Process Parameter Optimization Method for Trench Schottky Structure

Abstract

The present invention discloses a method for optimizing process parameters of a trench Schottky structure, and relates to the technical field of conductor manufacturing process optimization. The method comprises: obtaining a process group, which includes multiple process links; extracting a first process link and a corresponding predetermined process index group, optimizing parameters of these indexes through a first process database, and generating a process decision. Dynamically monitoring real-time process indicators and comparing them with process decisions to obtain process deviations. Extracting the first index in the real-time deviation, generating second process index calibration information through a weighted correlation coefficient, and used to calibrate the process decision of the second process link as the production standard of the process link. The technical problem that the existing process parameter optimization cannot adjust the process deviation in time is solved, and the technical effect of improving product consistency and stability and improving production efficiency is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of conductor manufacturing process optimization, and in particular to a method for optimizing process parameters of a trench Schottky structure. Background Art

[0002] In the manufacturing process of semiconductor devices, trench Schottky diodes are widely used because of their low forward voltage drop and fast reverse recovery characteristics. However, the preparation of trench Schottky structures involves multiple complex process links, such as lithography, etching, doping, deposition, etc. The process parameters of each process link have an important impact on the performance of the device, so the parameters of each process link need to be accurately controlled during the manufacturing process. However, traditional process parameter setting methods often lack dynamic adjustment capabilities, and it is difficult to optimize process parameters based on real-time monitoring of device performance, resulting in deviations in the process and affecting the performance and yield of the final device. Therefore, there is an urgent need for a method that can dynamically monitor and calibrate process parameters to improve the manufacturing accuracy and reliability of trench Schottky structure devices.

[0003] In the current related technologies, there is a technical problem that process parameter optimization cannot adjust process deviations in a timely manner. Summary of the invention

[0004] The present application provides a method for optimizing process parameters of a trench Schottky structure, adopts an acquisition process group, including multiple process links; extracts the first process link and its corresponding predetermined process indicator group, optimizes the parameters of these indicators through the first process database, and generates a process decision. Dynamically monitor the real-time process indicators and compare them with the process decision to obtain the process deviation. Extract the first indicator in the real-time deviation, generate the second process indicator calibration information through the weighted correlation coefficient, and use it to calibrate the process decision of the second process link as the production standard of the process link, and realize the precise control and optimization of each link in the process by dynamically monitoring and calibrating the process parameters, so as to achieve the technical effect of improving product consistency and stability and improving production efficiency.

[0005] The present application provides a method for optimizing process parameters of a trench Schottky structure, including:

[0006] Acquire a process group, wherein the process group includes multiple process links for manufacturing a target device; extract a first process link from the multiple process links, wherein the first process link corresponds to a first predetermined process indicator group; optimize parameters of the first predetermined process indicator group in combination with a first process database of the first process link to obtain a first process decision; dynamically monitor to obtain first real-time process indicator information of the target device, and compare it with the first process decision in real time to obtain a first process deviation, wherein the first process deviation includes the real-time deviation of each indicator in the first predetermined process indicator group; extract the first real-time deviation of the first indicator in the first process deviation, and weight the first correlation coefficient to obtain second process indicator calibration information, wherein the second process indicator calibration information refers to information after dynamic calibration of a second process decision of the second process link in the multiple process links; and use the second process indicator calibration information as the manufacturing standard of the target device in the second process link.

[0007] The process parameter optimization method of the trench Schottky structure proposed in this application is to first obtain a process group, which includes multiple process links; extract the first process link and its corresponding predetermined process indicator group, optimize the parameters of these indicators through the first process database, and generate a process decision. Dynamically monitor the real-time process indicators and compare them with the process decision to obtain the process deviation. Extract the first indicator in the real-time deviation, generate the second process indicator calibration information through the weighted correlation coefficient, and use it to calibrate the process decision of the second process link as the production standard of the process link, so as to achieve the technical effect of ensuring the precise control and optimization of each link in the process by dynamically monitoring and calibrating the process parameters, and improving product consistency and stability and improving production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0009] Figure 1 A schematic flow chart of a method for optimizing process parameters of a trench Schottky structure provided in an embodiment of the present application;

[0010] Figure 2 A schematic diagram of the first process decision process for obtaining the process parameter optimization method for the trench Schottky structure provided in an embodiment of the present application. DETAILED DESCRIPTION

