Thin film processing optimization control method and system

Abstract

The invention discloses a thin film processing optimization control method and system, and the method comprises the steps: obtaining and analyzing historical processing data of a thin film processing process, and determining key processing parameters which affect the thickness of a thin film; extracting features of the key processing parameters, and constructing a thickness prediction model based on the features and a preset model for prediction to obtain thickness prediction data; constructing a thickness distribution diagram of the thin film based on the thickness prediction data, and determining a thickness distribution area and thickness characteristics in the thickness distribution diagram; evaluating the thickness distribution of the film based on the thickness distribution area and the thickness characteristics to obtain a distribution quality evaluation value of the film, and determining an optimization coefficient based on the distribution quality evaluation value; and performing optimization adjustment on the real-time processing data of the thin film based on the optimization coefficient so as to control the processing of the thin film. According to the method, the distribution of the currently processed thin film is determined through prediction, the optimization coefficient is determined through analysis, the current processing process is optimized, the problems that the thickness of the thin film is not uniform and fluctuation is large are solved, the rejection rate is reduced, and the production efficiency and quality are improved.

Description

TECHNICAL FIELD

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

[0002] In the thin film processing industry, controlling the thickness distribution of the thin film is crucial because the thickness of the thin film directly affects its performance and application. In order to ensure the production of high-quality thin film products, effective optimization control methods need to be adopted to achieve accurate control and optimization of the thickness.

[0003] With the continuous development of technology, thin film processing technology is constantly innovating and progressing, involving various complex processes and equipment. However, in the traditional thin film processing process, there are often problems such as uneven thickness and large fluctuations, which may lead to unstable quality of thin film products, increasing production costs and waste rates. SUMMARY

[0004] In order to solve the above technical problems, the present application provides a thin film processing optimization control method and system, comprising: obtaining historical processing data of the thin film processing process and analyzing the historical processing data to determine key processing parameters affecting the thickness of the thin film; extracting features of the key processing parameters and constructing a thickness prediction model based on the features and a preset neural network model for prediction to obtain thickness prediction data of the thin film; constructing a thickness distribution map of the thin film based on the thickness prediction data of the thin film and analyzing the thickness distribution map to determine the thickness distribution area of the thin film and the corresponding thickness features; evaluating the thickness distribution of the thin film based on the thickness distribution area and the thickness features to obtain a distribution quality evaluation value of the thin film, and determining an optimization coefficient of the thin film thickness processing based on the distribution quality evaluation value; optimizing and adjusting the real-time processing data of the thin film based on the optimization coefficient, and controlling the processing of the thin film according to the optimized and adjusted processing data.

[0005] Further, the obtaining of the historical processing data of the thin film processing process and the analysis of the historical processing data to determine the key processing parameters affecting the thickness of the thin film comprises: dividing the historical processing data into a plurality of processing parameter data groups according to parameter types, and analyzing the correlation degree of each processing parameter data group and the thickness data; selecting the processing parameter data group with a correlation degree greater than a preset value as a candidate processing parameter affecting the thickness of the thin film in the historical processing data; for each candidate processing parameter, filtering the thickness data of the thin film when other candidate processing parameters are the same, and determining the change amount of the thickness data of the thin film; The candidate machining parameter with a change greater than a preset value is selected to determine a key machining parameter affecting the film thickness in the historical machining data.

[0006] Further, the feature of the key machining parameter is extracted, and a thickness prediction model is constructed based on the feature and a preset neural network model to perform prediction, to obtain thickness prediction data of the film, including: The feature of the key machining parameter and the key machining parameter are used to construct a data set, and the data set is input into a preset neural network model to construct a thickness prediction initial model of the film; The data set is divided into a training set and a test set according to a preset ratio, and the training set and the test set are input into the thickness prediction initial model of the film; The thickness prediction initial model of the film is trained and tested until the thickness prediction initial model of the film meets a preset convergence condition, to obtain a thickness prediction model of the film; Real-time machining parameters of the film are obtained, and the real-time machining parameters are input into the thickness prediction model of the film to perform prediction, to obtain thickness prediction data of the film.

[0007] Further, the thickness distribution map of the film is constructed based on the thickness prediction data of the film, and the thickness distribution map is analyzed to determine the thickness distribution region of the film and the corresponding thickness feature, including: The transverse thickness and the longitudinal thickness of the film are determined from the thickness prediction data of the film, and the thickness distribution map of the film is constructed based on the transverse thickness and the longitudinal thickness of the film; A preset required thickness of the film is obtained, and a first type region exceeding the required thickness of the film and a second type region not exceeding the required thickness of the film in the thickness distribution map are determined; The thickness values of each position in the first type region and the second type region are subjected to cluster analysis, and the corresponding thickness distribution region in the first type region and the second type region is determined according to the cluster analysis result; The difference between the thickness average value of all positions in each thickness distribution region in the first type region and the second type region and the required thickness of the film is calculated to obtain the thickness difference of each thickness distribution region, and the position of each thickness distribution region in the first type region and the second type region is determined; The thickness difference and the position of each thickness distribution region are determined as the thickness feature corresponding to the thickness distribution region of the film.

