Intelligent control system and method for preparing graphene conductive film

Abstract

The invention relates to the field of intelligent control, and particularly discloses an intelligent control system and method for preparing a graphene conductive film. An artificial intelligence technology based on the deep learning field is used for monitoring a monitoring video of spraying in a preset time period, and spraying distance values of a plurality of preset time points in the preset time period are calculated. And performing feature extraction and coding on the concentration values of the graphene suspension at the plurality of preset time points in the preset time period and the electric field intensity values at the plurality of preset time points in the preset time period so as to obtain a classification result that the electric field intensity at the current time point should be increased or decreased. Therefore, the uniformity of the graphene conductive film is improved by controlling the intensity of the electric field in real time.

Description

Technical Field

[0001] The present application relates to the field of intelligent control, and more specifically to an intelligent control system and method for preparing graphene conductive films. Background Art

[0002] Graphene is a single-layer, two-dimensional crystalline material composed of carbon atoms, possessing a unique structure and exceptional properties. It is composed of hexagonal units of six carbon atoms, forming a flat, honeycomb structure. Graphene also boasts extremely high electron mobility and conductivity, making it one of the best known conductive materials. Electrons in the lattice move extremely quickly, virtually unaffected by scattering and resistance. Graphene's high conductivity makes it suitable for the production of graphene conductive films. Graphene conductive films are thin films with excellent electrical conductivity and are widely used in electronic devices, solar cells, sensors, and other fields.

[0003] A common method for preparing graphene conductive films is electrostatic spray deposition (Electrospray Deposition). Electrostatic spray deposition utilizes an electric field to spray graphene nanosheets from a liquid onto a substrate to form a thin film. While this method offers advantages such as ease of preparation, high controllability, high efficiency, and excellent uniformity, the use of electrostatic spray deposition in preparing graphene conductive films can be challenging due to the difficulty in controlling the spray parameters, resulting in poor uniformity in the resulting graphene conductive film.

[0004] Therefore, an optimized intelligent control system for the preparation of graphene conductive films is expected. Summary of the Invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiments of the present application provide an intelligent control system and method for the preparation of graphene conductive film, which uses artificial intelligence technology based on the field of deep learning to extract and encode the monitoring video of the spray within a predetermined time period, the spray distance value at multiple predetermined time points within the predetermined time period, the concentration value of the graphene suspension at multiple predetermined time points within the predetermined time period, and the electric field strength value at multiple predetermined time points within the predetermined time period, so as to obtain a classification result of whether the electric field strength at the current time point should be increased or decreased. In this way, by controlling the magnitude of the electric field strength in real time, the uniformity of the graphene conductive film is improved.

[0006] According to one aspect of the present application, an intelligent control system for preparing a graphene conductive film is provided, comprising:

[0007] a spray data acquisition module, configured to acquire monitoring video of the spray within a predetermined time period, spray distance values at a plurality of predetermined time points within the predetermined time period, concentration values of the graphene suspension at a plurality of predetermined time points within the predetermined time period, and electric field intensity values at a plurality of predetermined time points within the predetermined time period;

[0008] a spray monitoring feature extraction module, configured to process the monitoring video of the spray within the predetermined time period to obtain a spray monitoring feature graph;

[0009] an electric field state feature extraction module, configured to process the spray distance values at a plurality of predetermined time points within the predetermined time period, the concentration values of the graphene suspension at a plurality of predetermined time points within the predetermined time period, and the electric field intensity values at a plurality of predetermined time points within the predetermined time period, to obtain an electric field state feature map;

[0010] The electric field intensity control result generation module is used to analyze the spray monitoring characteristic diagram and the electric field state characteristic diagram to obtain the result that the electric field intensity should be increased or decreased at the current time point.

[0011] According to another aspect of the present application, there is also provided an intelligent control method for preparing a graphene conductive film, comprising:

[0012] Obtaining a monitoring video of the spray within a predetermined time period, spray distance values at multiple predetermined time points within the predetermined time period, concentration values of the graphene suspension at multiple predetermined time points within the predetermined time period, and electric field intensity values at multiple predetermined time points within the predetermined time period;

[0013] Processing the monitoring video of the spray within the predetermined time period to obtain a spray monitoring characteristic graph;

[0014] Processing the spray distance values at a plurality of predetermined time points within the predetermined time period, the concentration values of the graphene suspension at a plurality of predetermined time points within the predetermined time period, and the electric field intensity values at a plurality of predetermined time points within the predetermined time period to obtain an electric field state characteristic diagram;

[0015] The spray monitoring characteristic diagram and the electric field state characteristic diagram are analyzed to obtain a result indicating whether the electric field intensity should be increased or decreased at the current time point.

[0016] In summary, the intelligent control system and method for preparing a graphene conductive film provided by this application utilizes artificial intelligence technology based on deep learning to extract and encode features from surveillance video of spraying within a predetermined time period, spray distance values at multiple predetermined time points within the predetermined time period, graphene suspension concentration values at multiple predetermined time points within the predetermined time period, and electric field intensity values at multiple predetermined time points within the predetermined time period, thereby obtaining a classification result indicating whether the electric field intensity should be increased or decreased at the current time point. This improves the uniformity of the graphene conductive film by controlling the electric field intensity in real time. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without inventive efforts. In the drawings, the same reference numerals generally represent the same components or steps.

[0018] Figure 1 4 is a block diagram of an intelligent control system for preparing graphene conductive film according to an embodiment of the present application.

