Environmentally friendly antioxidant production management system and method

Abstract

This application relates to the field of intelligent management, specifically disclosing a production management system and method for environmentally friendly antioxidants. The system obtains time series of consumption of multiple energy types during the production process of environmentally friendly antioxidants, including electricity, water, and steam, and uses artificial intelligence technology to perform temporal feature analysis and feature extraction on the time series of consumption of these multiple energy types, thereby intelligently identifying energy waste links. In this way, the system can identify patterns and anomalies in energy use, thereby optimizing energy efficiency during the production process to comply with the green production concept of environmentally friendly antioxidants, while also reducing unnecessary energy consumption and, consequently, production costs.

Description

Technical Field

[0001] The present application relates to the field of intelligent management, and more specifically, to an environmentally friendly antioxidant production management system and method. Background Art

[0002] Environmentally friendly antioxidants are additives used to prevent oxidation of materials, particularly during plastic manufacturing and processing. They effectively slow down the aging process of materials exposed to oxygen, light, and heat, thereby protecting the material from damage and extending its service life.

[0003] Energy usage is crucial in the production of environmentally friendly antioxidants, as it directly impacts production efficiency, costs, and the product's environmental performance. Energy waste in the production of environmentally friendly antioxidants can lead to multiple problems. First, it increases production costs, as energy is a significant expense in the production process. Second, excessive energy consumption can exacerbate environmental pollution, particularly when using fossil fuels, increasing greenhouse gas emissions and negatively impacting global climate change.

[0004] However, traditional energy management systems may lack sufficient connectivity, resulting in decentralized monitoring and a lack of centralized management, which can hinder timely detection of defects and alerts. Furthermore, most energy management systems only provide feedback on a company's energy usage and are unable to analyze and identify energy waste points, requiring in-depth analysis by professionals.

[0005] Therefore, an optimized environmentally friendly antioxidant production management system is expected. Summary of the Invention

[0006] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides an environmentally friendly antioxidant production management system and method, which obtains the time series of the consumption of multiple energy types in the production process of environmentally friendly antioxidants, wherein the energy types include electricity, water, and steam, and uses artificial intelligence technology to perform time series feature analysis and feature extraction on the time series of the consumption of the multiple energy types, so as to intelligently determine the links of energy waste. In this way, the system can identify patterns and anomalies in energy use, thereby optimizing the energy efficiency in the production process to comply with the green production concept of environmentally friendly antioxidants, while reducing unnecessary energy consumption and thus reducing production costs.

[0007] According to one aspect of the present application, there is provided an environmentally friendly antioxidant production management system, comprising:

[0008] The module for acquiring the time series of energy consumption during the production of environmentally friendly antioxidants is used to acquire the time series of consumption of various energy types during the production of environmentally friendly antioxidants, wherein the energy types include electricity, water, and steam;

[0009] an energy consumption time series feature extraction module, configured to extract energy consumption time series features after arranging the time series of the consumption of the multiple energy types into a plurality of energy consumption time series input vectors to obtain a plurality of energy consumption multi-scale time series feature vectors;

[0010] a cross-scale energy consumption feature analysis module, configured to arrange the plurality of energy consumption multi-scale time series feature vectors into an energy consumption type time series feature matrix and then perform cross-scale energy consumption feature analysis to obtain an energy consumption type multi-scale feature matrix;

[0011] An energy global feature capture module, configured to capture the energy global features of the energy consumption type multi-scale feature matrix to obtain an energy consumption type global feature matrix;

[0012] The energy waste link judgment module is used to determine the energy waste link based on the global characteristic matrix of energy consumption types.

[0013] According to another aspect of the present application, a method for managing the production of an environmentally friendly antioxidant is provided, comprising:

[0014] Obtaining a time series of consumption of various energy types during the production process of an environmentally friendly antioxidant, wherein the energy types include electricity, water, and steam;

[0015] Arranging the time series of the consumption of the multiple energy types into a plurality of energy consumption time series input vectors and then extracting energy consumption time series features to obtain a plurality of energy consumption multi-scale time series feature vectors;

[0016] Arranging the plurality of energy consumption multi-scale time series feature vectors into an energy consumption type time series feature matrix and then performing cross-scale energy consumption feature analysis to obtain an energy consumption type multi-scale feature matrix;

[0017] Capturing global energy features on the multi-scale feature matrix of energy consumption types to obtain a global feature matrix of energy consumption types;

[0018] Based on the global characteristic matrix of energy consumption types, the links of energy waste are determined.

