Production management system and method of oil refining catalyst
Through deep learning technology, the image and filtration speed of the refining catalyst are analyzed, and whether the filtration speed needs to be adjusted is determined, which solves the problem of blockage of the filtration device in the production of traditional refining catalysts, and improves production efficiency and product quality.
Patent Information
- Application Number
- CN202410926275.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-11
- Publication Date
- 2024-12-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
During the filtration process of traditional oil refining catalysts, due to the simple structure of the filter device, it is easy to cause the filter mesh to be blocked, reduce the filtration efficiency, and affect the production efficiency and product quality.
The image of the refining catalyst to be filtered is collected by the camera and the filtering speed value of the sensor is collected. The semantic correlation feature vector and filtering speed feature vector are extracted using deep learning technology to determine whether the filtration speed needs to be adjusted to ensure the stability and efficiency of the production process.
It improves the stability and efficiency of the production process of refining catalysts, improves product quality, reduces production costs, and avoids the problems of filter clogging and reduced filtration efficiency.
Smart Images

Figure CN119066507A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of production management, and more specifically, to a production management system and method for oil refining catalysts. Background Art
[0002] Oil refining is a process that distills crude oil or other oils without changing their molecular structure. In this process, refining catalysts play a key role. They are substances used to accelerate or control chemical reactions. These catalysts are usually composed of materials such as metals, metal oxides or zeolites. According to their functions, they can be divided into types such as hydrogenation catalysts, isomerization catalysts, cracking catalysts and reforming catalysts.
[0003] Refining catalysts reduce the activation energy of the reaction by providing a high-energy surface, thereby increasing the reaction rate and selectivity, and optimizing the distribution and quality of the products in the refining process. The production of traditional refining catalysts requires process steps such as crushing, filtering, stirring and drying. However, during the filtration process, due to the simple structure of the filtration device, when too much product to be filtered is poured in at one time, it is easy to cause clogging of the filter screen, reducing the filtration effect and thus reducing the filtration efficiency.
[0004] Therefore, a production management system and method for oil refining catalysts are desired. Summary of the invention
[0005] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a production management system and method for oil refining catalysts, which first obtains a filtering image of the oil refining catalyst to be filtered collected by a camera and a filtering speed value of the oil refining catalyst to be filtered at multiple predetermined time points collected by a sensor, and then uses deep learning technology to perform feature extraction and correlation analysis on the two, and finally uses a classifier to determine whether the filtering speed of the oil refining catalyst needs to be adjusted to ensure the stability and efficiency of the oil refining catalyst production process, thereby improving production efficiency and product quality.
[0006] According to one aspect of the present application, a production management system for oil refining catalyst is provided, comprising:
[0007] A refinery catalyst production management data acquisition module, used to acquire a filter image of the refinery catalyst to be filtered acquired by a camera and filter speed values of the refinery catalyst to be filtered at a plurality of predetermined time points acquired by a sensor;
[0008] A refinery catalyst production management data extraction module, used for extracting a refinery catalyst filter image block semantic association feature vector and a refinery catalyst filter speed feature vector from the refinery catalyst filter image to be filtered collected by the camera and the refinery catalyst filter speed values to be filtered at a plurality of predetermined time points collected by the sensor;
[0009] The oil refining catalyst filtering speed adjustment judgment module is used to judge whether the oil refining catalyst filtering speed needs to be adjusted based on the oil refining catalyst filtering image block semantic association feature vector and the oil refining catalyst filtering speed feature vector.
[0010] According to another aspect of the present application, a production management method for a refining catalyst is provided, comprising:
[0011] Acquire a filtering image of the oil refining catalyst to be filtered collected by a camera and filtering speed values of the oil refining catalyst to be filtered at a plurality of predetermined time points collected by a sensor;
[0012] Extracting a semantic association feature vector of a filter image block of a refinery catalyst and a filter speed feature vector of a refinery catalyst from the filter image of the refinery catalyst to be filtered collected by the camera and the filter speed values of the refinery catalyst to be filtered at a plurality of predetermined time points collected by the sensor;
[0013] Based on the semantic association feature vector of the oil refining catalyst filtering image block and the oil refining catalyst filtering speed feature vector, it is determined whether the oil refining catalyst filtering speed needs to be adjusted.
[0014] Compared with the prior art, the present application provides a production management system and method for a refining catalyst, which first obtains a filtration image of the refining catalyst to be filtered collected by a camera and a filtration speed value of the refining catalyst to be filtered at multiple predetermined time points collected by a sensor, and then uses deep learning technology to perform feature extraction and correlation analysis on the two, and finally uses a classifier to determine whether the filtration speed of the refining catalyst needs to be adjusted to ensure the stability and efficiency of the refining catalyst production process, thereby improving production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.