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0012] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0013] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0014] The present application embodiment provides a method for optimizing process parameters of a trench Schottky structure, such as Figure 1 As shown, the method includes:

[0015] Step S100, obtaining a process group, wherein the process group includes multiple process steps for manufacturing a target device. Specifically, when manufacturing a target device, the obtained process group includes multiple steps such as photolithography, etching, doping, deposition, metallization, annealing, passivation and dicing. The photolithography step accurately draws a circuit pattern on a semiconductor material by controlling a light source and a mask; the etching step selectively removes the material according to the photolithography pattern to form a specific structure; the doping step introduces impurity atoms to change the electrical properties of the material; the deposition step grows a thin film to ensure insulation and conductivity; the metallization step forms a good conductive path; the annealing step eliminates stress and activates impurities to improve performance stability; the passivation step protects the device surface; the dicing step cuts the wafer into individual chips. Through detailed analysis and data collection of each step, the process flow and parameters are fully understood, providing a basis for subsequent process parameter optimization.

[0016] In a possible implementation, a process group is obtained, and the process group includes multiple process links for making a target device. Step S100 further includes step S110, and the process group includes at least photolithography, etching, doping, deposition, metallization, annealing, passivation and scribing. Specifically, when making a target device, the process group includes at least photolithography, etching, doping, deposition, metallization, annealing, passivation and scribing. Photolithography accurately draws circuit patterns on semiconductor materials through a photolithography machine, including steps such as coating photoresist, exposure, development and inspection. Etching selectively removes materials based on the pattern defined by photolithography, and it is necessary to select an etching method, set parameters, monitor the process and perform post-processing. Doping changes the electrical properties of the material by introducing impurity atoms, and it is necessary to select a doping method, set parameters, monitor the process and anneal to activate the impurity atoms. Deposition is used to grow a thin film, and it is necessary to select a deposition method, set parameters, monitor the process and perform post-processing. Metallization forms a conductive path, and it is necessary to select a metal material and method, set parameters and monitor the process. Annealing can eliminate material stress and activate impurities, and requires selecting methods, setting parameters, monitoring the process, and performing post-processing. Passivation protects the device surface, and requires selecting materials and methods, setting parameters, and monitoring the process. Slicing cuts the wafer into individual chips, and requires selecting methods, setting parameters, monitoring the process, and performing post-processing. Each link requires precise control of parameters and strict quality monitoring to ensure the performance and reliability of the final device.

[0017] In a possible implementation, a process group is obtained, the process group includes multiple process links for making a target device, and step S100 further includes step S120, wherein the target device refers to an electronic device with a predetermined trench Schottky structure, and the predetermined trench Schottky structure is arranged in a circular mesh structure. Specifically, the target device is an electronic device with a predetermined trench Schottky structure, and its predetermined trench Schottky structure is arranged in a circular mesh structure. This structure can increase the effective contact area, improve the electric field distribution and improve the heat dissipation performance. In the design and manufacturing process, the structure is first designed according to the performance requirements and application scenarios, the parameters of the circular mesh structure and the Schottky contact materials and processes are determined, and then suitable semiconductor and metal materials are selected. The structure is realized by advanced semiconductor manufacturing processes such as lithography, etching, doping, deposition, and metallization, and the parameters of each process link are strictly controlled to ensure accuracy and quality. After the manufacturing is completed, performance tests are carried out, including electrical, thermal and reliability tests, etc., and the design and manufacturing process are optimized and adjusted according to the test results, such as adjusting the structural parameters, material selection or process parameters, to improve the device performance and reliability to meet the needs of different application scenarios.

[0018] Step S200, extracting the first process link among the multiple process links, wherein the first process link corresponds to the first predetermined process indicator group. Specifically, in the process of manufacturing the target device, extracting the first process link among the multiple process links is a key step in optimizing the process parameters. The selection of the first process link can be problem-oriented, such as the link where problems frequently occur in the early production or have a great impact on the device performance; considering the link that has an important impact on the key performance indicators; or giving priority to the earlier links that have a great impact on the subsequent links according to the process sequence. To determine the first predetermined process indicator group, it is necessary to quantify each indicator for accurate measurement and evaluation based on the characteristics of the first process link, while comprehensively considering multiple factors, such as performance, efficiency and cost. This can clarify the goals and directions for the subsequent process parameter optimization.