[0008] Further, the thickness values of each position in the first type region and the second type region are subjected to cluster analysis, and the corresponding thickness distribution region in the first type region and the second type region is determined according to the cluster analysis result, including: A data set is established according to the thickness values of each position in the first type region and the second type region, and k initial cluster centers of the data set are randomly selected; calculating the Euclidean distance of the thickness value in the data set to the initial cluster center, and dividing each position to the corresponding cluster according to the Euclidean distance of the thickness value in the data set to the initial cluster center; calculating the thickness average value in each cluster, and re-determining the cluster center according to the thickness average value in each cluster; repeating the above steps until the cluster center no longer changes or the number of iterations reaches a preset iteration threshold, obtaining k clusters, wherein the cluster represents a thickness distribution region; calculating the thickness average value of each thickness distribution region, and classifying the thickness distribution region with a positive thickness average value to a first type region and the thickness distribution region with a negative thickness average value to a second type region, and finally obtaining the corresponding thickness distribution region in the first type region and the second type region.

[0009] Further, the thickness distribution of the film is evaluated based on the thickness distribution region and the thickness feature, and a distribution quality evaluation value of the film is obtained, including: respectively evaluating the thickness difference of each thickness distribution region in the first type region and the second type region to obtain a first thickness evaluation value of each thickness distribution region in the first type region and a second thickness evaluation value of each thickness distribution region in the second type region; determining the weight of each thickness distribution region based on the position of each thickness distribution region, and calculating the distribution quality evaluation value of the film based on the weight, the first thickness evaluation value, the second thickness evaluation value and the second thickness evaluation value.

[0010] Further, the calculation formula of the distribution quality evaluation value of the film is: P= , wherein P is the distribution quality evaluation value of the film, a is the weight of each thickness distribution region, xi is the first thickness evaluation value of the i-th thickness distribution region in the first type region, n is the number of thickness distribution regions in the first type region, yi is the second thickness evaluation value of the i-th thickness distribution region in the second type region, and m is the number of thickness distribution regions in the second type region.

[0011] Further, the optimization coefficient of the film thickness processing is determined based on the distribution quality evaluation value of the film, including: previously setting a preset optimization coefficient-distribution quality evaluation value interval corresponding relationship, and the preset optimization coefficient-distribution quality evaluation value interval corresponding relationship is associated with a corresponding preset optimization coefficient for each distribution quality evaluation value interval; The distribution quality evaluation value of the film is obtained, and a preset optimization coefficient corresponding to the distribution quality evaluation value interval is selected as the optimization coefficient for the film thickness processing based on the mapping relationship in the preset optimization coefficient-distribution quality evaluation value interval corresponding relationship within the distribution quality evaluation value interval to which the distribution quality evaluation value belongs.

[0012] Further, the real-time processing data of the film is optimized and adjusted based on the optimization coefficient, and the processing of the film is controlled according to the processing data after the optimization and adjustment, including: The real-time processing data of the film is obtained, the optimization coefficient is multiplied by the real-time processing parameter to obtain the processing data after the optimization and adjustment, and the processing of the film is controlled according to the data after the optimization and adjustment.

[0013] The application also provides a film processing optimization control system, including: The acquisition module is used for obtaining the historical processing data of the film processing process, and analyzing the historical processing data to determine the key processing parameters affecting the film thickness. The prediction module is used for extracting the features of the key processing parameters, and constructing a thickness prediction model based on the features and a preset neural network model to make a prediction and obtain the thickness prediction data of the film. The analysis module is used for constructing a thickness distribution map of the film based on the thickness prediction data of the film, and analyzing the thickness distribution map to determine the thickness distribution area and the corresponding thickness features of the film. The evaluation module is used for evaluating the thickness distribution of the film based on the thickness distribution area and the thickness features, obtaining the distribution quality evaluation value of the film, and determining the optimization coefficient for the film thickness processing based on the distribution quality evaluation value. The control module is used for optimizing and adjusting the real-time processing data of the film based on the optimization coefficient, and controlling the processing of the film according to the processing data after the optimization and adjustment.

[0014] Compared with the prior art, the film processing optimization control method and system of the application have the following advantages: The application can make more accurate decisions based on actual data through analysis and modeling of historical processing data, rather than relying on experience and intuition. The application can accurately identify the key processing parameters affecting the film thickness by analyzing historical data, helping to optimize and control the processing process. The thickness prediction model constructed based on neural network and other machine learning technologies can accurately predict the thickness of the film, providing a basis for real-time control. The application realizes real-time optimization control according to the prediction model and the optimization coefficient by real-time monitoring and adjusting the processing parameters, ensuring that the thickness of the film meets the target requirements. The present application integrates data analysis and machine learning techniques, enabling automation and intelligentization of the film processing process, improving production efficiency and product quality. The present application can effectively control the film thickness distribution, improve product quality, and reduce waste rate by establishing a prediction model and real-time control system. The present application can gradually improve the processing process, improve production efficiency and product quality, and achieve continuous improvement and optimization by continuously optimizing the model and algorithm. In summary, the present application can help the film production process to be more intelligent and precise, improve the film production efficiency and product quality, reduce the cost and risk, and lay a foundation for future technological innovation and development. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 is a flow structure diagram of the film processing optimization control method in the embodiment of the present application. Figure 2 is a composition diagram of the film processing optimization control system in the embodiment of the present application. DETAILED DESCRIPTION

[0016] The specific embodiments of the present application will be further described in detail below in conjunction with the drawings and examples. The following examples are used to illustrate the present application, but not to limit the scope of the present application.