[0016] Figure 1 It is a block diagram of a production management system for oil refining catalysts according to an embodiment of the present application.
[0017] Figure 2 The present invention is a block diagram of a refinery catalyst production management data extraction module in a refinery catalyst production management system according to an embodiment of the present application.
[0018] Figure 3The present invention is a block diagram of a refinery catalyst filtering speed adjustment judgment module in a refinery catalyst production management system according to an embodiment of the present application.
[0019] Figure 4 The present invention is a flowchart of a production management method for a refinery catalyst according to an embodiment of the present application.
[0020] Figure 5 is a block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION
[0021] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.
[0022] Exemplary Systems
[0023] Figure 1 Schematic diagram of the production management system of the oil refining catalyst of the embodiment of the present application. Figure 1 As shown, the production management system 100 of the refining catalyst according to the embodiment of the present application includes: a refining catalyst production management data acquisition module 110, which is used to obtain the refining catalyst filtering image to be filtered collected by the camera and the refining catalyst filtering speed values to be filtered at multiple predetermined time points collected by the sensor; a refining catalyst production management data extraction module 120, which is used to extract the refining catalyst filtering image block semantic association feature vector and the refining catalyst filtering speed feature vector from the refining catalyst filtering image to be filtered collected by the camera and the refining catalyst filtering speed values to be filtered at multiple predetermined time points collected by the sensor; a refining catalyst filtering speed adjustment judgment module 130, which is used to judge whether the refining catalyst filtering speed needs to be adjusted based on the refining catalyst filtering image block semantic association feature vector and the refining catalyst filtering speed feature vector.
[0024] In the above-mentioned production management system 100 of the oil refining catalyst, the oil refining catalyst production management data acquisition module 110 is used to obtain the filtering image of the oil refining catalyst to be filtered collected by the camera and the filtering speed value of the oil refining catalyst to be filtered at multiple predetermined time points collected by the sensor. It should be understood that oil refining is a process for distilling crude oil or other oils without changing their molecular structure. In this process, oil refining catalysts play a vital role. They are a class of substances used to accelerate or control chemical reactions. These catalysts are usually composed of materials such as metals, metal oxides or zeolites, and can be divided into different types such as hydrogenation catalysts, isomerization catalysts, cracking catalysts and reforming catalysts according to their functions. The role of the oil refining catalyst is to reduce the activation energy of the reaction by providing a high-energy surface, thereby increasing the reaction rate and selectivity, and optimizing the distribution and quality of the product in the oil refining process. The production of traditional oil refining catalysts requires multiple process steps such as crushing, filtering, stirring and drying. However, during the filtering process, due to the simple structure of the filtering device, when too much product to be filtered is poured in at one time, it is easy to cause the filter screen to be blocked, reducing the filtering effect, thereby reducing the filtering efficiency. In the technical solution of the present application, images of the refining catalyst to be filtered are collected by a camera, and the filtration speed values at multiple predetermined time points are collected by a sensor. These data are analyzed and processed using deep learning technology to determine whether the filtration speed of the refining catalyst needs to be adjusted, thereby improving production efficiency and product quality.
[0025] Specifically, the filtered images collected by the camera provide intuitive visual information, which can show the real-time status and characteristics of the refining catalyst production process in the filtration process, such as particle size, distribution, etc. These images can characterize the quality and filtration effect of the refining catalyst. The filtration speed value collected by the sensor provides the key parameter in the filtration process of the refining catalyst, namely, speed information. By collecting speed values at multiple predetermined time points, the changes in the filtration speed of the refining catalyst at different stages can be obtained, thereby revealing the dynamic changes and trends in the production process. Taking into account the filtered images collected by the camera and the speed values collected by the sensor, comprehensive monitoring and analysis of the refining catalyst production process can be achieved to more accurately evaluate the filtration status of the catalyst, timely discover potential problems and take corresponding measures, thereby providing a rich information basis, which helps to optimize production efficiency, improve product quality and reduce production costs.
[0026] In the above-mentioned production management system 100 of the oil refining catalyst, the oil refining catalyst production management data extraction module 120 is used to extract the semantic association feature vector of the oil refining catalyst filtering image block and the oil refining catalyst filtering speed feature vector from the oil refining catalyst filtering image to be filtered collected by the camera and the oil refining catalyst filtering speed values to be filtered at multiple predetermined time points collected by the sensor. It should be understood that by extracting the image block semantic association feature vector, the system can convert the image information into a vector representation with semantic association. And by converting the speed value into a feature vector, the system can better understand the movement characteristics and changes of the oil refining catalyst during the filtering process. By comprehensively utilizing the oil refining catalyst filtering image block semantic association feature vector and the oil refining catalyst filtering speed feature vector, the system can realize comprehensive monitoring and analysis of the oil refining catalyst production process. In this way, the system can more accurately evaluate the filtering state and performance of the oil refining catalyst, and realize refined management and optimization of the production process.