[0019] Figure 2 4 is a block diagram of a spray monitoring feature extraction module in an intelligent control system for preparing graphene conductive films according to an embodiment of the present application.

[0020] Figure 3 4 is a block diagram of an electric field state feature extraction module in an intelligent control system for preparing graphene conductive films according to an embodiment of the present application.

[0021] Figure 4 Flowchart of an intelligent control method for preparing a graphene conductive film according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] Below, in order to enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of this application.

[0023] Figure 1 FIG is a block diagram of an intelligent control system for preparing a graphene conductive film according to an embodiment of the present application. Figure 1As shown, the intelligent control system 100 for preparing graphene conductive film according to the embodiment of the present application includes: a spray data acquisition module 110, which is used to obtain the monitoring video of the spray within a predetermined time period, the spray distance value at multiple predetermined time points within the predetermined time period, the concentration value of the graphene suspension at multiple predetermined time points within the predetermined time period, and the electric field strength value at multiple predetermined time points within the predetermined time period; a spray monitoring feature extraction module 120, which is used to process the monitoring video of the spray within the predetermined time period to obtain a spray monitoring characteristic graph; an electric field state feature extraction module 130, which is used to process the spray distance value at multiple predetermined time points within the predetermined time period, the concentration value of the graphene suspension at multiple predetermined time points within the predetermined time period, and the electric field strength value at multiple predetermined time points within the predetermined time period to obtain an electric field state characteristic graph; and an electric field strength control result generation module 140, which is used to analyze the spray monitoring characteristic graph and the electric field state characteristic graph to obtain a result that the electric field strength at the current time point should be increased or decreased.

[0024] In the above-mentioned intelligent control system 100 for preparing graphene conductive film, the spray data acquisition module 110 is used to obtain the monitoring video of the spray in a predetermined time period, the spray distance value at a plurality of predetermined time points in the predetermined time period, the concentration value of the graphene suspension at a plurality of predetermined time points in the predetermined time period, and the electric field strength value at a plurality of predetermined time points in the predetermined time period. As described in the above-mentioned background technology, electrostatic spraying (ElectrosprayDeposition) is to utilize the electric field effect to spray the graphene nanosheet layer in the liquid onto a substrate to form a film. Electrostatic spraying has the characteristics of being easy to prepare, highly controllable, highly efficient, and having good uniformity. However, when using electrostatic spraying to prepare graphene conductive film, there is also the problem that the spray parameters are difficult to control, resulting in the poor uniformity of the final graphene conductive film. Therefore, it is desirable to have an intelligent control system for the preparation of an optimized graphene conductive film.

[0025] To address the above technical issues, an intelligent control system for graphene conductive film production has been proposed. This system uses deep learning-based artificial intelligence to extract and encode features from surveillance video of the spray within a predetermined time period, the spray distance values at multiple predetermined time points within the predetermined time period, the concentration values of the graphene suspension at multiple predetermined time points within the predetermined time period, and the electric field strength values at multiple predetermined time points within the predetermined time period. This system then categorizes whether the electric field strength at the current time point should be increased or decreased. This system improves the uniformity of the graphene conductive film by controlling the electric field strength in real time.

[0026] In the electrostatic spraying method, graphene is first dispersed in a solution to form a suspension of graphene nanosheets. Then, a high-voltage electric field is applied to spray the graphene nanosheets in the suspension onto a substrate. During the spraying process, the graphene nanosheets in the solution gradually dry and deposit on the substrate, forming a conductive graphene film. The electric field strength is controlled by adjusting the voltage of the high-voltage power supply. A higher electric field strength increases the spray velocity and kinetic energy of the spray particles, helping to form thinner conductive graphene films. However, excessively high electric field strength can lead to unstable spraying and particle aggregation, affecting film uniformity and quality. The spray distance refers to the distance between the nozzle and the substrate. A longer spray distance increases the time the spray particles spend in the air, thereby increasing their kinetic energy. When the spray particles enter the electric field, their greater kinetic energy makes it easier for the particles to overcome the resistance of the electric field, resulting in a faster deposition rate on the substrate surface. Therefore, a longer spray distance results in a higher electric field strength. The solution concentration refers to the graphene content in the graphene suspension. A higher solution concentration means more graphene particles are present in the spray, which increases the number of spray particles. When the spray particles enter the electric field, the higher particle count leads to more particles deposited on the substrate surface, increasing the thickness and uniformity of the film. Therefore, a higher solution concentration can lead to a higher electric field strength. By monitoring the video, the morphology and uniformity of the graphene conductive film can be evaluated.

[0027] Specifically, in the technical solution of the present application, a camera device or a monitoring system is used to record the spraying conditions during the preparation process of the graphene conductive film. For example, a high-speed camera or other suitable equipment is used to capture the details of the spray. A laser rangefinder or other distance measuring device can be used to measure the distance between the sprayer and the target surface. Within a predetermined time period, the spray distance values at multiple time points are recorded. Using an appropriate concentration measuring instrument or method, the concentration of the graphene suspension is measured. Within a predetermined time period, the concentration values at multiple time points are recorded. Using an electric field strength measuring instrument or method, the electric field strength during the preparation process is measured. Within a predetermined time period, the electric field strength values at multiple time points are recorded.