[0017] In the description of the present application, it should be understood that the terms "center", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the platform or element must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0018] The terms "first", "second" are only used for description purposes, and cannot be understood as indicating or implying the relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features limited by "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, unless otherwise specified, the meaning of "multiple" is two or more.

[0019] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a cooling connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0020] like Figure 1 As shown in the embodiments of this application, a thin film processing optimization control method is provided, including: S100: acquiring historical processing data of the thin film processing process, analyzing the historical processing data, and determining key processing parameters affecting the film thickness; S200: extracting features of the key processing parameters, and constructing a thickness prediction model based on the features and a preset neural network model to predict the thickness of the thin film; S300: constructing a thickness distribution map of the thin film based on the thickness prediction data, and analyzing the thickness distribution map to determine the thickness distribution area of ​​the thin film and the corresponding thickness features; S400: evaluating the thickness distribution of the thin film based on the thickness distribution area and thickness features to obtain a distribution quality evaluation value of the thin film, and determining the optimization coefficient of the thin film thickness processing based on the distribution quality evaluation value; S500: optimizing and adjusting the real-time processing data of the thin film based on the optimization coefficient, and controlling the processing of the thin film according to the optimized and adjusted processing data.

[0021] Furthermore, this invention, through the analysis and modeling of historical processing data, can make more accurate decisions based on actual data, rather than relying solely on experience and intuition. By analyzing historical data, this invention can accurately identify key processing parameters affecting film thickness, aiding in the optimization and control of the processing process. The thickness prediction model built based on machine learning technologies such as neural networks can accurately predict film thickness, providing a basis for real-time control. This invention ensures that the film thickness meets target requirements by real-time monitoring and adjustment of processing parameters, and by performing real-time optimization control based on the prediction model and optimization coefficients. This invention integrates data analysis and machine learning technologies, enabling the automation and intelligentization of the film processing process, improving production efficiency and product quality. By establishing a prediction model and a real-time control system, this invention can effectively control film thickness distribution, improve product quality, and reduce scrap rates. By continuously optimizing the model and algorithm, this invention can gradually improve the processing process, increase production efficiency and product quality, and achieve continuous improvement and optimization. In summary, this invention can help make the film production process more intelligent and precise, improve film production efficiency and product quality, reduce costs and risks, and lay the foundation for future technological innovation and development.

[0022] In the embodiments of the present application, a thin film processing optimization control method is provided, the historical processing data of the thin film processing process is obtained, and the historical processing data is analyzed to determine the key processing parameters affecting the thickness of the thin film, including: dividing the historical processing data into a plurality of processing parameter data groups according to the parameter types, and analyzing the correlation degree of each processing parameter data group and the thickness data; selecting the processing parameter data group with a correlation degree greater than a preset value as a candidate processing parameter affecting the thickness of the thin film in the historical processing data; for each candidate processing parameter, filtering the thickness data of the thin film when other candidate processing parameters are the same, and determining the change amount of the thickness data of the thin film; selecting the candidate processing parameter with a change amount greater than a preset value as a key processing parameter affecting the thickness of the thin film in the historical processing data.

[0023] Specifically, the historical processing data is divided into different processing parameter data groups according to the parameter types, the relationship between different types of processing parameters and the thickness of the thin film is independently analyzed, and the influence of each parameter on the thickness of the thin film is better understood; the correlation analysis is performed on each processing parameter data group and the thickness data to understand the correlation degree between them, and through the analysis of the correlation degree, it can be determined which processing parameters have a significant relationship with the thickness of the thin film, providing a basis for subsequent screening of candidate processing parameters; the processing parameter data group with a correlation degree greater than a preset value is selected as a candidate processing parameter affecting the thickness of the thin film, the range is narrowed, and the processing parameters that may have an important influence on the thickness of the thin film are determined, preparing for the subsequent screening of key processing parameters; for each candidate processing parameter, the thickness data of the thin film when other candidate processing parameters are the same is filtered, and the change amount of the thickness data of the thin film is determined; the candidate processing parameter with a change amount greater than a preset value is selected as a key processing parameter affecting the thickness of the thin film, and the key processing parameter that has the greatest influence on the thickness of the thin film under specific conditions is accurately identified, providing an important basis for the optimization of the processing process. This step can accurately identify the key processing parameters affecting the thickness of the thin film by filtering and analyzing the historical data, helping manufacturers to carry out targeted optimization control; by determining the key processing parameters, unnecessary adjustments and tests can be avoided, thereby improving production efficiency, reducing waste rate, and reducing production cost; precise control of the key processing parameters can ensure the stability and consistency of the thickness of the thin film, improve product quality, and enhance market competitiveness; relying on data analysis and correlation evaluation, decisions can be made based on facts and evidence, avoiding the interference of subjective factors, and improving the accuracy and credibility of the decisions. In the embodiments of the present application, a thin film processing optimization control method is provided, which extracts features of key processing parameters, and constructs a thickness prediction model based on the features and a preset neural network model to make a prediction, to obtain thickness prediction data of the thin film, including: extracting features of key processing parameters and key processing parameters to construct a data set, and inputting the data set into a preset neural network model to construct a thickness prediction initial model of the thin film; dividing the data set into a training set and a test set according to a preset ratio, and inputting the training set and the test set into the thickness prediction initial model of the thin film; training and testing the thickness prediction initial model of the thin film until the thickness prediction initial model of the thin film meets a preset convergence condition, to obtain a thickness prediction model of the thin film; obtaining real-time processing parameters of the thin film, and inputting the real-time processing parameters into the thickness prediction model of the thin film to make a prediction, to obtain thickness prediction data of the thin film.