[0027] Figure 2 FIG. 1 is a block diagram of a refinery catalyst production management data extraction module in a refinery catalyst production management system according to an embodiment of the present application. Figure 2 As shown, in a specific embodiment of the present application, the refinery catalyst production management data extraction module 120 includes: a catalyst filter image preprocessing unit 121, used for performing data preprocessing on the refinery catalyst filter image to be filtered collected by the camera to obtain multiple refinery catalyst filter image block feature vectors; a filter image block feature encoding unit 122, used for feature encoding the multiple refinery catalyst filter image block feature vectors to obtain the refinery catalyst filter image block semantic association feature vector; a filter speed value feature encoding unit 123, used for feature encoding the refinery catalyst filter speed values to be filtered at multiple predetermined time points collected by the sensor to obtain the refinery catalyst filter speed feature vector.
[0028] It should be understood that data preprocessing may include steps such as image denoising, graying, edge detection, and image segmentation to eliminate interference and noise in the image and highlight key features. By segmenting the filtered image of the refining catalyst to be filtered into multiple image blocks, local features in the image, such as the shape, size, and distribution of particles, can be captured more finely. Among them, each image block can be regarded as an independent sub-image with its own characteristics and information. By extracting features from these image blocks, the content and characteristics of the entire image can be more comprehensively described. The process of extracting feature vectors of multiple refining catalyst filter image blocks may include feature extraction for each image block, such as using a convolutional neural network to extract visual features of the image block, or applying traditional image processing techniques to extract color histograms, texture features, etc. These feature vectors may contain various information of the image block, such as shape, texture, color, etc., which is helpful for a deeper analysis and comparison of the image block. By comprehensively utilizing multiple refining catalyst filter image block feature vectors, a more detailed and comprehensive analysis of the refining catalyst filtering process can be achieved.
[0029] Furthermore, feature encoding of multiple refinery catalyst filter image block feature vectors can convert the feature information of the image block into a more semantically relevant and representative vector representation, so as to better describe and distinguish the semantic information between different image blocks. Feature encoding is to obtain higher-level abstract features by transforming and extracting the original features, which can better reflect the semantic and structural characteristics of the image block. By feature encoding the refinery catalyst filter image block feature vector, the original image block features can be converted into a more compact and representative semantically relevant feature vector. Such a feature vector can not only reduce the dimension and complexity of the data, but also better express the semantic association between image blocks, which helps the system to more accurately understand and analyze the content and structure of the image block.
[0030] Furthermore, the filtering speed values of the refining catalyst to be filtered at multiple predetermined time points collected by the sensor can be converted into a more representative and information-rich feature vector representation, so as to better describe and analyze the changing laws and characteristics of the filtering speed of the refining catalyst. For the filtering speed values of the refining catalyst to be filtered at multiple predetermined time points, the feature encoding can include pre-processing steps such as smoothing, normalization, and dimensionality reduction of the data to eliminate noise and outliers, so as to make the data more reliable and stable. Then, the extracted features can be learned and encoded by applying a machine learning algorithm or a deep learning model to convert the original speed data into a more representative and distinguishing feature vector representation. The refining catalyst filtering speed feature vector obtained in this way can better describe the changing characteristics of the refining catalyst filtering speed, including information such as the trend, periodicity, and mutation of the speed.
[0031] In a specific embodiment of the present application, the catalyst filter image preprocessing unit 121 includes: constructing the refinery catalyst filter image to be filtered collected by the camera into a refinery catalyst filter two-dimensional image sequence block; passing the refinery catalyst filter two-dimensional image sequence block through a refinery catalyst filter learnable linear embedding feature extractor to obtain the multiple refinery catalyst filter image block feature vectors.
[0032] It should be understood that constructing two-dimensional image sequence blocks can help the system better understand and utilize the spatiotemporal information in the image data, and realize comprehensive monitoring and analysis of the refining catalyst filtration process. Among them, constructing two-dimensional image sequence blocks can divide the image data into multiple spatial blocks or regions, thereby realizing local feature extraction and analysis of different regions. This spatial block method helps the system to more comprehensively understand the detail information and structural features in the refining catalyst filtration image, and improve the system's ability to understand and recognize image content.
[0033] Furthermore, the refinery catalyst filtering two-dimensional image sequence blocks are passed through a refinery catalyst filtering learnable linear embedding feature extractor, which can use deep learning technology to extract high-level abstract feature representations from image data to achieve a deeper understanding and analysis of the refinery catalyst filtering process. Among them, the linear embedding feature extractor can automatically extract important information for distinguishing the characteristics of different refinery catalyst filtering image blocks by learning the correlation and pattern between image data, thereby achieving effective encoding and compression of image data.