[0028] Currently, deep learning and neural networks are widely used in fields such as computer vision, natural language processing, and speech signal processing. Furthermore, deep learning and neural networks have demonstrated capabilities approaching or even surpassing those of humans in areas such as image classification, object detection, semantic segmentation, and text translation.

[0029] In recent years, the development of deep learning and neural networks has provided new solutions and options for intelligent control systems used in the preparation of graphene conductive films.

[0030] Specifically, first, a monitoring video of the spray within a predetermined time period, spray distance values at multiple predetermined time points within the predetermined time period, concentration values of the graphene suspension at multiple predetermined time points within the predetermined time period, and electric field strength values at multiple predetermined time points within the predetermined time period are obtained.

[0031] In the above-mentioned intelligent control system 100 for the preparation of graphene conductive film, the spray monitoring feature extraction module 120 is used to process the monitoring video of the spray within the predetermined time period to obtain a spray monitoring feature map. By processing the video, dynamic changes in the spray process can be captured, such as spray shape, distribution, motion trajectory, etc. This dynamic information is crucial for understanding the evolution of the spray and analyzing the spray state. The extracted spray feature vector can be fused and analyzed with other key data, such as spray distance, suspension concentration, and electric field strength. By comprehensively analyzing the correlation between different features, the current spray state and trend can be judged more accurately, and adjustment suggestions for the electric field strength can be generated.

[0032] More specifically, to extract key features relevant to the spray process from the video, the spray monitoring video within a predetermined time period is processed through a spray feature extraction module based on a temporal attention mechanism model to generate a spray monitoring feature map. This helps better understand and represent the dynamic changes of the spray. The spray monitoring video contains time series information about the spray process. Using this temporal attention mechanism model, each frame in the video is weighted, giving higher attention and weight to features in key frames or key time periods. This approach allows for focus on the most important moments or key frames in the spray process, while ignoring irrelevant or redundant information, thereby improving the effectiveness and accuracy of feature extraction. The spray monitoring feature map generated by the spray feature extraction module can be viewed as an abstract representation of the spray process, capturing key characteristics of the spray process, such as spray shape, diffusion, and uniformity. These feature maps can be used as input for subsequent processing, fused with other feature vectors, and associated encoding to further improve the precision and accuracy of electric field intensity control.

[0033] Specifically, in an embodiment of the present application, the spray monitoring feature extraction module is used to pass the monitoring video of the spray within the predetermined time period through the spray feature extraction module based on the time attention mechanism model to obtain a spray monitoring feature map.

[0034] Figure 2 FIG. 1 is a block diagram of a spray monitoring feature extraction module in an intelligent control system for preparing a graphene conductive film according to an embodiment of the present application. Figure 2As shown, the spray monitoring feature extraction module 120 includes: a sampling unit 121, which is used to extract multiple spray monitoring key frames from the monitoring video of the spray within the predetermined time period at a predetermined sampling frequency; an adjacent frame extraction unit 122, which is used to extract the adjacent first spray monitoring key frame and the second spray monitoring key frame from the multiple spray monitoring key frames; a first convolution encoding unit 123, which is used to pass the first spray monitoring key frame and the second spray monitoring key frame through the first convolution layer and the second convolution layer of the spray feature extraction module respectively to obtain a first convolution feature map and a second convolution feature map; a time attention subunit 124, which is used to for calculating the position-wise multiplication between the first convolution feature map and the second convolution feature map to obtain a temporal attention map; an attention activation subunit 125 for inputting the temporal attention map into a Softmax activation function to obtain a temporal attention feature map; a second convolution encoding subunit 126 for passing the second spray monitoring key frame through the third convolution layer of the spray feature extraction module to obtain a third convolution feature map; and an attention application subunit 127 for calculating the position-wise multiplication between the third convolution feature map and the temporal attention feature map to obtain a temporal attention feature map corresponding to the second spray monitoring key frame.

[0035] In the above-mentioned intelligent control system 100 for preparing graphene conductive film, the electric field state feature extraction module 130 is used to process the spray distance values at multiple predetermined time points within the predetermined time period, the concentration values of the graphene suspension at multiple predetermined time points within the predetermined time period, and the electric field intensity values at multiple predetermined time points within the predetermined time period to obtain an electric field state characteristic diagram. Spray distance, suspension concentration, and electric field intensity are important parameters in the preparation process of graphene conductive film. By processing and fusing the numerical values of these parameters, the influence of different parameters on the conductive film preparation process can be comprehensively considered to obtain a more comprehensive electric field state feature. At the same time, there may be a certain correlation between different parameters, such as the influence of spray distance and suspension concentration on electric field intensity. By analyzing and fusing the numerical values of these parameters, the correlation between different parameters can be revealed, thereby better understanding the characteristics and changing laws of the electric field state.

[0036] More specifically, in order to integrate the characteristic information of the time series into a vector representation, the spray distance values, graphene suspension concentration values, and electric field strength values at multiple predetermined time points within a predetermined time period are arranged according to the time dimension as spray distance input vectors, suspension concentration input vectors, and electric field strength input vectors, respectively, for subsequent feature fusion and model training. During the preparation process of graphene conductive film, parameters such as spray distance, suspension concentration, and electric field strength change over time. By arranging these parameters as input vectors according to the time dimension, their temporal relationship, that is, the order of parameter values at different time points, can be preserved. By arranging the spray distance values, suspension concentration values, and electric field strength values as input vectors, the characteristic information of the time series can be converted into a fixed-length vector representation. The advantage of this is that the changing patterns and trends of the time series can be encoded into the vector, allowing the model to better capture the relevant features in the time dimension.