[0019] Compared to existing technologies, the present application provides an environmentally friendly antioxidant production management system and method. This system intelligently identifies energy waste by acquiring time series of consumption of multiple energy types during the production process, including electricity, water, and steam. The time series uses artificial intelligence technology to perform temporal feature analysis and feature extraction on these energy types, thereby intelligently identifying areas of energy waste. This system can identify patterns and anomalies in energy use, thereby optimizing energy efficiency during production to comply with the green production philosophy of environmentally friendly antioxidants. It can also reduce unnecessary energy consumption and, consequently, production costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and other purposes, features, and advantages of the present application will become more apparent through a more detailed description of the embodiments of the present application in conjunction with the accompanying drawings. The accompanying drawings are intended 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 drawings, the same reference numerals generally represent the same components or steps.

[0021] Figure 1 4 is a block diagram of an environmentally friendly antioxidant production management system according to an embodiment of the present application.

[0022] Figure 2 Schematic diagram of the structure of the environmentally friendly antioxidant production management system according to an embodiment of the present application.

[0023] Figure 3 4 is a block diagram of an energy consumption time series feature extraction module in an environmentally friendly antioxidant production management system according to an embodiment of the present application.

[0024] Figure 4 4 is a block diagram of an energy waste determination module in an environmentally friendly antioxidant production management system according to an embodiment of the present application.

[0025] Figure 5 Flowchart of an environmentally friendly antioxidant production management method according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] The following describes embodiments of the present disclosure in more detail with reference to the accompanying drawings. While the drawings illustrate certain embodiments of the present disclosure, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of protection of the present disclosure.

[0027] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in a different order and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0028] It should be noted that the terms "first, second, and third" in the embodiments of the present application are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It is understood that the terms "first, second, and third" can be interchanged to represent a specific order or precedence where permitted. It should be understood that the objects distinguished by "first, second, and third" can be interchanged where appropriate, such that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0029] As used in this disclosure and the claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not intended to refer to the singular but may include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0030] Various exemplary embodiments, features, and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise indicated.

[0031] Environmentally friendly antioxidants generally refer to those that are environmentally friendly and free of harmful substances. They can be extracted from natural resources or synthesized, but all must meet strict environmental standards. For example, some environmentally friendly antioxidants are extracted from plants, such as certain flavonoids and polyphenols, which have strong antioxidant properties and are safe for both humans and the environment. Furthermore, environmentally friendly antioxidants also include specific synthetic compounds that are designed with their environmental impact in mind, ensuring that they do not release harmful substances during use. Therefore, the development and application of environmentally friendly antioxidants is a crucial component of the plastics industry's sustainable development strategy. They can effectively slow the aging process of materials exposed to oxygen, light, and heat, thereby protecting them from damage and extending their service life.

[0032] Energy usage is crucial in the production of environmentally friendly antioxidants, as it directly impacts production efficiency, costs, and the environmental performance of the product. Energy waste in this process can lead to multiple problems. First, it increases production costs, as energy is a significant expense in the production process. Second, excessive energy consumption can exacerbate environmental pollution, particularly when using fossil fuels, leading to increased greenhouse gas emissions and negatively impacting global climate change. Furthermore, energy waste can lead to unsustainable resource use, as most energy resources are finite. In the chemical industry, particularly in antioxidant production, the application of energy-saving and consumption-reducing technologies can significantly reduce energy consumption, improve energy efficiency, and minimize environmental pollution and production costs. Therefore, developing and implementing more efficient production technologies and energy management systems is crucial. This not only helps protect the environment but also enhances a company's competitiveness and market sustainability.

[0033] However, traditional energy management systems may lack sufficient connectivity, resulting in decentralized monitoring and a lack of centralized management, which can hinder timely detection of defects and alerts. Furthermore, most energy management systems only provide feedback on a company's energy usage and are unable to analyze and identify energy waste points, requiring in-depth analysis by professionals.

[0034] Therefore, an optimized environmentally friendly antioxidant production management system is desired. This system, by acquiring time series of consumption of multiple energy types (including electricity, water, and steam) during the production process, and using artificial intelligence (AI) technology to perform temporal feature analysis and feature extraction on these time series, can intelligently identify areas of energy waste. In this way, the system can identify patterns and anomalies in energy usage, thereby optimizing energy efficiency during production to meet the green production philosophy of environmentally friendly antioxidants, while also reducing unnecessary energy consumption and, consequently, production costs.