[0034] In a specific embodiment of the present application, the filtering image block feature encoding unit 122 includes: passing the multiple refinery catalyst filtering image block feature vectors through a converter-based refinery catalyst filtering semantic encoder to obtain multiple refinery catalyst filtering image block semantic feature vectors; cascading the multiple refinery catalyst filtering image block semantic feature vectors to obtain the refinery catalyst filtering image block semantic association feature vector.
[0035] It should be understood that the converter-based refinery catalyst filter semantic encoder can utilize techniques such as the self-attention mechanism to interact and integrate multiple refinery catalyst filter image block feature vectors, thereby capturing the semantic associations and contextual information between image blocks. Such an encoder can convert image block features into a more semantically informative representation, which helps the system to more deeply understand the meaning and association of image content. Through the converter-based refinery catalyst filter semantic encoder, semantic modeling and encoding of refinery catalyst filter image blocks can be achieved, and image features can be converted into semantic feature vectors. The semantic feature vectors obtained in this way can better describe the semantic meaning and information of the image block, provide the system with a richer and more abstract image representation, and help to achieve a deeper understanding and analysis of image content.
[0036] Furthermore, cascading multiple refinery catalyst filter image block semantic feature vectors can effectively integrate the semantic information of each image block into a unified feature representation. Doing so helps to comprehensively consider the semantic associations and interactions between different image blocks, thereby more comprehensively capturing the semantic association features between image blocks and improving the ability to characterize the overall image block. Specifically, the converter-based Bert model of the converter-based refinery catalyst filter semantic encoder is used to perform global context semantic encoding on each refinery catalyst filter image block feature vector in the multiple refinery catalyst filter image block feature vectors to obtain multiple feature vectors; and, the multiple feature vectors are cascaded to obtain the multiple refinery catalyst filter image block semantic feature vectors.
[0037] In a specific embodiment of the present application, the filtering speed value feature encoding unit 123 includes: arranging the filtering speed values of the refinery catalyst to be filtered at multiple predetermined time points collected by the sensor into a refinery catalyst filtering speed input vector; passing the refinery catalyst filtering speed input vector through a refinery catalyst filtering speed convolutional neural network as a feature encoder to obtain the refinery catalyst filtering speed feature vector.
[0038] It should be understood that arranging the filtering speed values of the refining catalyst to be filtered at multiple predetermined time points as input vectors can realize the integration and organization of speed data. Such a vector form can organically combine the speed values at different time points to form a continuous numerical sequence, which better reflects the changing trend and law of the speed value.
[0039] Furthermore, through the refinery catalyst filtering speed convolutional neural network, feature extraction and abstract representation of speed data can be achieved. Among them, the convolutional neural network has a strong feature learning ability, which can automatically learn the spatial and temporal features in the data, thereby converting the original speed data into a higher-level feature representation. Such a feature extraction process helps to reduce the dimension of the data, remove noise and redundant information, and extract key features in the data. By converting the speed input vector into a speed feature vector through the convolutional neural network, the hierarchical representation and abstract modeling of the speed data can be achieved. The multi-layer structure of the neural network can gradually extract different levels of features of the data, from low-level to high-level abstract representation, making the representation of the data richer and more meaningful. Such a feature vector can better capture the complex patterns and laws in the speed data. Specifically, each layer of the refinery catalyst filtering speed convolutional neural network used as a feature encoder performs convolution processing, pooling processing and nonlinear activation processing on the input data in the forward transfer of the layer to output the refinery catalyst filtering speed feature vector by the last layer of the refinery catalyst filtering speed convolutional neural network as a feature encoder, wherein the input of the refinery catalyst filtering speed convolutional neural network as a feature encoder is the refinery catalyst filtering speed input vector.
[0040] In the production management system 100 of the above-mentioned oil refining catalyst, the oil refining catalyst filtering speed adjustment judgment module 130 is used to judge whether the oil refining catalyst filtering speed needs to be adjusted based on the semantic association feature vector of the oil refining catalyst filtering image block and the oil refining catalyst filtering speed feature vector. It should be understood that the visual information about the filter state and the distribution of the oil refining catalyst can be provided based on the semantic association feature vector of the oil refining catalyst filtering image block. These feature vectors can capture key features in the image block, such as particle distribution, blockage, etc., and reflect the actual working state of the oil refining catalyst filter. Combined with the oil refining catalyst filtering speed feature vector, the oil refining catalyst filtering state can be evaluated from the perspective of speed change. Among them, the speed feature vector can reflect the changing trend and law of speed data, and provide important clues about whether the filtering speed is normal. By analyzing the speed feature vector, it can be found that the speed fluctuates abnormally or continues to decline, and then it can be judged whether there are problems such as filter blockage and oil refining catalyst loss, so as to decide whether speed adjustment is needed. In this way, by comprehensively considering the information of the semantic association feature vector of the refinery catalyst filter image block and the refinery catalyst filter speed feature vector, the refinery catalyst filter status can be evaluated more comprehensively, and potential problems can be discovered in time, providing timely decision-making references for operators.