[0037] Figure 3 FIG. 1 is a block diagram of an electric field state feature extraction module in an intelligent control system for preparing a graphene conductive film according to an embodiment of the present application. Figure 3 As shown, the electric field state feature extraction module 130 includes: an input vector construction unit 131, which is used to arrange the spray distance values at multiple predetermined time points within the predetermined time period, the concentration values of the graphene suspension at multiple predetermined time points within the predetermined time period, and the electric field strength values at multiple predetermined time points within the predetermined time period into a spray distance input vector, a suspension concentration input vector, and an electric field strength input vector according to the time dimension; a spray distance feature extraction unit 132, which is used to pass the spray distance input vector through a spray distance feature extractor including multiple fully connected layers and one-dimensional convolution layers to obtain a spray distance feature vector; a suspension concentration feature extraction unit 133, used to pass the suspension concentration input vector through a suspension concentration feature extractor comprising multiple fully connected layers and one-dimensional convolution layers to obtain a suspension concentration feature vector; an electric field strength feature extraction unit 134, used to pass the electric field strength input vector through an electric field strength feature extractor comprising multiple fully connected layers and one-dimensional convolution layers to obtain an electric field strength feature vector; and an electric field state fusion unit 135, used to fuse the spray distance feature vector, the suspension concentration feature vector and the electric field strength feature vector to obtain an electric field state fusion feature matrix.

[0038] More specifically, to extract key features from the spray distance input vector, the spray distance input vector is transformed into a more expressive and discriminative spray distance feature vector through a spray distance feature extractor consisting of multiple fully connected layers and one-dimensional convolutional layers. Spray distance is a critical parameter in the preparation of graphene conductive films, impacting film uniformity and performance. The spray distance feature extractor extracts effective features related to the preparation process and film quality from the spray distance input vector. The combination of multiple fully connected layers and one-dimensional convolutional layers achieves nonlinear transformation and feature extraction of the input vector. The fully connected layers learn complex relationships and combinational features in the input vector, while the one-dimensional convolutional layers capture local features and patterns in the input vector. By stacking and combining these layers, higher-level features are gradually extracted, resulting in a more expressive spray distance feature vector. The design and training process of the spray distance feature extractor can be adjusted and optimized based on the specific task and data. By appropriately configuring the network structure and parameters, the performance and effectiveness of the spray distance feature extractor can be improved, enabling it to better capture the correlation between spray distance and electric field intensity.

[0039] Specifically, in an embodiment of the present application, the spray distance feature extraction unit 132 includes: a fully connected encoding unit, which is used to encode the spray distance input vector using the fully connected layer of the spray distance feature extractor to extract high-dimensional implicit features of the eigenvalues of each position in the spray distance input vector; and a one-dimensional convolutional encoding unit, which is used to encode the spray distance input vector using the one-dimensional convolutional layer of the spray distance feature extractor to extract high-dimensional implicit correlation features of the correlation between the eigenvalues of each position in the spray distance input vector.

[0040] More specifically, in order to extract the key features in the suspension concentration input vector, the suspension concentration input vector is converted into a more expressive and discriminative suspension concentration feature vector through a suspension concentration feature extractor comprising multiple fully connected layers and one-dimensional convolutional layers. Through the suspension concentration feature extractor, effective features related to the preparation process and membrane quality can be extracted from the suspension concentration input vector. In order to extract the key features in the electric field strength input vector, the electric field strength input vector is converted into a more expressive and discriminative electric field strength feature vector through an electric field strength feature extractor comprising multiple fully connected layers and one-dimensional convolutional layers. Specifically, the processing method for the suspension concentration input vector and the electric field strength input vector refers to the processing process for the spray distance input vector.

[0041] More specifically, to integrate these diverse feature information, the spray distance feature vector, suspension concentration feature vector, and electric field intensity feature vector are fused to generate a fused electric field state feature matrix, forming a more comprehensive and integrated representation of the electric field state during graphene film fabrication. Spray distance, suspension concentration, and electric field intensity are important parameters influencing the graphene film fabrication process, and they exhibit complex and nonlinear relationships. By fusing the feature vectors of these parameters, we can comprehensively consider their interactions and combined effects, yielding a more comprehensive representation of the electric field state. The fused feature matrix can be generated in various ways, such as by concatenating feature vectors column-wise to form a matrix or by weighted fusion of feature vectors using a feature fusion network. The resulting fused electric field state feature matrix encompasses information from different features and exhibits enhanced expressiveness and discriminability. Using the fused electric field state feature matrix, we can more comprehensively describe and represent the electric field state during graphene film fabrication. This helps extract richer and more accurate feature information, thereby optimizing the fabrication process and improving the performance and quality of graphene films.

[0042] Specifically, in an embodiment of the present application, the electric field state fusion unit 135 includes: an electric field state fusion subunit, which is used to arrange the spray distance feature vector, the suspension concentration feature vector and the electric field intensity feature vector in two dimensions to obtain an electric field state fusion feature matrix; and an electric field feature filtering subunit, which is used to pass the electric field state fusion feature matrix through an electric field feature filtering module based on a convolutional neural network model to obtain an electric field state feature map.