[0019] Step S300, combine the first process database of the first process link to perform parameter optimization on the first predetermined process indicator group to obtain the first process decision. Specifically, the first process database stores a large amount of historical data related to the first process link and is an important resource for parameter optimization. First, the data in the database is sorted and analyzed, abnormal data is removed and standardized, potential relationships and trends are explored using data analysis tools, and the objective function that can measure the quality of process indicators is determined, and the importance of multiple process indicators is comprehensively considered for weighting. Then, according to the characteristics of the problem and the nature of the data, a suitable optimization algorithm is selected, such as a genetic algorithm, a simulated annealing algorithm, etc., and the parameters are initialized. Then, the algorithm is used to perform iterative optimization, calculate the objective function value and update the parameters, and record the parameter values ​​and objective function values ​​of each iteration to observe the convergence. When the algorithm converges or reaches the preset number of iterations, the first process decision, that is, a set of optimized process parameters, is determined so that the first predetermined process indicator group reaches the optimal or near-optimal state, providing support for improving the manufacturing quality and performance of the target device.

[0020] In a possible implementation, the first predetermined process index group is optimized by combining the first process database of the first process link to obtain a first process decision, such as Figure 2 As shown, step S300 further includes step S310, randomly acquiring the first data group in the first process database. Specifically, first, the first data group is randomly acquired from the first process database, which stores a large amount of historical data related to the first process link, and the data covers the process results under different parameter settings and the corresponding device performance data. The randomly acquired data group can ensure a certain degree of randomness and representativeness in the subsequent analysis, and avoid deviations caused by selecting specific data. The first data group contains various information related to the first process link, such as process parameters, process results, performance indicators, etc. The data can provide rich information resources for subsequent analysis, helping us to better understand the characteristics and laws of the first process link.

[0021] Step S320, traverse the first predetermined process indicator group in the first data group to obtain the first historical process indicator information. Specifically, each process link corresponds to different process indicators. The first predetermined process indicator group is a group of specific process indicators determined for the first process link. The indicators reflect the key performance and quality requirements of the first process link. By traversing the first predetermined process indicator group in the first data group, historical process indicator information related to these indicators can be found. The traversal process is to compare and match each indicator in the first predetermined process indicator group with the data in the first data group one by one. In this process, the historical data value corresponding to each indicator can be found, thereby obtaining the first historical process indicator information. The information can help us understand the value and change trend of each indicator of the first process link under different historical circumstances.

[0022] Step S330, forming a first training data group based on the first historical process indicator information and the first historical performance index in the first data group. Specifically, the first data group also includes a first historical performance index, which reflects the overall performance of the first process link under historical conditions. The first historical process indicator information is combined with the first historical performance index to form a first training data group. This data group will serve as the basis for subsequent supervised learning and training. The first training data group contains multiple samples, each of which consists of a set of process indicator values ​​and corresponding performance indexes. The samples can help us establish a relationship model between process indicators and performance indexes, and provide a basis for subsequent performance simulation prediction.

[0023] Step S340, supervised learning and training are performed on the first training data group to obtain a performance simulation prediction model. Specifically, supervised learning is performed on the first training data group, using a suitable machine learning algorithm, such as linear regression, decision tree, neural network, etc. The principle of supervised learning is to train the model through known input (process indicator value) and output (performance index), so that the model can learn the mapping relationship between input and output. During the training process, the parameters of the model are continuously adjusted to minimize the prediction error. Through multiple iterations and optimizations, the model gradually converges to a better state, and can accurately predict the performance index under a given process indicator. After the training is completed, a performance simulation prediction model is obtained, and the model can be used to simulate and predict the performance of the first process link.

[0024] Step S350, the first process decision is obtained by performing simulation prediction optimization analysis through the performance simulation prediction model. Specifically, the performance simulation prediction model is used to simulate and predict different combinations of process parameters, a set of process parameters is input into the model, and the model will output the corresponding performance index prediction value. By predicting multiple parameter combinations, the performance under different conditions can be obtained, and the results of the simulation prediction are optimized to find the process parameter combination that can make the performance index reach the optimal or near-optimal. The process can use optimization algorithms such as genetic algorithms, simulated annealing algorithms, etc. The optimal process parameter combination finally determined is the first process decision. This decision will be used as an optimization plan for the first process link to guide the actual production process, improve the performance and quality of the first process link, and effectively use the data in the first process database. Through supervised learning and simulation prediction optimization, the first process decision is obtained, providing a scientific basis and method support for optimizing the first process link.

