detect and correct potential errors and inconsistencies.

3. The detection method for accurately controlling current output according to claim 1, characterized in that: Parse the pattern file and extract the hierarchical structure of the pattern. The specific steps are as follows: Parse file header information to extract relevant information by parsing the first few lines or header area of ​​the file to ensure a comprehensive understanding of the file structure; According to the number and structure information of layers obtained from the file header information, the data of each layer is parsed layer by layer, and the specific content is parsed by traversing each layer data block in the file and converting it into a data structure suitable for processing; In the process of parsing the layer data, the geometric information in each layer is extracted, and the complete pattern representation of each layer is constructed by extracting the geometric information; The parsed layers and their geometric information are organized according to the hierarchical structure of the design file to form a complete hierarchical model.

4. The detection method for accurately controlling current output according to claim 1, characterized in that: The geometric information and hierarchical structure of the pattern are extracted. After analyzing the geometric information and hierarchical structure, a geometric complexity index and a hierarchical complexity index are generated respectively. Based on the analyzed geometric complexity index and hierarchical complexity index, a pattern complexity coefficient is generated through a machine learning model. The complexity of the acquired pattern is evaluated through the pattern complexity coefficient.

5. A detection method for accurately controlling current output according to claim 4, characterized in that: The pattern complexity coefficient generated after analyzing the pattern to be engraved is compared and analyzed with the preset pattern complexity coefficient reference threshold, and the pattern is divided according to complexity. The division process is as follows: If the pattern complexity coefficient is greater than or equal to the pattern complexity coefficient reference threshold, the pattern is classified as a high-complexity pattern according to its complexity; If the pattern complexity coefficient is less than the pattern complexity coefficient reference threshold, the pattern is classified as a general pattern according to the complexity.

6. A detection method for accurately controlling current output according to claim 5, characterized in that: For the lithography process classified as a highly complex pattern, the specific steps for adjusting the actual current control frequency through the intelligent system based on the standard current control frequency are as follows: Get the preset standard current control frequency f from the system sta and the complexity coefficient of the current pattern Pattern ξ At the same time, obtain the pattern complexity coefficient reference threshold C ref , providing a basis for subsequent regulatory processes; According to the pattern complexity coefficient Pattern ξ and the pattern complexity coefficient reference threshold C ref The difference is used to calculate the adjustment coefficient δ, which is used to determine the correction amount of the actual current control frequency. The calculation expression is: According to the adjustment coefficient δ, the initial adjustment frequency is calculated. The calculation expression is: f adj =f sta ·δ, where f adj Indicates the initial adjustment frequency; Combine the standard current control frequency with the initial adjustment frequency to calculate the actual current control frequency. The calculation expression is: In the formula, f actual Indicates the actual current regulation frequency.

7. A detection method for accurately controlling current output according to claim 4, characterized in that: After analyzing the geometric information, the specific steps to generate the geometric complexity index are as follows: Extract geometric information from pattern data and establish a geometric element set based on the basic information of geometric elements, where each element g i The geometric information includes vertex coordinates (x i，j ,y i，j ), side length L i and area A i , where (x ij ,y ij ) represents the coordinates of vertex j, n i Represents the geometric element g i The number of vertices, L i is the geometric element g i The area and side length of i is the geometric element g i area; For each geometric element g i Calculate its complexity index C i The complexity index is calculated based on the number of vertices, edge lengths, and areas of geometric elements. The calculation expression is: Based on the complexity index C i Calculate the geometric complexity index, the calculation expression is: In the formula, C complexity represents the geometric complexity index, and N represents the total number of geometric elements.

8. A detection method for accurately controlling current output according to claim 4, characterized in that: After analyzing the hierarchy, the specific steps to generate the hierarchy complexity index are as follows: For each layer, calculate the layer complexity, the calculation expression is: In the formula, C L Indicates the complexity of the layer, represents the complexity weight of the kth geometric element, Indicates the resolution requirement of the kth geometric element, N E Indicates the number of geometric elements in each layer; Calculate the interaction complexity between layers. The interaction complexity between layers is expressed as the sum of the interaction effects between all adjacent layers. The calculation expression is: In the formula, I L Indicates the complexity of interaction between layers, I p，q represents the interaction between the pth layer and the qth layer, f(I p，q ) is a function that represents the interaction effect, N L Indicates the number of layers, that is, the total number of layers in the pattern file; The complexity of all layers and the complexity of interactions between layers are combined to calculate the hierarchy complexity index. The calculation expression is: In the formula, I complexity represents the hierarchy complexity index, Indicates the layer complexity of the p-th layer.

9. A detection device for accurately controlling current output, used to implement a detection method for accurately controlling current output as described in any one of claims 1 to 8, characterized in that: It includes a pattern data acquisition module, a pattern data preprocessing module, a pattern analysis and complexity evaluation module, a pattern complexity classification module and a current control module; A pattern data acquisition module, which acquires the current pattern data to be engraved from the control system of the lithography equipment; A pattern data preprocessing module preprocesses the acquired pattern data to improve the pattern quality; A pattern analysis and complexity evaluation module analyzes the pattern file, extracts the geometric information and hierarchical structure of the pattern, analyzes the geometric information and hierarchical structure, and then evaluates the complexity of the acquired pattern based on the analyzed geometric information and hierarchical structure; The pattern complexity classification module classifies the patterns according to the results of complexity analysis, and divides the patterns into highly complex patterns and general patterns according to their complexity; The current control module controls the photolithography process classified as a general pattern based on the standard current control frequency. For the photolithography process classified as a highly complex pattern, the actual current control frequency is adjusted through an intelligent system based on the standard current control frequency.

Images