[0043] Specifically, to further extract and filter key features related to the electric field state, the electric field state fusion feature matrix is passed through an electric field feature filtering module based on a convolutional neural network model to generate an electric field state feature map. This reduces redundant information and noise, resulting in a more discriminative and expressive electric field state feature map. The electric field state fusion feature matrix contains comprehensive information from multiple features, but may contain some redundant and irrelevant features, making it ineffective for discriminating and expressing the electric field state. The electric field feature filtering module based on a convolutional neural network model utilizes convolution operations and nonlinear activation functions to extract and emphasize important features in the electric field state feature matrix. Convolutional neural network models have powerful feature extraction and representation capabilities in image and sequence data processing. By using convolutional layers and activation functions in the electric field feature filtering module, local patterns and spatial relationships in the electric field state feature matrix can be captured, thereby extracting more discriminative features. The electric field state feature map obtained after processing by the electric field feature filtering module better represents key information about the electric field state while reducing the influence of irrelevant and redundant features. Such feature vectors can be better used in the classification, prediction and optimization of electric field states, and improve the control accuracy and effect of the graphene conductive film preparation process.

[0044] Specifically, in an embodiment of the present application, the electric field feature filtering subunit is used to: use each layer of the electric field feature filtering module to perform convolution processing and nonlinear activation processing on the input data in the forward pass of the layer so as to output the electric field state feature map by the last layer of the electric field feature filtering module, wherein the input of the first layer of the electric field feature filtering module is the electric field state fusion feature matrix.

[0045] In the above-mentioned intelligent control system 100 for preparing graphene conductive film, the electric field intensity control result generation module 140 is used to analyze the spray monitoring characteristic diagram and the electric field state characteristic diagram to obtain the result of whether the electric field intensity should be increased or decreased at the current time point. By analyzing the spray monitoring characteristic diagram and the electric field state fusion characteristic matrix, the correlation between the spray state and the electric field state can be revealed. For example, a specific spray shape, density, and distribution may require a specific electric field intensity to achieve the best effect. By analyzing these correlations, the appropriate electric field intensity adjustment direction under the current spray state can be determined.

[0046] Specifically, in an embodiment of the present application, the electric field strength control result generation module 140 includes: a spray feature fusion unit, used to associate and encode the electric field state characteristic diagram and the spray monitoring characteristic diagram to obtain a spray comprehensive characteristic diagram; and an electric field strength control result generation unit, used to pass the spray comprehensive characteristic diagram through a classifier to obtain a classification result, and the classification result is used to indicate whether the electric field strength at the current time point should increase or decrease.

[0047] More specifically, to combine information from both the electric field state and spray monitoring, the electric field state feature map and the spray monitoring feature map are associatively encoded to produce a comprehensive spray feature map. This creates a more comprehensive and integrated feature representation, enabling a better description and understanding of the spray characteristics during graphene film fabrication. The electric field state feature map contains characteristic information related to the electric field state, while the spray monitoring feature map contains characteristic information related to the spray process. Both types of information are important for the graphene film fabrication process and have certain correlations and influences on each other. By associatively encoding the electric field state feature map and the spray monitoring feature map, their information can be fused to form a comprehensive feature matrix. This comprehensive spray feature map can more comprehensively describe and represent the spray characteristics during graphene film fabrication, including the interaction and combined effects of the electric field state and the spray process. Associative encoding can be implemented in various ways, such as concatenating the electric field state feature map and the spray monitoring feature map column by column to form a matrix, or performing a weighted fusion of them using a feature fusion network. The resulting comprehensive spray feature map can contain information from different features and has higher expressive power and discriminability. The comprehensive spray profile allows for a more comprehensive understanding and analysis of the spray characteristics during graphene film fabrication. This helps extract richer and more accurate characteristic information, enabling better optimization of the fabrication process and improved performance and quality of the graphene film.

[0048] In particular, in the technical solution of the present application, the difference in dimensionality and scale between the characteristic manifolds of the electric field state characteristic map and the spray monitoring characteristic map in the high-dimensional feature space unit is taken into account, mainly because the objects and features described by the two are essentially different. The electric field state characteristic map is obtained based on the electric field intensity values at multiple predetermined time points within a predetermined time period, and it describes the spatial distribution of the electric field intensity. The electric field intensity values are usually continuous and smooth, so the manifold in the feature space may be continuous and relatively flat. At the same time, the dimensionality of the electric field intensity values is usually low, because the distribution of the electric field intensity can usually be represented by a small number of characteristic dimensions. The spray monitoring characteristic map is obtained from the spray monitoring video, and it describes the changes in the spray in time and space. The morphology and movement of the spray are usually more complex and diverse, so the manifold in the feature space may be more complex, diverse and nonlinear. Due to the differences in dimensionality and scale between the electric field state characteristic map and the spray monitoring characteristic map, some technical problems may arise when attempting to fuse them. First, the dimensionality difference may lead to information loss during the feature fusion process. Without proper processing measures, the lower-dimensional electric field state characteristic map may not be able to fully capture the complexity and diversity in the spray monitoring characteristic map, resulting in local structural collapse of the fused spray comprehensive characteristic map. Secondly, scale differences may lead to pathological alignment problems in the feature fusion process. When the numerical ranges of different features differ greatly, alignment difficulties may arise between different features during the fusion process, making the fusion results inaccurate or unreliable. In order to solve these problems, the present application performs a high-dimensional space unit manifold sub-dimensional hyperconvex correlation measurement on the electric field state characteristic map and the spray monitoring characteristic map to obtain a spray comprehensive characteristic map.