[0035] Figure 1 4 is a block diagram of an environmentally friendly antioxidant production management system according to an embodiment of the present application. Figure 2 FIG. 1 is a schematic diagram of the structure of the environmentally friendly antioxidant production management system according to an embodiment of the present application. Figure 1 and Figure 2As shown, according to an embodiment of the present application, the environmentally friendly antioxidant production management system 100 includes: an environmentally friendly antioxidant production energy consumption time series acquisition module 110, which is used to obtain the time series of consumption of multiple energy types in the production process of the environmentally friendly antioxidant, wherein the energy types include electricity, water, and steam; an energy consumption time series feature extraction module 120, which is used to arrange the time series of consumption of the multiple energy types into multiple energy consumption time series input vectors and then perform energy consumption time series feature extraction to obtain multiple energy consumption multi-scale time series feature vectors; a cross-scale energy consumption feature analysis module 130, which is used to arrange the multiple energy consumption multi-scale time series feature vectors into an energy consumption type time series feature matrix and then perform cross-scale energy consumption feature analysis to obtain an energy consumption type multi-scale feature matrix; an energy global feature capture module 140, which is used to perform energy global feature capture on the energy consumption type multi-scale feature matrix to obtain an energy consumption type global feature matrix; and an energy waste link judgment module 150, which is used to determine the energy waste link based on the energy consumption type global feature matrix.

[0036] In an embodiment of the present application, the energy consumption time series acquisition module 110 for the production of environmentally friendly antioxidants is used to obtain the time series of the consumption of various energy types in the production process of environmentally friendly antioxidants, wherein the energy types include electricity, water, and steam. It should be understood that considering that the electricity consumption is the energy consumption generated by the operation of various motors, machinery, lighting, computer systems, automation equipment, and electronic equipment in the factory. The water consumption is the energy consumption used for cooling, heating, cleaning, process, boiler feed water, etc. in the production process of environmentally friendly antioxidants. The steam consumption is usually generated by boilers and used for heating, disinfection, driving steam turbines, or as a direct energy source for certain processes. These electricity, water, and steam are three basic energy types commonly found in industrial production, and their consumption has a significant impact on the cost and production efficiency of the production process of environmentally friendly antioxidants. Based on this, in the technical solution of the present application, the time series of the consumption of various energy types in the production process of environmentally friendly antioxidants is obtained, wherein the energy types include electricity, water, and steam, and the time series analysis and mutual correlation are performed on them, which can provide a good basis and support for the determination of subsequent energy waste links.

[0037] In an embodiment of the present application, the energy consumption time series feature extraction module 120 is used to arrange the time series of the consumption of the multiple energy types into multiple energy consumption time series input vectors and then perform energy consumption time series feature extraction to obtain multiple energy consumption multi-scale time series feature vectors. Figure 3 FIG is a block diagram of an energy consumption time series feature extraction module in an environmentally friendly antioxidant production management system according to an embodiment of the present application. Specifically, in the embodiment of the present application, Figure 3As shown, the energy consumption time series feature extraction module 120 includes: an energy consumption time sample arrangement unit 121, which is used to arrange the time series of the consumption of the multiple energy types into the multiple energy consumption time series input vectors according to the time dimension and the sample dimension; and a time series feature extraction unit 122, which is used to input the multiple energy consumption time series input vectors into the energy consumption time series feature extraction network based on 1D-CNN to obtain the multiple energy consumption multi-scale time series feature vectors.

[0038] Specifically, the energy consumption time sample arrangement unit 121 is configured to arrange the time series of the consumption of the multiple energy types according to the time dimension and the sample dimension into the multiple energy consumption time series input vectors. Accordingly, considering that each of the multiple energy types has time series characteristic information on a time scale, that is, the consumption of each energy type changes over time, and that the multiple energy sources are also interrelated in terms of time series, in order to preserve the time series characteristic information of each of the multiple energy types and enable separate analysis and processing of each energy source, in the technical solution of the present application, the time series of the multiple energy types are arranged according to the time dimension and the sample dimension into the multiple energy consumption time series input vectors. Specifically, arranging the time series of the multiple energy types according to the time dimension ensures that the time series input vectors are arranged in the order in which the time series were collected, providing more accurate input for subsequent analysis of time series characteristics, such as periodic changes in energy sources, short-term fluctuations in energy sources, and so on. Arranging the time series of the consumption of the multiple energy types according to the sample dimension enables independent analysis and understanding of each energy source, thereby enabling more accurate analysis of the time series characteristic information of each energy source.