[0025] In a possible implementation, the first process decision is obtained by performing optimization analysis of simulation prediction through the performance simulation prediction model, and step S350 further includes step S351, step a: generating first simulation information according to the first predetermined process indicator group. Specifically, the first predetermined process indicator group is a group of process indicators determined for a specific process link, reflecting the key performance and quality requirements of the link. In this step, the first simulation information is generated based on the first predetermined process indicator group, and is achieved by randomly generating different process parameter combinations within a reasonable range or generating different process parameter combinations according to certain rules. The generated first simulation information should be diverse and cover different parameter value ranges so as to fully explore possible process solutions. For example, if the first predetermined process indicator group includes parameters such as temperature, pressure and time, multiple different parameter combinations can be generated as the first simulation information by randomly taking values ​​within a certain range.

[0026] Step S352, step b: Analyze the first simulation information through the performance simulation prediction model to obtain a first predicted performance index. Specifically, the performance simulation prediction model is obtained by supervised learning and training of historical data, and can predict the corresponding performance index according to the input process parameters. In this step, the first simulation information generated in step a is input into the performance simulation prediction model. The model will analyze the information and output the corresponding first predicted performance index. The first predicted performance index reflects the process performance under specific simulation information. It can be a single value or a comprehensive evaluation index of multiple performance indicators. By analyzing the predicted performance index, the degree of influence of different process parameter combinations on performance can be understood.

[0027] Step S353, step c: repeat step a to step b until a predetermined repetition threshold is reached, and a prediction performance index set is formed according to the obtained first prediction performance index. Specifically, in order to more comprehensively explore the performance of different process parameter combinations, it is necessary to repeat step a and step b multiple times, and each repetition will generate new simulation information, and obtain the corresponding prediction performance index through the model. By continuously repeating this process, more performance data can be gradually accumulated to provide richer information for subsequent analysis and decision-making. The predetermined repetition threshold is set to control the number of repetitions and avoid unlimited repetitions. The threshold can be set according to actual conditions, for example, it can be determined according to computing resources, time limits or performance improvement requirements. When the number of repetitions reaches the predetermined repetition threshold, the repetition process is stopped. During the repetition process, the first prediction performance index obtained each time is collected to form a prediction performance index set. The index set contains performance information under different process parameter combinations, which can be used for subsequent analysis and decision-making.

[0028] Step S354, step d: filter the simulation information when the performance index in the predicted performance index set is the largest, and use it as the first process decision. Specifically, in the predicted performance index set, the simulation information with the largest performance index represents the optimal process solution within the current model and parameter range. The maximum performance index is used as the screening basis, and the corresponding simulation information is found from the index set. The filtered simulation information is used as the first process decision. The decision will be used as an optimization solution for a specific process link to guide the actual production process. In this way, the best solution can be found among many possible combinations of process parameters, the performance and quality of the process link can be improved, and the performance simulation prediction model can be used to gradually explore and optimize the process parameters, and finally the first process decision is determined, providing a scientific basis for the improvement and enhancement of the process link.

[0029] In a possible implementation, a first training data group is formed based on the first historical process indicator information and the first historical performance index in the first data group, and step S330 further includes step S331, wherein the first data group also includes a first historical performance parameter. Specifically, the first data group includes not only process parameters and process indicator information related to the first process link, but also first historical performance parameters, which reflect the actual performance of the first process link in the past production process, for example, may include data on production efficiency, product quality, equipment stability, etc. The first historical performance parameters are crucial for understanding the overall performance and optimization direction of the first process link. By analyzing the parameters, the advantages and disadvantages of the process link can be understood, providing a basis for subsequent optimization.