[0049] Specifically, in the embodiment of the present application, the spray feature fusion unit includes: a position mean subunit, which is used to calculate the position mean feature map between the electric field state feature map and the spray monitoring feature map; a position difference subunit, which is used to calculate the position difference feature map between the electric field state feature map and the spray monitoring feature map; a differential center-of-difference feature calculation subunit, which is used to respectively calculate the differential feature map between the electric field state feature map and the spray monitoring feature map and the position mean feature map to obtain a first differential center-of-difference feature map and a second differential center-of-difference feature map; a logarithmic differential center-of-difference calculation subunit, which is used to calculate the The logarithmic function value with base 2 of the characteristic value of each position in the first differential off-center characteristic map and the second differential off-center characteristic map is used to obtain the first logarithmic differential off-center characteristic map and the second logarithmic differential off-center characteristic map; a characteristic correction subunit is used to divide the first logarithmic differential off-center characteristic map by the position-based sum characteristic map between the first logarithmic differential off-center characteristic map and the second logarithmic differential off-center characteristic map to obtain a correction characteristic map; and a position-weighted sum subunit is used to calculate the position-weighted sum between the correction characteristic map and the logarithmic position-based differential characteristic map of the position-based differential characteristic map to obtain the spray comprehensive characteristic map.

[0050] That is, considering that there are differences in dimension and scale between the characteristic manifolds of the electric field state characteristic map and the spray monitoring characteristic map in the high-dimensional feature space unit, in the process of fusing the electric field state characteristic map and the spray monitoring characteristic map, technical problems such as local structural collapse or pathological alignment may occur in the fused electric field state characteristic map due to the differences in dimension and scale.

[0051] In response to the above technical problems, in the technical solution of the present application, the electric field state characteristic graph and the spray monitoring characteristic graph are subjected to a high-dimensional space unit manifold sub-dimension hyperconvex correlation measurement, which uses the mean characteristic graph of the electric field state characteristic graph and the spray comprehensive characteristic graph as the pseudo-clustering center of the characteristic manifold, and constructs a hyperconvex correlation measurement function of the characteristic manifold based on the characteristic graph, so that the eigenvalue of each position between the characteristic graphs can maintain consistency with the pseudo-clustering center of the characteristic manifold in its sub-dimension, thereby realizing the hyperconvex correlation matching of the characteristic manifold of the characteristic graph, which effectively measures the similarity and difference of the characteristic manifolds between the feature graphs, enhances the hyperconvex correlation of the characteristic manifold of the feature graph, and improves the robustness and accuracy of the characteristic manifold of the feature graph.

[0052] More specifically, the spray's comprehensive characteristic map is passed through a classifier to obtain a classification result, which is then used to indicate whether the electric field strength at the current time point should be increased or decreased. The comprehensive information from the spray characteristics is used to predict and guide the adjustment of the electric field strength, thereby achieving intelligent control of the electric field strength during the graphene conductive film preparation process. A classifier is a machine learning model that can classify samples into different categories based on the input feature vector. In this case, the spray's comprehensive characteristic map is used as input, and after processing by the classifier, a classification result is obtained, which is used to indicate the decision of whether the electric field strength should be increased or decreased at the current time point.

[0053] Specifically, in the embodiment of the present application, the electric field intensity control result generating unit is used to: use the classifier to process the spray comprehensive feature map using the following classification formula to obtain the classification result; wherein, the classification formula is: O = softmax{(W c ,B c )|Project(F)}, wherein Project(F) represents the projection of the spray comprehensive feature map into a vector, W c is the weight matrix, B c represents a bias vector, softmax represents a normalized exponential function, and O represents the classification result.

[0054] In summary, the intelligent control system for preparing a graphene conductive film according to the embodiments of the present application has been described. It uses artificial intelligence technology based on deep learning to extract and encode features from monitoring video of the spray within a predetermined time period, spray distance values at multiple predetermined time points within the predetermined time period, concentration values of the graphene suspension at multiple predetermined time points within the predetermined time period, and electric field strength values at multiple predetermined time points within the predetermined time period, thereby obtaining a classification result indicating whether the electric field strength should be increased or decreased at the current time point. In this way, by controlling the magnitude of the electric field strength in real time, the uniformity of the graphene conductive film is improved.

[0055] As described above, the intelligent control system 100 for preparing a graphene conductive film according to an embodiment of the present application can be implemented in various terminal devices, such as an intelligent control server for preparing a graphene conductive film. In one example, the intelligent control system 100 for preparing a graphene conductive film according to an embodiment of the present application can be integrated into a terminal device as a software module and / or a hardware module. For example, the intelligent control system 100 for preparing a graphene conductive film can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the intelligent control system 100 for preparing a graphene conductive film can also be one of the many hardware modules of the terminal device.

[0056] Alternatively, in another example, the intelligent control system 100 for preparing graphene conductive film and the terminal device may also be separate devices, and the intelligent control system 100 for preparing graphene conductive film may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0057] Based on the same inventive concept, an embodiment of the present application also provides an intelligent control method for preparing a graphene conductive film, which can be used to implement the system described in the above embodiment, as described in the following embodiment.