[0041] Figure 3 FIG. 1 is a block diagram of a filter speed adjustment judgment module for a refinery catalyst in a production management system for a refinery catalyst according to an embodiment of the present application. Figure 3As shown, in a specific embodiment of the present application, the refinery catalyst filtering speed adjustment judgment module 130 includes: a filtering feature fusion unit 131, which is used to fuse the semantic association feature vector of the refinery catalyst filtering image block and the refinery catalyst filtering speed feature vector to obtain a refinery catalyst filtering speed adjustment judgment feature vector; a filtering feature optimization unit 132, which is used to perform deep feature enhancement on the refinery catalyst filtering speed adjustment judgment feature vector based on the class label probability space to obtain an optimized refinery catalyst filtering speed adjustment judgment feature vector; a filtering speed adjustment judgment unit 133, which is used to pass the optimized refinery catalyst filtering speed adjustment judgment feature vector through a classifier to obtain a classification result, and the classification result is used to judge whether the refinery catalyst filtering speed needs to be adjusted.
[0042] It should be understood that fusing the semantic association feature vector of the refinery catalyst filter image block and the refinery catalyst filter speed feature vector can comprehensively evaluate the refinery catalyst filter status from multiple dimensions to provide a more accurate and reliable basis for speed adjustment judgment. Among them, fusing the semantic association feature vector of the refinery catalyst filter image block and the filter speed feature vector can achieve the complementarity and cross-validation of multi-source information. Specifically, the image feature vector provides visual information about the state of the refinery catalyst filter, and the speed feature vector reflects the change law of the speed data. The combination of the two can comprehensively consider the refinery catalyst filter situation in the spatial and temporal dimensions, and improve the accuracy and robustness of the judgment.
[0043] Furthermore, the refinery catalyst filtration speed adjustment judgment feature vector is subjected to deep feature enhancement based on the class label probability space to obtain an optimized refinery catalyst filtration speed adjustment judgment feature vector, including: calculating the square of the absolute value of the eigenvalue at each position of the refinery catalyst filtration speed adjustment judgment feature vector and then performing a logarithmic operation to obtain a logarithmic transformation refinery catalyst filtration speed adjustment judgment feature vector; accumulating and summing all eigenvalues of the logarithmic transformation refinery catalyst filtration speed adjustment judgment feature vector to obtain the sum of the logarithmic transformation refinery catalyst filtration speed adjustment judgment feature vector; taking the absolute value of the eigenvalue at each position in the refinery catalyst filtration speed adjustment judgment feature vector through a softmax function to obtain a pseudo classification label value; multiplying the eigenvalue at each position of the refinery catalyst filtration speed adjustment judgment feature vector by the sum of the logarithmic transformation refinery catalyst filtration speed adjustment judgment feature vector and then dividing by the pseudo classification label value to obtain the optimized refinery catalyst filtration speed adjustment judgment feature vector.
[0044] In particular, in the technical solution of the present application, by fusing the semantic feature vector of the refinery catalyst filter image block and the feature vector of the refinery catalyst filter speed, a refinery catalyst filter speed adjustment judgment feature vector is obtained to judge whether the filter speed needs to be adjusted. In order to ensure the accuracy and reliability of the judgment, it is necessary to focus on improving the convergence of the refinery catalyst filter speed adjustment judgment feature vector relative to the class probability label space. Specifically, by improving the convergence of the refinery catalyst filter speed adjustment judgment feature vector relative to the class probability label space, the accuracy of the classifier in judging whether the filter speed needs to be adjusted can be increased. A better refinery catalyst filter speed adjustment judgment feature vector representation helps the classifier to better distinguish the differences between different categories and improve the accuracy of classification. The higher the convergence of the refinery catalyst filter speed adjustment judgment feature vector and the class probability label space, the lower the possibility of the classifier making errors in the judgment process. This helps to reduce the risk of misjudgment and ensure that the filter speed adjustment judgment result is more reliable. Optimizing the convergence of the feature vector helps to improve the generalization ability of the model, so that it can process new refinery catalyst filter data and achieve better classification results, rather than being limited to samples in the training data set. Improving the convergence of the refinery catalyst filtration speed adjustment judgment feature vector relative to the class probability label space can enhance the robustness of the model, making it more resistant to noise and interference, thereby better adapting to classification tasks in different situations. Based on this, the refinery catalyst filtration speed adjustment judgment feature vector is subjected to deep feature enhancement based on the class label probability space to obtain an optimized refinery catalyst filtration speed adjustment judgment feature vector.