[0030] Step S332, performing variation analysis on the first historical performance parameter to obtain a first weight distribution of a predetermined performance indicator. Specifically, the purpose of performing variation analysis on the first historical performance parameter is to determine the importance of different performance indicators in the overall performance. Since different performance indicators may have different influences on the process links, it is necessary to determine their weight distribution by analyzing historical data. Variation analysis can adopt a variety of methods, such as statistical analysis, principal component analysis, etc. Through the above methods, the coefficient of variation, contribution rate and other indicators of each performance indicator can be calculated to determine their relative importance in the overall performance. According to the analysis results, the corresponding first weight is allocated to the predetermined performance indicator. The first weight distribution reflects the importance of each predetermined performance indicator in the first process link. Indicators with larger weights will have greater influence in subsequent performance evaluation and optimization.

[0031] Step S333, weighted calculation is performed on the first historical performance parameter in combination with the first weight allocation to obtain the first historical performance index. Specifically, weighted calculation is performed on the first historical performance parameter in combination with the first weight allocation. The specific calculation method may be to multiply the value of each performance indicator by its corresponding weight, and then add the weighted values ​​of all indicators to obtain the first historical performance index. The first historical performance index is a comprehensive evaluation indicator of the historical performance of the first process link, which can more comprehensively and accurately reflect the overall performance of the process link, and provide a basis for subsequent process optimization and decision-making. The first historical performance parameters in the first data group are fully utilized, and the weight allocation is determined through variation analysis. The weighted calculation is performed to obtain the first historical performance index, which provides strong support for the performance evaluation and optimization of the first process link.

[0032] In a possible implementation, the first historical performance parameter is subjected to variation analysis to obtain a first weight distribution of a predetermined performance indicator, and step S332 further includes step S3321, wherein the predetermined performance indicator includes forward conduction characteristics, breakdown voltage, reverse leakage characteristics, and thermal stability. Specifically, the predetermined performance indicators include forward conduction characteristics, breakdown voltage, reverse leakage characteristics, and thermal stability. These indicators are crucial to trench Schottky structure electronic devices, and their selection is based on the device working principle and application requirements. In the process parameter optimization, each indicator needs to be considered, such as improving the forward conduction characteristics by adjusting the doping concentration and metallization process; controlling the material quality, structural uniformity, and electric field distribution to improve the breakdown voltage; optimizing the passivation process, improving the quality of the insulating material, and controlling the surface state to reduce the reverse leakage characteristics; and considering the thermal expansion coefficient and thermal conductivity of the material to ensure thermal stability. By optimizing these indicators, the device performance and reliability can be improved to meet different application requirements.

[0033] Step S400, dynamically monitor and obtain the first real-time process indicator information of the target device, and compare it with the first process decision in real time to obtain the first process deviation, which includes the real-time deviation of each indicator in the first predetermined process indicator group. Specifically, in order to realize the dynamic monitoring of the target device, it is necessary to establish a monitoring system including sensors, data acquisition equipment, data transmission network and data analysis software, determine the appropriate monitoring frequency and accuracy, and process the real-time data to extract the first real-time process indicator information. The first process decision is the optimized process parameter obtained by parameter optimization, and the real-time process indicator information is compared with it. The first process deviation is obtained by calculating the difference between the actual and the corresponding indicators in the decision, including the real-time deviation of each indicator in the first predetermined process indicator group, reflecting the gap between the actual process and the optimal solution, providing a basis for process adjustment, and strongly supporting the timely adjustment of process parameters and improving process quality.

[0034] Step S500, extract the first real-time deviation of the first indicator in the first process deviation, and weight the first correlation coefficient to obtain the second process indicator calibration information, the second process indicator calibration information refers to the information after the second process decision of the second process link in the multiple process links is dynamically calibrated. Specifically, first determine and extract the first real-time deviation of any one of the first predetermined process indicators, that is, the first indicator, from the first process deviation. Then calculate the correlation coefficient between the first indicator and other related indicators, and perform weighted processing on the first real-time deviation after determining the first correlation coefficient. The second process link is interrelated with the first process link, and its second process indicator calibration information is the information after the second process decision is dynamically calibrated, which is obtained by applying the weighted first real-time deviation to the adjustment process of the second process decision. Dynamic calibration is a continuous process. According to the real-time data and process deviation changes, the calibration information is used to adjust the second process decision to ensure that the process indicators of the second process link remain in the optimal or near-optimal state, provide accurate data support for it, and improve the quality and efficiency of the entire process.