[0058] Figure 4 FIG. 1 is a flow chart of an intelligent control method for preparing a graphene conductive film according to an embodiment of the present application. Figure 4 As shown, the intelligent control method for preparing graphene conductive film according to the embodiment of the present application includes the steps of: S110, obtaining the monitoring video of the spray within a predetermined time period, the spray distance values at multiple predetermined time points within the predetermined time period, the concentration values of the graphene suspension at multiple predetermined time points within the predetermined time period, and the electric field strength values at multiple predetermined time points within the predetermined time period; S120, processing the monitoring video of the spray within the predetermined time period to obtain a spray monitoring characteristic diagram; S130, processing the spray distance values at multiple predetermined time points within the predetermined time period, the concentration values of the graphene suspension at multiple predetermined time points within the predetermined time period, and the electric field strength values at multiple predetermined time points within the predetermined time period to obtain an electric field state characteristic diagram; and S140, analyzing the spray monitoring characteristic diagram and the electric field state characteristic diagram to obtain a result that the electric field strength at the current time point should be increased or decreased.

[0059] Here, those skilled in the art will appreciate that the specific operations of each step in the above-mentioned intelligent control method for preparing a graphene conductive film have been described in detail above. Figures 1 to 3 The intelligent control system for graphene conductive film preparation has been described in detail, and therefore, its repeated description will be omitted.

[0060] In summary, the intelligent control method for preparing a graphene conductive film according to the embodiments of the present application has been described. It uses artificial intelligence technology based on deep learning to extract and encode features from monitoring video of the spray within a predetermined time period, spray distance values at multiple predetermined time points within the predetermined time period, concentration values of the graphene suspension at multiple predetermined time points within the predetermined time period, and electric field strength values at multiple predetermined time points within the predetermined time period, to obtain a classification result indicating whether the electric field strength should be increased or decreased at the current time point. In this way, by controlling the magnitude of the electric field strength in real time, the uniformity of the graphene conductive film is improved.

[0061] References to "first," "second," "third," "fourth," and the like (if any) herein are intended to distinguish similar objects and are not necessarily intended to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions, e.g., a process, method, or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such processes, methods, or apparatus.

[0062] It should be noted that the descriptions of "first", "second", etc. in this application are for descriptive purposes only and should not be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" or "second" may explicitly or implicitly include at least one of such features. In addition, the technical solutions between the various embodiments can be combined with each other, but this must be based on the fact that they can be implemented by ordinary technicians in this field. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0063] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

[0064] The block diagrams of the devices, devices, equipment, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As will be appreciated by those skilled in the art, these devices, devices, equipment, and systems can be connected, arranged, or configured in any manner. Words such as "include," "comprise," "have," and the like are open-ended words, meaning "including but not limited to," and can be used interchangeably therewith. The words "or" and "and" used herein refer to the words "and / or" and can be used interchangeably therewith, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to," and can be used interchangeably therewith.

[0065] It should also be noted that in the apparatus, device, and method of the present application, each component or each step can be decomposed and / or recombined, and such decomposition and / or recombination should be regarded as equivalent solutions of the present application.

[0066] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0067] The above description has been provided for the purpose of illustration and description. Furthermore, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although a number of example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. An intelligent control system for preparing graphene conductive film, characterized in that: include: a spray data acquisition module, configured to acquire monitoring video of the spray within a predetermined time period, spray distance values at a plurality of predetermined time points within the predetermined time period, concentration values of the graphene suspension at a plurality of predetermined time points within the predetermined time period, and electric field intensity values at a plurality of predetermined time points within the predetermined time period; a spray monitoring feature extraction module, configured to process the monitoring video of the spray within the predetermined time period to obtain a spray monitoring feature graph; an electric field state feature extraction module, configured to process the spray distance values at a plurality of predetermined time points within the predetermined time period, the concentration values of the graphene suspension at a plurality of predetermined time points within the predetermined time period, and the electric field intensity values at a plurality of predetermined time points within the predetermined time period, to obtain an electric field state feature map; The electric field intensity control result generation module is used to analyze the spray monitoring characteristic diagram and the electric field state characteristic diagram to obtain the result that the electric field intensity should be increased or decreased at the current time point.

2. The intelligent control system for preparing graphene conductive film according to claim 1, characterized in that: The spray monitoring feature extraction module is used to: The monitoring video of the spray within the predetermined time period is passed through a spray feature extraction module based on a temporal attention mechanism model to obtain a spray monitoring feature map.

3. The intelligent control system for preparing graphene conductive film according to claim 2, characterized in that: The spray monitoring feature extraction module includes: A sampling unit, configured to extract a plurality of spray monitoring key frames from the monitoring video of the spray within the predetermined time period at a predetermined sampling frequency; An adjacent frame extraction unit, configured to extract an adjacent first spray monitoring key frame and a second spray monitoring key frame from the plurality of spray monitoring key frames; a first convolutional encoding unit, configured to pass the first spray monitoring key frame and the second spray monitoring key frame through the first convolutional layer and the second convolutional layer of the spray feature extraction module respectively to obtain a first convolutional feature map and a second convolutional feature map; a temporal attention unit, configured to calculate a temporal attention map by computing a position-wise multiplication between the first convolutional feature map and the second convolutional feature map; an attention activation unit, configured to input the temporal attention map into a Softmax activation function to obtain a temporal attention feature map; a second convolutional encoding unit, configured to pass the second spray monitoring key frame through a third convolutional layer of the spray feature extraction module to obtain a third convolutional feature map; An attention applying unit is used to calculate the position point multiplication between the third convolution feature map and the temporal attention feature map to obtain the temporal attention feature map corresponding to the second spray monitoring key frame.