[0045] Specifically, the refinery catalyst filtration speed adjustment judgment feature vector is subjected to deep feature enhancement based on the class label probability space to obtain an optimized refinery catalyst filtration speed adjustment judgment feature vector, including: performing deep feature enhancement based on the class label probability space on the refinery catalyst filtration speed adjustment judgment feature vector according to the following optimization formula, wherein the optimization formula is:
[0046]
[0047] Among them, V i is the eigenvalue of the i-th position of the refinery catalyst filtration speed adjustment judgment eigenvector, L is the length of the refinery catalyst filtration speed adjustment judgment eigenvector, log2 represents the logarithmic function value with base 2, softmax represents the normalized exponential function, V' i It is the eigenvalue of the ith position of the optimization oil refining catalyst filtration speed adjustment judgment eigenvector.
[0048] Here, in order to improve the convergence of the refinery catalyst filtration speed adjustment judgment feature vector relative to the predetermined class label in the class label probability space during the classification judgment process of the refinery catalyst filtration speed adjustment judgment feature vector through the classifier, in the technical solution of the present application, the refinery catalyst filtration speed adjustment judgment feature vector is subjected to deep feature enhancement based on the class label probability space, and the abstract features of the refinery catalyst filtration speed adjustment judgment feature vector at the position are represented by the logarithmic value of the absolute phase stack of the eigenvalues at each position in the refinery catalyst filtration speed adjustment judgment feature vector, and the abstract features of the eigenvalues at all positions in the refinery catalyst filtration speed adjustment judgment feature vector are aggregated by a wavelet-like function. In order to obtain a structural depth stacking factor, and in coordination with the pseudo-classification labels at each position in the refinery catalyst filtration speed adjustment judgment feature vector to imitate the label response space, the features at each position in the refinery catalyst filtration speed adjustment judgment feature vector are subjected to deep feature enhancement based on the class label probability space. In this way, the high-dimensional features of the features at each position of the refinery catalyst filtration speed adjustment judgment feature vector are approximated and mapped to the class label probability space so that each element of the refinery catalyst filtration speed adjustment judgment feature vector can approach its optimal position during the iteration process, thereby improving the convergence of the optimized refinery catalyst filtration speed adjustment judgment feature vector in the class label probability space relative to the predetermined class label.
[0049] Furthermore, by inputting the optimized feature vector into the classifier, the automatic judgment of the adjustment of the refining catalyst filtration speed can be realized. During the training process, the classifier can establish a mapping relationship between the feature vector and the speed adjustment demand by learning a large number of data samples, so that in practical applications, the corresponding classification results can be given according to the new feature vector. This machine learning-based method can more objectively and accurately evaluate the state of the refining catalyst filtration speed, avoiding the errors and uncertainties caused by subjective judgment.
[0050] In summary, the embodiment of the present application first obtains the filtering image of the refining catalyst to be filtered collected by the camera and the filtering speed values of the refining catalyst to be filtered at multiple predetermined time points collected by the sensor, and then uses deep learning technology to perform feature extraction and correlation analysis on the two. Finally, a classifier is used to determine whether the filtering speed of the refining catalyst needs to be adjusted to ensure the stability and efficiency of the refining catalyst production process, thereby improving production efficiency and product quality.
[0051] As described above, the production management system 100 of the oil refining catalyst according to the embodiment of the present application can be implemented in various terminal devices. In one example, the production management system 100 of the oil refining catalyst can be integrated into the terminal device as a software module and / or a hardware module. For example, the production management system 100 of the oil refining catalyst 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 production management system 100 of the oil refining catalyst can also be one of the many hardware modules of the terminal device.
[0052] Alternatively, in another example, the production management system 100 of the refinery catalyst and the terminal device may also be separate devices, and the production management system 100 of the refinery catalyst 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.
[0053] Exemplary Methods
[0054] Figure 4 Flow chart of the production management method of the oil refining catalyst according to the embodiment of the present application. Figure 4 As shown, according to the production management method of the refining catalyst of the embodiment of the present application, it includes: S110, obtaining the filtering image of the refining catalyst to be filtered collected by the camera and the filtering speed values of the refining catalyst to be filtered at multiple predetermined time points collected by the sensor; S120, extracting the refining catalyst filtering image block semantic association feature vector and the refining catalyst filtering speed feature vector from the refining catalyst filtering image to be filtered collected by the camera and the refining catalyst filtering speed values at multiple predetermined time points collected by the sensor; S130, judging whether the refining catalyst filtering speed needs to be adjusted based on the refining catalyst filtering image block semantic association feature vector and the refining catalyst filtering speed feature vector.