[0024] Specifically, features of these parameters are extracted from historical processing data as inputs of the model, which are used to describe the relationship between processing parameters and the influence on the thickness of the thin film; the extracted key processing parameter features are constructed into a data set, and input into a preset neural network model to construct a thin film thickness prediction initial model, the neural network model will learn the complex nonlinear relationship between the processing parameters and the thickness of the thin film in the historical data, so as to realize the prediction of the thickness of the thin film; the constructed data set is divided into a training set and a test set according to a preset ratio, which is used to train and verify the performance of the model, the training set is used for parameter learning of the neural network model, and the test set is used to evaluate the generalization ability and prediction accuracy of the model; input the training set and the test set into the thin film thickness prediction initial model to train and test, until the model meets the preset convergence condition, by continuously adjusting the model parameters and structure, the neural network model is optimized to improve the prediction accuracy and stability of the thickness of the thin film, to obtain the thin film thickness prediction model; input the real-time processing parameters into the trained thin film thickness prediction model to make a prediction, to obtain the thickness prediction data of the thin film, which can realize the real-time monitoring and prediction of the thickness of the thin film in the real-time processing process, to help adjust and optimize the processing process. This step can realize accurate prediction of the thickness of the thin film by constructing a neural network model, to help realize more accurate control in the production process; the thin film thickness prediction model can realize rapid response and prediction of real-time processing parameters, to help real-time adjustment and optimization in the production process; through the application of the prediction model, the time of trial and error and adjustment can be reduced, to improve the production efficiency and product quality; the establishment of the prediction model makes the production process more intelligent, reduces human intervention, improves the production automation level, and reduces human error.

[0025] In the embodiments of the present application, a thin film processing optimization control method is provided, which constructs a thickness distribution map of the thin film based on thickness prediction data of the thin film, and analyzes the thickness distribution map to determine thickness distribution regions of the thin film and corresponding thickness characteristics, including: determining a transverse thickness and a longitudinal thickness of the thin film from the thickness prediction data of the thin film, and constructing a thickness distribution map of the thin film based on the transverse thickness and the longitudinal thickness of the thin film; obtaining a required thickness of the thin film, and determining a first type of region in the thickness distribution map that exceeds the required thickness of the thin film and a second type of region that does not exceed the required thickness of the thin film; performing clustering analysis on the thickness values of each position in the first type of region and the second type of region, and determining corresponding thickness distribution regions in the first type of region and the second type of region according to the results of the clustering analysis; calculating the difference between the thickness average value of all positions in each thickness distribution region in the first type of region and the second type of region and the required thickness of the thin film, to obtain the thickness difference of each thickness distribution region, and determining the positions of each thickness distribution region in the first type of region and the second type of region; determining the thickness difference and the positions of each thickness distribution region as the thickness characteristics corresponding to the thickness distribution regions of the thin film.

[0026] Specifically, the transverse thickness and the longitudinal thickness of the thin film are determined from the thickness prediction data of the thin film, which describe the thickness distribution of the thin film in the plane; based on the transverse thickness and the longitudinal thickness of the thin film, a thickness distribution map of the thin film is constructed to visually display the thickness of the thin film at different positions; the required thickness of the thin film is obtained, and the first type of region and the second type of region in the thickness distribution map that exceed and do not exceed the required thickness of the thin film are determined; clustering analysis is performed on the thickness values of each position in the first type of region and the second type of region, and the thickness distribution regions are determined according to the clustering results, which are divided into sub-regions with similar characteristics to better understand the distribution of the thickness of the thin film; the difference between the thickness average value of all positions in each thickness distribution region in the first type of region and the second type of region and the required thickness of the thin film is calculated to obtain the thickness difference of each thickness distribution region, and the positions and the thickness difference of each thickness distribution region are determined as the thickness characteristics corresponding to the thickness distribution regions of the thin film. This step can better understand the thickness of the thin film at different positions by analyzing and classifying the distribution characteristics of the thickness data, and provides a basis for further optimizing the processing process; by real-time prediction and analysis of the thickness of the thin film, the regions that exceed and do not exceed the required thickness can be found in time to help real-time adjustment and optimization of the production process; by analyzing the characteristics of the thickness distribution regions, the positions and degrees of the thickness difference can be determined to help guide the quality control and the development of improvement measures in the production process.