4. The intelligent control system for preparing graphene conductive film according to claim 3, characterized in that: The electric field state feature extraction module includes: an input vector construction unit, configured to arrange the spray distance values at a plurality of predetermined time points within the predetermined time period, the concentration values of the graphene suspension at a plurality of predetermined time points within the predetermined time period, and the electric field intensity values at a plurality of predetermined time points within the predetermined time period into a spray distance input vector, a suspension concentration input vector, and an electric field intensity input vector, respectively, according to a time dimension; a spray distance feature extraction unit, configured to pass the spray distance input vector through a spray distance feature extractor comprising a plurality of fully connected layers and a one-dimensional convolutional layer to obtain a spray distance feature vector; a suspension concentration feature extraction unit, configured to pass the suspension concentration input vector through a suspension concentration feature extractor comprising a plurality of fully connected layers and a one-dimensional convolutional layer to obtain a suspension concentration feature vector; an electric field strength feature extraction unit, configured to pass the electric field strength input vector through an electric field strength feature extractor comprising a plurality of fully connected layers and a one-dimensional convolutional layer to obtain an electric field strength feature vector; The electric field state fusion unit is used to fuse the spray distance characteristic vector, the suspension concentration characteristic vector and the electric field intensity characteristic vector to obtain an electric field state characteristic map.

5. The intelligent control system for preparing graphene conductive film according to claim 4, characterized in that: The spray distance feature extraction unit comprises: a fully connected encoding unit, configured to encode the spray distance input vector using the fully connected layer of the spray distance feature extractor to extract high-dimensional implicit features of the eigenvalues at each position in the spray distance input vector; A one-dimensional convolutional encoding unit is used to encode the spray distance input vector using the one-dimensional convolutional layer of the spray distance feature extractor to extract high-dimensional implicit correlation features of the correlation between the feature values of each position in the spray distance input vector.

6. The intelligent control system for preparing graphene conductive film according to claim 5, characterized in that: The electric field state fusion unit includes: An electric field state fusion subunit, configured to perform a two-dimensional arrangement on the spray distance feature vector, the suspension concentration feature vector, and the electric field intensity feature vector to obtain an electric field state fusion feature matrix; The electric field feature filtering subunit is used to pass the electric field state fusion feature matrix through an electric field feature filtering module based on a convolutional neural network model to obtain an electric field state feature map.

7. The intelligent control system for preparing graphene conductive film according to claim 6, characterized in that: The electric field intensity control result generating module includes: A spray feature fusion unit, configured to perform high-dimensional space unit manifold sub-dimension hyperconvex correlation measurement on the electric field state feature map and the spray monitoring feature map to obtain a spray comprehensive feature map; The electric field intensity control result generating unit is used to pass the spray comprehensive characteristic diagram through a classifier to obtain a classification result, and the classification result is used to indicate whether the electric field intensity at the current time point should be increased or decreased.

8. The intelligent control system for preparing graphene conductive film according to claim 7, characterized in that: The spray feature fusion unit comprises: A position mean subunit is used to calculate a position mean characteristic diagram between the electric field state characteristic diagram and the spray monitoring characteristic diagram; A position difference subunit, for calculating a position difference characteristic diagram between the electric field state characteristic diagram and the spray monitoring characteristic diagram; a differential off-center feature calculation subunit, configured to respectively calculate the electric field state feature map and the differential feature map between the spray monitoring feature map and the position mean feature map to obtain a first differential off-center feature map and a second differential off-center feature map; a logarithmic difference offset center calculation subunit, configured to calculate a logarithmic function value with base 2 of the eigenvalues at each position in the first differential offset center feature map and the second differential offset center feature map to obtain a first logarithmic difference offset center feature map and a second logarithmic difference offset center feature map; a feature correction subunit, configured to divide the first logarithmic difference off-center feature map by a position-based sum feature map between the first logarithmic difference off-center feature map and the second logarithmic difference off-center feature map to obtain a correction feature map; The position-weighted sum subunit is used to calculate the position-weighted sum between the correction characteristic map and the logarithmic position-differential characteristic map of the position-differential characteristic map to obtain the spray comprehensive characteristic map.

9. An intelligent control method for preparing a graphene conductive film, characterized in that: include: Obtaining a monitoring video of the spray within a predetermined time period, spray distance values at multiple predetermined time points within the predetermined time period, concentration values of the graphene suspension at multiple predetermined time points within the predetermined time period, and electric field strength values at multiple predetermined time points within the predetermined time period; Processing the monitoring video of the spray within the predetermined time period to obtain a spray monitoring characteristic graph; Processing the spray distance values at a plurality of predetermined time points within the predetermined time period, the concentration values of the graphene suspension at a plurality of predetermined time points within the predetermined time period, and the electric field intensity values at a plurality of predetermined time points within the predetermined time period to obtain an electric field state characteristic diagram; The spray monitoring characteristic diagram and the electric field state characteristic diagram are analyzed to obtain a result indicating whether the electric field intensity should be increased or decreased at the current time point.

10. The intelligent control method for preparing a graphene conductive film according to claim 9, characterized in that: Processing the monitoring video of the spray within the predetermined time period to obtain a spray monitoring characteristic graph includes: The monitoring video of the spray within the predetermined time period is passed through a spray feature extraction module based on a temporal attention mechanism model to obtain a spray monitoring feature map.

Images