[0055] Here, those skilled in the art will appreciate that the specific operations of each step in the above-mentioned production management method of the oil refining catalyst have been described in the above reference. Figures 1 to 3 The description of the production management system of the oil refining catalyst has been described in detail, and therefore, its repeated description will be omitted.
[0056] Exemplary Electronic Devices
[0057] Below, reference Figure 5 To describe an electronic device according to an embodiment of the present application.
[0058] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned production management method of oil refining catalyst when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.
[0059] See also Figure 5 , Figure 5 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:
[0060] The processor 11 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solution provided in the embodiment of the present application; the memory 12 can be implemented in the form of a read-only memory (Read Only Memory, ROM), a static storage device, a dynamic storage device or a random access memory (Random Access Memory, RAM), etc.
[0061] The memory 12 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 12, and the processor 11 is called to execute the production management method of the oil refining catalyst in the embodiment of this application;
[0062] Input / output interface 13, used to implement information input and output;
[0063] The communication interface 14 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);
[0064] A bus 15 that transmits information between the various components of the device (e.g., the processor 11, the memory 12, the input / output interface 13, and the communication interface 14);
[0065] The processor 11 , the memory 12 , the input / output interface 13 and the communication interface 14 are connected to each other in communication within the device via a bus 15 .
[0066] An embodiment of the present application further provides a storage medium, which is a computer-readable storage medium and stores a computer program. When the computer program is executed by a processor, the production management method of the above-mentioned oil refining catalyst is implemented.
[0067] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0068] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.
[0069] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.
[0070] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0071] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.
[0072] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0073] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.
[0074] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0075] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0076] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method of each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs. The preferred embodiment of the embodiment of the present application is described above with reference to the accompanying drawings, and the scope of rights of the embodiment of the present application is not limited thereto. Any modifications, equivalent substitutions and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall be within the scope of rights of the embodiments of the present application.
Claims
1. A production management system for oil refining catalysts, characterized in that: include: A refinery catalyst production management data acquisition module, used to acquire a filter image of the refinery catalyst to be filtered acquired by a camera and filter speed values of the refinery catalyst to be filtered at a plurality of predetermined time points acquired by a sensor; A refinery catalyst production management data extraction module, used for extracting a refinery catalyst filter image block semantic association feature vector and a refinery catalyst filter speed feature vector from the refinery catalyst filter image to be filtered collected by the camera and the refinery catalyst filter speed values to be filtered at a plurality of predetermined time points collected by the sensor; The oil refining catalyst filtering speed adjustment judgment module is used to judge whether the oil refining catalyst filtering speed needs to be adjusted based on the oil refining catalyst filtering image block semantic association feature vector and the oil refining catalyst filtering speed feature vector.
2. The production management system of oil refining catalyst according to claim 1, characterized in that: The oil refining catalyst production management data extraction module includes: A catalyst filter image preprocessing unit, used for performing data preprocessing on the oil refining catalyst filter image to be filtered collected by the camera to obtain a plurality of oil refining catalyst filter image block feature vectors; A filter image block feature encoding unit, used for performing feature encoding on the plurality of refinery catalyst filter image block feature vectors to obtain a semantic association feature vector of the refinery catalyst filter image block; The filtering speed value characteristic encoding unit is used to perform characteristic encoding on the filtering speed values of the oil refining catalyst to be filtered at multiple predetermined time points collected by the sensor to obtain the filtering speed characteristic vector of the oil refining catalyst.
3. The production management system of oil refining catalyst according to claim 2, characterized in that: The catalyst filtering image preprocessing unit comprises: The oil refining catalyst filtered image to be filtered collected by the camera is constructed into an oil refining catalyst filtered two-dimensional image sequence block; The refinery catalyst filtered two-dimensional image sequence blocks are passed through a refinery catalyst filtered learnable linear embedding feature extractor to obtain the plurality of refinery catalyst filtered image block feature vectors.
4. The production management system of oil refining catalyst according to claim 3, characterized in that: The filtering image block feature encoding unit comprises: Passing the plurality of refinery catalyst filter image block feature vectors through a converter-based refinery catalyst filter semantic encoder to obtain a plurality of refinery catalyst filter image block semantic feature vectors; The multiple oil refining catalyst filter image block semantic feature vectors are cascaded to obtain the oil refining catalyst filter image block semantic association feature vector.