[0027] In the embodiments of the present application, a thin film processing optimization control method is provided, which performs cluster analysis on the thickness values of each position in the first type region and the second type region, and determines the corresponding thickness distribution regions in the first type region and the second type region according to the results of the cluster analysis, including: establishing a data set according to the thickness values of each position in the first type region and the second type region, and randomly selecting k initial cluster centers of the data set; calculating the Euclidean distance of the thickness values in the data set to the initial cluster centers, and dividing each position to the corresponding cluster cluster according to the Euclidean distance of the thickness values in the data set to the initial cluster centers; calculating the average thickness value in each cluster cluster, and re-determining the cluster center according to the average thickness value in each cluster cluster; repeating the above steps until the cluster center no longer changes or the iteration number reaches a preset iteration threshold, obtaining k cluster clusters, wherein the cluster cluster represents a thickness distribution region; calculating the average thickness value of each thickness distribution region, and dividing the thickness distribution region with a positive average thickness value to the first type region and dividing the thickness distribution region with a negative average thickness value to the second type region, finally obtaining the corresponding thickness distribution regions in the first type region and the second type region.

[0028] Specifically, a data set is established according to the thickness values of each position in the first type region and the second type region, and k initial cluster centers of the data set are randomly selected, k representing a preset number of cluster clusters; the Euclidean distance of each thickness value in the data set to the initial cluster centers is calculated, and each position is divided into the corresponding cluster cluster according to the distance, and the thickness values in the data set are divided into different cluster clusters according to similarity, so as to facilitate subsequent cluster analysis; the average value of the thickness values in each cluster cluster is calculated, and the cluster center is re-determined according to the average values; by calculating the average value and updating the cluster center, the characteristics of the thickness values in different cluster clusters can be better represented; the above steps are repeated until the cluster center no longer changes or the preset iteration number threshold is reached, the cluster result is continuously optimized, and the stability and accuracy of the cluster cluster are ensured; the average thickness value of each thickness distribution region is calculated, and it is divided into the first type region or the second type region according to the positive and negative of the average value, the region with a positive average value is divided into the first type region, and the region with a negative average value is divided into the second type region, and finally the corresponding thickness distribution regions in the first type region and the second type region are determined. This step can more carefully divide the thin film thickness data into different cluster clusters through cluster analysis, which helps to more clearly understand the distribution of the thin film thickness; through iterative optimization and automatic division, efficient processing and analysis of a large amount of data can be realized, and the influence of manual intervention and subjective judgment is reduced; according to the positive and negative of the average value, the thickness distribution region is divided into different categories, which helps to more accurately identify the regions exceeding the required thickness and the regions meeting the requirements; the thickness distribution region characteristics obtained through cluster analysis can provide more data support for decision-making in the production process, helping to optimize the production process and control product quality.

[0029] In the embodiments of the present application, a film processing optimization control method is provided, which evaluates the thickness distribution of a film based on thickness distribution regions and thickness characteristics to obtain a distribution quality evaluation value of the film, including: evaluating and obtaining a first thickness evaluation value of each thickness distribution region in a first type of region and a second thickness evaluation value of each thickness distribution region in a second type of region by respectively evaluating the thickness difference of each thickness distribution region in the first type of region and the second type of region; determining the weight of each thickness distribution region based on the position of each thickness distribution region, and calculating the distribution quality evaluation value of the film based on the weight, the first thickness evaluation value, the second thickness evaluation value and the second thickness evaluation value.

[0030] Specifically, for each thickness distribution region in the first type of region and the second type of region, the thickness difference is evaluated, and the corresponding evaluation value is obtained to quantify the difference degree of the thickness inside each region, which provides a basis for subsequent distribution evaluation; based on the position and characteristics of each thickness distribution region, the weight of each region is determined, and the weight is determined according to the importance of the region in the film to reflect the contribution degree of different regions to the overall film quality; the distribution quality evaluation value of the film is calculated according to the weight, the first thickness evaluation value and the second thickness evaluation value, and the thickness difference, the position weight and the evaluation value of each region are comprehensively considered to obtain the evaluation of the overall film thickness distribution quality. This step can quantitatively evaluate the quality of the film thickness distribution by evaluating the thickness difference and calculating the evaluation value, which provides a basis for further improvement and control; considering the position and importance of different regions in the film, the weight can more accurately reflect the influence of each region on the overall film quality; considering the thickness difference, the position weight and the evaluation value comprehensively, the distribution quality evaluation value can comprehensively reflect the quality status of the film thickness distribution, which provides guidance for the optimization of the production process; by quantitatively evaluating the distribution quality of the film, more data support can be provided for decision-making in the production process, which helps to optimize the production process and improve product quality.