[0035] In a possible implementation, the first real-time deviation of the first indicator in the first process deviation is extracted, and the first correlation coefficient is weighted to obtain the calibration information of the second process indicator, the calibration information of the second process indicator refers to the information after the second process decision of the second process link in the multiple process links is dynamically calibrated, and step S500 further includes step S510, obtaining the second process database of the second process link, wherein the second process link refers to the adjacent downstream process link of the first process link in the multiple process links. Specifically, among the multiple process links, the second process link is the adjacent downstream process link of the first process link, which means that the output of the first process link directly affects the input of the second process link, and there is a close correlation between the two links. The second process database of the second process link is obtained, which provides a data basis for analyzing the relationship between the two links. The second process database contains various historical data related to the second process link, such as process parameters, process indicators, performance data, etc. The data reflects the operating status and results of the second process link under different conditions, which can help us understand the characteristics and laws of the second process link.

[0036] Step S520, obtaining any indicator in the second predetermined process indicator group. Specifically, an indicator is randomly selected from the second predetermined process indicator group for analysis, and the indicator can be selected according to actual needs, for example, an indicator that has a greater impact on the performance of the second process link, or an indicator that has a high correlation with the first process link can be selected. The selected arbitrary indicator represents the performance or characteristics of a specific aspect of the second process link. By analyzing this indicator, the operation of the second process link and its potential relationship with the first process link can be understood.

[0037] Step S530, matching any indicator parameter corresponding to the any indicator in the second data group, the second data group refers to any historical data group in the second process database. Specifically, the second data group refers to any historical data group in the second process database, in which we can find the parameter value corresponding to the selected any indicator, and by matching the parameters, we can obtain specific data about the indicator in different historical situations. The matching process can be achieved through data query, screening and other methods. According to the characteristics of the selected indicator and the structure of the database, determine the appropriate matching method to ensure that the corresponding parameter value can be accurately found.

[0038] Step S540, performing a correlation analysis on the first real-time indicator parameter of the first indicator and the arbitrary indicator parameter of the arbitrary indicator to obtain a correlation analysis result. Specifically, performing a correlation analysis on the first real-time indicator parameter of the first indicator and the arbitrary indicator parameter of the arbitrary indicator is intended to determine the strength and direction of the relationship between the two indicators. Correlation analysis can help us understand the degree of mutual influence between the first process link and the second process link. A variety of correlation analysis methods can be used, such as the Pearson correlation coefficient, the Spearman correlation coefficient, etc. The method can calculate the linear or nonlinear relationship between the two variables, thereby determining their degree of correlation. The correlation analysis result can be a numerical value representing the correlation coefficient between the two indicators. The coefficient ranges from -1 to 1, where -1 indicates a complete negative correlation, 1 indicates a complete positive correlation, and 0 indicates no correlation. Through the analysis results, we can understand the type and strength of the relationship between the two indicators.

[0039] Step S550, analyzing the correlation analysis result to obtain the first correlation coefficient. Specifically, the correlation analysis result is deeply analyzed to interpret the meaning and practical significance of the correlation coefficient, and the closeness and influence direction of the relationship between the first indicator and any indicator are judged according to the size and sign of the correlation coefficient. According to the analysis result, the first correlation coefficient is determined, and the coefficient will be used to perform weighted processing on the first real-time deviation of the first indicator to obtain the second process indicator calibration information. The determination of the first correlation coefficient needs to comprehensively consider multiple factors, such as the reliability of the correlation analysis result, the actual situation of the process link, etc., to accurately determine the first correlation coefficient, which provides an important basis for the subsequent process indicator calibration and optimization.

[0040] Step S600, using the second process indicator calibration information as the production standard of the target device in the second process link. Specifically, the second process indicator calibration information is obtained by weighting the real-time deviation of the first indicator and combining the correlation analysis of the two process links. It reflects the process indicator requirements that should be achieved in the second process link under the current production conditions. It is used as the production standard of the target device in the second process link because it is based on process relevance, data accuracy and dynamic adjustability. During the implementation process, the standard information should be conveyed to relevant personnel, the process parameters should be adjusted according to the calibration information, and the production process should be monitored in real time to ensure that the process indicators meet the standards. At the same time, this is a process of continuous improvement, and the calibration information can be continuously optimized to improve the accuracy and adaptability of the production standards to ensure the quality and performance of the target device.