[0039] Specifically, the time series feature extraction unit 122 is used to input the multiple energy consumption time series input vectors into the energy consumption time series feature extraction network based on 1D-CNN to obtain the multiple energy consumption multi-scale time series feature vectors. It should be understood that, considering that in energy consumption data, specific patterns and trends in the time series are crucial for determining the links of energy waste. That is, the multiple energy consumption time series input vectors have time series characteristics about each energy on a local time scale, and considering that the 1D-CNN is suitable for processing time series data, it can effectively capture the local time series pattern and feature information in the time series. Therefore, in the technical solution of the present application, the multiple energy consumption time series input vectors are input into the energy consumption time series feature extraction network based on 1D-CNN to obtain the multiple energy consumption multi-scale time series feature vectors. In this way, the local time series feature information about each energy type in each energy consumption time series input vector can be captured from the multiple energy consumption time series input vectors, thereby obtaining multiple energy consumption multi-scale time series feature vectors containing more abundant time series feature information.

[0040] In particular, in one achievable embodiment of the present application, the time series data of the consumption of multiple energy types are arranged into multiple energy consumption time series input vectors, and then energy consumption time series feature extraction is performed to obtain multiple energy consumption multi-scale time series feature vectors. This can be achieved in the following manner. First, the time series data of the consumption of multiple energy types are organized into multiple energy consumption time series input vectors, each vector representing the time series data of a single energy type. Next, feature extraction is performed on each energy consumption time series input vector using traditional time series feature extraction methods such as autocorrelation and lagged correlation. Specifically, autocorrelation refers to the correlation between time series data and itself at different time points. If a time series is correlated at one time point with itself at another time point, the relationship between the two time points can be described by an autocorrelation function. Specifically, the autocorrelation function can be measured by calculating the correlation coefficient of the time series at different time lags. Common methods include the Pearson correlation coefficient and the Spearman correlation coefficient, which help us understand periodicity, trends, and other important patterns in time series data. Lagged correlation refers to the correlation between time series data at different time points. By comparing the values ​​of a time series at different time points, the lagged relationships between the data can be revealed. Specifically, the lag correlation can be measured by calculating the correlation coefficient of the time series at different time lags. By calculating the correlation coefficients at different lag orders, the lag relationship between the data can be understood, which can be used to predict future data trends, identify causal relationships between data, etc. Then, the characteristics of each energy type are merged into a multi-scale time series feature vector. Simple splicing or weighted averaging methods can be used to obtain the multiple energy consumption multi-scale time series feature vectors. Through the above steps, the feature extraction and processing of the time series data of the consumption of multiple energy types can be realized to obtain multiple energy consumption multi-scale time series feature vectors, which can be used for the intelligent identification and optimization of energy waste links in the production management system of environmentally friendly antioxidants.

[0041] In an embodiment of the present application, the cross-scale energy consumption feature analysis module 130 is used to arrange the multiple energy consumption multi-scale time series feature vectors into an energy consumption type time series feature matrix and then perform cross-scale energy consumption feature analysis to obtain the energy consumption type multi-scale feature matrix. Specifically, in an embodiment of the present application, the cross-scale energy consumption feature analysis module is used to: arrange the multiple energy consumption multi-scale time series feature vectors into the energy consumption type time series feature matrix and then input the matrix into a cross-scale energy consumption feature analysis network to obtain the energy consumption type multi-scale feature matrix. In particular, the cross-scale energy consumption feature analysis network includes: a first convolution layer, a second convolution layer, and a cascade layer, wherein the first convolution layer is parallel to the second convolution layer, the cascade layer is connected to the first convolution layer and the second convolution layer, and the first convolution layer and the second convolution layer use different convolution kernel scales. It should be understood that, considering the direct temporal correlation between the multiple energy consumption multi-scale time series feature vectors, in the technical solution of the present application, the multiple energy consumption multi-scale time series feature vectors are arranged into the energy consumption type time series feature matrix. Then, considering that the energy consumption type time series feature matrix contains the time series patterns and characteristic laws of different time scales in the energy data, in order to capture and mine the characteristics of energy consumption at different time scales in the energy consumption type time series feature matrix, in the technical solution of the present application, the energy consumption type time series feature matrix is ​​input into the cross-scale energy consumption feature analysis network to obtain the energy consumption type multi-scale feature matrix. In other words, the energy consumption type time series features of different scales in the energy consumption type time series feature matrix can provide information at different levels and angles, thereby capturing characteristic information at different scales, helping to fully understand the complex patterns in the energy consumption data, thereby better describing the time series patterns and change laws of the energy data, and thus improving the discrimination and representation ability of the energy consumption type multi-scale feature matrix.