[0031] In the embodiments of the present application, a film processing optimization control method is provided, and the calculation formula of the distribution quality evaluation value of the film is: P= , Wherein, P is the distribution quality evaluation value of the film, a is the weight of each thickness distribution region, xi is the first thickness evaluation value of the i-th thickness distribution region in the first type of region, n is the number of thickness distribution regions in the first type of region, yi is the second thickness evaluation value of the i-th thickness distribution region in the second type of region, and m is the number of thickness distribution regions in the second type of region.

[0032] In the embodiments of the present application, a film processing optimization control method is provided, which determines an optimization coefficient of film thickness processing based on a distribution quality evaluation value of the film, and includes the following steps: presetting a preset optimization coefficient-distribution quality evaluation value interval correspondence relationship, wherein each distribution quality evaluation value interval is associated with a corresponding preset optimization coefficient; obtaining the distribution quality evaluation value of the film, and selecting the preset optimization coefficient corresponding to the distribution quality evaluation value interval as the optimization coefficient of the film thickness processing based on the mapping relationship of the distribution quality evaluation value interval of the distribution quality evaluation value in the preset optimization coefficient-distribution quality evaluation value interval correspondence relationship.

[0033] Specifically, for different distribution evaluation value intervals, corresponding optimization coefficients are preset; the distribution evaluation value of the film is calculated, and the distribution evaluation value interval to which the value belongs is determined; the distribution evaluation value interval is mapped to the preset optimization coefficient-distribution evaluation value interval correspondence relationship, and the preset optimization coefficient corresponding to the interval is selected as the optimization coefficient of the film thickness processing. This step can automatically select the appropriate optimization coefficient according to the distribution evaluation value of the film through the preset optimization coefficient-distribution evaluation value interval correspondence relationship, without manual intervention; different distribution evaluation value intervals can correspond to different optimization coefficients, realizing personalized film thickness processing optimization and adjusting parameters according to actual conditions; selecting the preset optimization coefficient corresponding to the distribution evaluation value interval can more accurately guide the film thickness processing process, improving production efficiency and product quality; dynamically adjusting the optimization parameter according to the real-time distribution evaluation value realizes real-time monitoring and adjustment of the production process, ensuring the stability and consistency of the film thickness processing.

[0034] In the embodiments of the present application, a film processing optimization control method is provided, which determines an optimization coefficient of film thickness processing based on a distribution quality evaluation value of the film, and includes the following steps: presetting a preset optimization coefficient-distribution quality evaluation value interval correspondence relationship, wherein each distribution quality evaluation value interval is associated with a corresponding preset optimization coefficient; obtaining the distribution quality evaluation value of the film, and selecting the preset optimization coefficient corresponding to the distribution quality evaluation value interval as the optimization coefficient of the film thickness processing based on the mapping relationship of the distribution quality evaluation value interval of the distribution quality evaluation value in the preset optimization coefficient-distribution quality evaluation value interval correspondence relationship.

[0035] Specifically, during the film production process, the processing data of the film is obtained in real time through sensors, monitoring devices, etc.; the preset optimization coefficient is multiplied by the real-time obtained processing parameters to obtain the processing data after optimization adjustment, and the processing parameters can be adjusted according to the real-time situation to achieve better film processing effect; the processing data after optimization adjustment is applied to the film processing process to control the production equipment to ensure that the film processing meets the optimization requirements. This step realizes real-time optimization adjustment of the processing parameters by multiplying the optimization coefficient and the real-time processing parameters, improves the flexibility and adaptability of the production process; the processing data after optimization adjustment can more accurately guide the film processing process, improve the stability and consistency of product quality; real-time data acquisition and optimization calculation can realize partially automated production control, reduce human intervention, and improve production efficiency; by optimizing and adjusting the processing data, resources can be more effectively utilized, and waste in the production process can be reduced, thereby improving production efficiency and reducing costs; through real-time optimization adjustment, problems can be found and adjusted in time to ensure that the film processing quality meets the requirements, and the competitiveness and market acceptance of the product are improved.

[0036] As shown in Figure 2 In the embodiments of the present application, a film processing optimization control system is provided, comprising: an acquisition module for acquiring historical processing data of the film processing process and analyzing the historical processing data to determine key processing parameters affecting the thickness of the film; a prediction module for extracting features of the key processing parameters and constructing a thickness prediction model based on the features and a preset neural network model to predict, to obtain thickness prediction data of the film; an analysis module for constructing a thickness distribution map of the film based on the thickness prediction data of the film and analyzing the thickness distribution map to determine the thickness distribution area of the film and the corresponding thickness features; an evaluation module for evaluating the thickness distribution of the film based on the thickness distribution area and the thickness features, obtaining a distribution quality evaluation value of the film, and determining an optimization coefficient of the film thickness processing based on the distribution quality evaluation value; a control module for optimizing and adjusting the real-time processing data of the film based on the optimization coefficient, and controlling the processing of the film according to the processing data after optimization adjustment.