5. The production management system of oil refining catalyst according to claim 4, characterized in that: The filtering speed value feature encoding unit comprises: Arranging the filtration speed values of the oil refining catalyst to be filtered at a plurality of predetermined time points collected by the sensor as an oil refining catalyst filtration speed input vector; The refinery catalyst filtration rate input vector is passed through a refinery catalyst filtration rate convolutional neural network as a feature encoder to obtain the refinery catalyst filtration rate feature vector.
6. The production management system of oil refining catalyst according to claim 5, characterized in that: The refinery catalyst filtration speed input vector is passed through a refinery catalyst filtration speed convolutional neural network as a feature encoder to obtain the refinery catalyst filtration speed feature vector, including: Each layer of the refinery catalyst filter speed convolutional neural network used as a feature encoder performs convolution processing, pooling processing and nonlinear activation processing on the input data in the forward pass of the layer, so that the refinery catalyst filter speed feature vector is output by the last layer of the refinery catalyst filter speed convolutional neural network used as a feature encoder, wherein the input of the refinery catalyst filter speed convolutional neural network used as a feature encoder is the refinery catalyst filter speed input vector.
7. The production management system of oil refining catalyst according to claim 6, characterized in that: The oil refining catalyst filtering speed adjustment judgment module comprises: A filtering feature fusion unit, used for fusing the semantic association feature vector of the oil refining catalyst filtering image block and the oil refining catalyst filtering speed feature vector to obtain an oil refining catalyst filtering speed adjustment judgment feature vector; A filtering feature optimization unit, used for performing deep feature enhancement on the oil refining catalyst filtering speed adjustment judgment feature vector based on the class label probability space to obtain an optimized oil refining catalyst filtering speed adjustment judgment feature vector; The filtering speed adjustment judgment unit is used to pass the optimized refining catalyst filtering speed adjustment judgment feature vector through a classifier to obtain a classification result, and the classification result is used to judge whether the refining catalyst filtering speed needs to be adjusted.
8. The production management system of oil refining catalyst according to claim 7, characterized in that: The filtering feature optimization unit comprises: Calculating the square of the absolute value of the eigenvalue at each position of the refinery catalyst filtration speed adjustment judgment eigenvector and then performing a logarithmic operation to obtain a logarithmic transformation refinery catalyst filtration speed adjustment judgment eigenvector; Cumulatively summing all eigenvalues of the logarithmic transformation oil refining catalyst filtration speed adjustment judgment eigenvector to obtain a sum value of the logarithmic transformation oil refining catalyst filtration speed adjustment judgment eigenvector; The eigenvalues of each position in the oil refining catalyst filtration speed adjustment judgment eigenvector are passed through a softmax function and then the absolute values are taken to obtain a pseudo classification label value; The eigenvalues of each position of the refinery catalyst filtration speed adjustment judgment feature vector are multiplied by the sum of the logarithmic transformation refinery catalyst filtration speed adjustment judgment feature vectors and then divided by the pseudo classification label value to obtain the optimized refinery catalyst filtration speed adjustment judgment feature vector.
9. A production management method for oil refining catalysts, characterized in that: include: Acquire a filtering image of the oil refining catalyst to be filtered collected by a camera and filtering speed values of the oil refining catalyst to be filtered at a plurality of predetermined time points collected by a sensor; Extracting a semantic association feature vector of a filter image block of a refinery catalyst and a filter speed feature vector of a refinery catalyst from the filter image of the refinery catalyst to be filtered collected by the camera and the filter speed values of the refinery catalyst to be filtered at a plurality of predetermined time points collected by the sensor; Based on the semantic association feature vector of the oil refining catalyst filtering image block and the oil refining catalyst filtering speed feature vector, it is determined whether the oil refining catalyst filtering speed needs to be adjusted.
10. The production management method of oil refining catalyst according to claim 9, characterized in that: Extracting a semantic association feature vector of a filter image block of a refinery catalyst and a filter speed feature vector of a refinery catalyst from the filter image of the refinery catalyst to be filtered collected by the camera and the filter speed values of the refinery catalyst to be filtered at a plurality of predetermined time points collected by the sensor, including: Performing data preprocessing on the oil refining catalyst filter image to be filtered collected by the camera to obtain a plurality of oil refining catalyst filter image block feature vectors; Performing feature encoding on the plurality of refinery catalyst filter image block feature vectors to obtain a semantically associated feature vector of the refinery catalyst filter image block; The filtering speed values of the oil refining catalyst to be filtered at a plurality of predetermined time points collected by the sensor are feature-coded to obtain the filtering speed feature vector of the oil refining catalyst.
Citation Information
Patent Citations
Flow velocity control system for filling PET bottle with purified water and control method thereof
CN115082837A
Ingredient control management system and method in detergent production process
CN116187952A
Intelligent preparation method and system of modified PBT material
CN116690823A