[0041] The embodiment of the present application adopts the method of obtaining a process group, which includes multiple process links; extracting the first process link and its corresponding predetermined process indicator group, optimizing the parameters of these indicators through the first process database, and generating process decisions. Dynamically monitor the real-time process indicators and compare them with the process decisions to obtain process deviations. Extract the first indicator in the real-time deviation, generate the second process indicator calibration information through the weighted correlation coefficient, and use it to calibrate the process decision of the second process link as the production standard of the process link, so as to achieve the technical effect of ensuring the precise control and optimization of each link in the process by dynamically monitoring and calibrating the process parameters, and improving product consistency and stability and improving production efficiency.

[0042] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A process parameter optimization method for a trench Schottky structure, characterized in that: include: Step S100, obtaining a process group, wherein the process group includes multiple process steps for manufacturing a target device; Step S200, extracting a first process link from the multiple process links, wherein the first process link corresponds to a first predetermined process indicator group; Step S300, performing parameter optimization on the first predetermined process indicator group in combination with the first process database of the first process link to obtain a first process decision; Step S400, dynamically monitoring and obtaining first real-time process indicator information of the target device, and performing real-time comparison with the first process decision to obtain a first process deviation, wherein the first process deviation includes a real-time deviation of each indicator in the first predetermined process indicator group; Step S500, extracting a first real-time deviation of a first indicator in the first process deviation, and weighting a first correlation coefficient to obtain second process indicator calibration information, wherein the second process indicator calibration information refers to information after dynamically calibrating a second process decision of a second process link in the multiple process links; Step S600, using the second process index calibration information as a manufacturing standard for the target device in the second process link; Step S300 includes: Step S310, randomly acquiring a first data group from the first process database; Step S320, traversing the first predetermined process indicator group in the first data group to obtain first historical process indicator information; Step S330, forming a first training data group based on the first historical process indicator information and the first historical performance index in the first data group; Step S340, performing supervised learning and training on the first training data group to obtain a performance simulation prediction model; Step S350, performing optimization analysis of simulation prediction through the performance simulation prediction model to obtain the first process decision; Step S500 includes: Acquire a second process database of the second process link, wherein the second process link refers to a downstream process link adjacent to the first process link among the multiple process links; Obtaining any indicator in the second predetermined process indicator group; matching any indicator parameter corresponding to the any indicator in a second data group, wherein the second data group refers to any historical data group in the second process database; Performing a correlation analysis on the first real-time indicator parameter of the first indicator and the arbitrary indicator parameter of the arbitrary indicator to obtain a correlation analysis result; The correlation analysis result is analyzed to obtain the first correlation coefficient.

2. The process parameter optimization method of the trench Schottky structure according to claim 1, characterized in that: The process group includes at least photolithography, etching, doping, deposition, metallization, annealing, passivation and scribing.

3. The process parameter optimization method of the trench Schottky structure according to claim 1, characterized in that: The target device refers to an electronic device having a predetermined trench Schottky structure, and the predetermined trench Schottky structure is arranged in a circular mesh structure.

4. The process parameter optimization method of the trench Schottky structure according to claim 1, characterized in that: Step S350 includes: Step a: generating first simulation information according to the first predetermined process indicator group; Step b: Analyzing the first simulation information through the performance simulation prediction model to obtain a first prediction performance index; Step c: repeating step a to step b until a predetermined repetition threshold is reached, and forming a prediction performance index set according to the obtained first prediction performance index; Step d: Filter the simulation information when the performance index in the predicted performance index set is the largest, and use it as the first process decision.

5. The process parameter optimization method of the trench Schottky structure according to claim 1, characterized in that: Step S330 includes: The first data set also includes a first historical performance parameter; Performing variation analysis on the first historical performance parameter to obtain a first weight distribution of a predetermined performance indicator; The first historical performance parameter is weightedly calculated in combination with the first weight allocation to obtain the first historical performance index.

6. The process parameter optimization method of the trench Schottky structure according to claim 5, characterized in that: The predetermined performance indicators include forward conduction characteristics, breakdown voltage, reverse leakage characteristics and thermal stability.

Images