[0037] In summary, the embodiment of the present application provides a film processing optimization control method and system, which comprises: obtaining and analyzing historical processing data of a film processing process to determine key processing parameters affecting film thickness; extracting features of the key processing parameters, and constructing a thickness prediction model based on the features and a preset model to obtain thickness prediction data through prediction; constructing a thickness distribution map of the film based on the thickness prediction data, and determining thickness distribution regions and thickness features therein; evaluating the thickness distribution of the film based on the thickness distribution regions and the thickness features to obtain a distribution quality evaluation value of the film, and determining an optimization coefficient based thereon; and optimizing and adjusting real-time processing data of the film based on the optimization coefficient to control the processing of the film. The present application determines the current film distribution through prediction, analyzes and determines the optimization coefficient to optimize the current processing process, solves the problems of uneven film thickness and large fluctuations, reduces the waste rate, and improves the production efficiency and quality.

[0038] Finally, it should be noted that: obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

[0039] The above only describes one embodiment of the present application, but cannot limit the scope of the present application. Any structural changes made according to the present application, as long as the essence of the present application is not lost, should be considered to fall within the protection scope of the present application. Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the platform and the related description described above can refer to the corresponding process in the foregoing platform embodiment, which will not be described here.

[0040] The term "comprising" or any other similar word is intended to cover non-exclusive inclusion, so that the process, platform, article or device / platform including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or includes the elements inherent to these processes, platforms, articles or devices / platforms.

[0041] So far, the technical solutions of the present application have been described in combination with the further embodiments shown in the drawings, but those skilled in the art can easily understand that the protection scope of the present application is obviously not limited to these specific embodiments. Those skilled in the art can make equivalent changes or replacements to the related technical features without deviating from the principles of the present application, and the technical solutions after these changes or replacements will fall within the protection scope of the present application.

[0042] The above only describes the preferred embodiments of the present application, and is not intended to limit the protection scope of the present application.

Claims

1. A method for optimizing control of thin film processing, characterized by, The application relates to a method for evaluating and optimizing the thickness distribution of a film. The method comprises the following steps: acquiring historical processing data of a film processing process, and analyzing the historical processing data to determine key processing parameters affecting the thickness of the film; extracting features of the key processing parameters, and constructing a thickness prediction model based on the features and a preset neural network model to perform prediction, thereby obtaining thickness prediction data of the film; constructing a thickness distribution map of the film based on the thickness prediction data of the film, and analyzing the thickness distribution map to determine a thickness distribution region of the film and corresponding thickness features; evaluating the thickness distribution of the film based on the thickness distribution region and the thickness features, thereby obtaining a distribution quality evaluation value of the film, and determining an optimization coefficient of the film thickness processing based on the distribution quality evaluation value; 2. The method of claim 1, wherein the method further comprises: optimizing and adjusting real-time processing data of the film based on the optimization coefficient, and controlling the processing of the film according to the optimized and adjusted processing data. The method of acquiring historical processing data of a film processing process and analyzing the historical processing data to determine key processing parameters affecting the thickness of the film comprises the following steps: dividing the historical processing data into a plurality of processing parameter data groups according to parameter types, and analyzing the correlation degree of each processing parameter data group and thickness data; selecting processing parameter data groups with a correlation degree greater than a preset value as candidate processing parameters affecting the thickness of the film in the historical processing data; for each candidate processing parameter, filtering thickness data of the film when other candidate processing parameters are the same, and determining a change amount of the thickness data of the film; 3. The method of claim 2, wherein the control signal is a function of the difference between the measured value and the target value. selecting a candidate processing parameter with a change amount greater than a preset value as a key processing parameter affecting the thickness of the film in the historical processing data. The method of extracting features of the key processing parameters, and constructing a thickness prediction model based on the features and a preset neural network model to perform prediction, thereby obtaining thickness prediction data of the film, comprises the following steps: extracting features of the key processing parameters and constructing a data set of the key processing parameters, and inputting the data set into a preset neural network model to construct a thickness prediction initial model of the film; dividing the data set into a training set and a test set according to a preset proportion, and inputting the training set and the test set into the thickness prediction initial model of the film; training and testing the thickness prediction initial model of the film until the thickness prediction initial model of the film meets a preset convergence condition, thereby obtaining a thickness prediction model of the film; 4. The method of claim 3, wherein the step of determining the set of process parameters comprises the step of: determining the set of process parameters based on the set of process parameters and the set of process parameters. acquiring real-time processing parameters of the film, and inputting the real-time processing parameters into the thickness prediction model of the film to perform prediction, thereby obtaining thickness prediction data of the film. The method of constructing a thickness distribution map of the film based on the thickness prediction data of the film, and analyzing the thickness distribution map to determine a thickness distribution region of the film and corresponding thickness features, comprises the following steps: determining a transverse thickness and a longitudinal thickness of the film from the thickness prediction data of the film, and constructing a thickness distribution map of the film based on the transverse thickness and the longitudinal thickness of the film; acquiring a preset required thickness of the film, and determining a first type region exceeding the required thickness of the film and a second type region not exceeding the required thickness of the film in the thickness distribution map; performing clustering analysis on the thickness value of each position in the first type region and the second type region, and determining corresponding thickness distribution regions in the first type region and the second type region according to the clustering analysis result; respectively, the thickness average value of all positions in each thickness distribution region in the first type region and the second type region is calculated, and the thickness difference of each thickness distribution region is obtained, and the position of each thickness distribution region in the first type region and the second type region is determined; the thickness difference and the position of each thickness distribution region are determined as the thickness feature corresponding to the thickness distribution region of the film.

5. A method of thin film process optimization control according to claim 4, wherein, The clustering analysis is performed on the thickness value of each position in the first type region and the second type region, and the corresponding thickness distribution region in the first type region and the second type region is determined according to the result of the clustering analysis, and the method comprises the following steps: A data set is established according to the thickness value of each position in the first type region and the second type region, and k initial clustering centers of the data set are randomly selected; The Euclidean distance of the thickness value in the data set to the initial clustering center is calculated, and each position is divided into a corresponding clustering cluster according to the Euclidean distance of the thickness value in the data set to the initial clustering center; The thickness average value in each clustering cluster is calculated, and the clustering center is re-determined according to the thickness average value in each clustering cluster; The above steps are repeated iteratively until the clustering center no longer changes or the iteration number reaches a preset iteration threshold, and k clustering clusters are obtained, wherein the clustering cluster represents a thickness distribution region; The thickness average value of each thickness distribution region is calculated, and the thickness distribution region with a positive thickness average value is classified into the first type region, and the thickness distribution region with a negative thickness average value is classified into the second type region, and finally the corresponding thickness distribution region in the first type region and the second type region is obtained.

6. The method of claim 4, wherein the step of determining the set of process parameters is performed by a neural network. The thickness distribution of the film is evaluated based on the thickness distribution region and the thickness feature, and the distribution quality evaluation value of the film is obtained, and the method comprises the following steps: The thickness difference of each thickness distribution region in the first type region and the second type region is evaluated and valued respectively, and the first thickness evaluation value of each thickness distribution region in the first type region and the second thickness evaluation value of each thickness distribution region in the second type region are obtained; The weight of each thickness distribution region is determined based on the position of each thickness distribution region, and the distribution quality evaluation value of the film is calculated based on the weight, the first thickness evaluation value, the second thickness evaluation value and the second thickness evaluation value.

7. A method of thin film process optimization control according to claim 6, wherein The calculation formula of the distribution quality evaluation value of the film is: P= ， wherein, P is the distribution quality evaluation value of the film, α is the weight of each thickness distribution region, xi is the first thickness evaluation value of the i-th thickness distribution region in the first type region, n is the number of thickness distribution regions in the first type region, yi is the second thickness evaluation value of the i-th thickness distribution region in the second type region, and m is the number of thickness distribution regions in the second type region.

8. The method of claim 6, wherein the step of determining the set of process parameters is performed by a neural network. The optimization coefficient of the film thickness processing is determined based on the distribution quality evaluation value of the film, and the method comprises the following steps: A preset optimization coefficient-distribution quality evaluation value interval corresponding relationship is set in advance, and for each distribution quality evaluation value interval, a corresponding preset optimization coefficient is associated. The distribution quality evaluation value of the film is obtained, and a preset optimization coefficient corresponding to a distribution quality evaluation value interval to which the distribution quality evaluation value belongs is selected as an optimization coefficient for film thickness processing based on a mapping relationship in a preset optimization coefficient-distribution quality evaluation value interval corresponding relationship.

9. A method of thin film process optimization control according to claim 8, wherein, The real-time processing data of the film is optimized and adjusted based on the optimization coefficient, and the processing of the film is controlled according to the processing data after the optimization and adjustment, which comprises: The real-time processing data of the film is obtained, the optimization coefficient is multiplied by the real-time processing parameter to obtain the processing data after the optimization and adjustment, and the processing of the film is controlled according to the data after the optimization and adjustment.

10. A thin film process optimization control system, characterized by, It comprises: The acquisition module is used to obtain the historical processing data of the film processing process, analyze the historical processing data, and determine the key processing parameters affecting the film thickness; The prediction module is used to extract the features of the key processing parameters, and construct a thickness prediction model based on the features and a preset neural network model to predict the thickness prediction data of the film; The analysis module is used to construct the thickness distribution map of the film based on the thickness prediction data of the film, analyze the thickness distribution map, determine the thickness distribution area of the film and the corresponding thickness features, and analyze the thickness distribution map; The evaluation module is used to evaluate the thickness distribution of the film based on the thickness distribution area and the thickness features, obtain the distribution quality evaluation value of the film, and determine the optimization coefficient for the film thickness processing based on the distribution quality evaluation value; The control module is used to optimize and adjust the real-time processing data of the film based on the optimization coefficient, and control the processing of the film according to the processing data after the optimization and adjustment.

Images