[0042] In an embodiment of the present application, the energy global feature capture module 140 is used to capture the energy global features of the energy consumption type multi-scale feature matrix to obtain the energy consumption type global feature matrix. Specifically, in an embodiment of the present application, the energy global feature capture module is used to: input the energy consumption type multi-scale feature matrix into the energy global feature capture module based on the non-local neural network to obtain the energy consumption type global feature matrix. Accordingly, considering that the energy consumption type multi-scale feature matrix is ​​obtained by arranging multiple energy consumption multi-scale time series feature vectors and performing multi-scale extraction, and the multi-scale extraction captures the local time series information of different scales in the energy consumption type multi-scale feature matrix, therefore, in order to better represent the overall information of the energy consumption type multi-scale feature matrix on a global scale, thereby improving the model's understanding and representation ability of global information, in the technical solution of the present application, the energy consumption type multi-scale feature matrix is ​​input into the energy global feature capture module based on the non-local neural network to obtain the energy consumption type global feature matrix. It is worth mentioning that the non-local neural network, by introducing non-local operations, can help the model better capture the correlation and dependency between global features, thereby improving the model's effect in processing global information. That is to say, the non-local neural network can establish dependencies between features in the global scope of the multi-scale feature matrix of the energy consumption type, thereby helping the model to better understand the structure and pattern of the overall energy consumption data, which helps to improve the model's ability to capture global information, thereby improving the model's performance and accuracy in energy consumption data analysis and energy waste judgment tasks.

[0043] In the embodiment of the present application, the energy waste link judgment module 150 is used to determine the energy waste link based on the global characteristic matrix of the energy consumption type. Figure 4 FIG. 1 is a block diagram of an energy waste determination module in an environmentally friendly antioxidant production management system according to an embodiment of the present application. Specifically, in the embodiment of the present application, Figure 4 As shown, the energy waste link judgment module 150 includes: a feature optimization unit 151, which is used to optimize the global feature matrix of the energy consumption type to obtain an optimized global feature matrix of the energy consumption type; and a link classification unit 152, which is used to input the optimized global feature matrix of the energy consumption type into a classifier-based energy link discriminator to obtain a classification result of the energy waste link.

[0044] Specifically, the feature optimization unit 151 is used to optimize the global feature matrix of the energy consumption type to obtain an optimized global feature matrix of the energy consumption type. Specifically, in an embodiment of the present application, the global feature matrix of the energy consumption type is subjected to structural depth stacking based on the feature distribution space to obtain the optimized global feature matrix of the energy consumption type. In particular, in the technical solution of the present application, considering that energy consumption data often comes from multiple energy types, each energy type has its own unique characteristics and consumption patterns. By improving the information aggregation degree of the logarithmic transformation of the global feature matrix of the energy consumption type, the information of various energy types can be better integrated, so that the model can fully understand the relationship and influence between different energy types. Energy consumption data usually contains features of different time scales, such as short-term fluctuations and long-term trends. By improving the information aggregation degree, feature information of different scales can be better integrated, so that the model can simultaneously capture energy consumption characteristics at different time scales, and improve the comprehensive understanding of energy consumption changes. In the process of energy consumption data processing, information loss is a common problem, especially in the data processing of multiple energy types and multiple time scales. By improving the information aggregation degree of the global feature matrix of the logarithmic transformation of energy consumption types, the possibility of information loss can be reduced, ensuring that the model can fully utilize all available information for accurate energy consumption analysis. In the classification task of energy waste links, accurately identifying the links of energy waste is crucial for saving energy and reducing waste. By improving the information aggregation degree of the global feature matrix of the logarithmic transformation of energy consumption types, the model's ability to understand and characterize energy consumption data can be improved, thereby improving the accuracy and reliability of the classification of energy waste links. Based on this, in the technical solution of the present application, the global feature matrix of energy consumption types is subjected to structural depth stacking based on the feature distribution space to obtain the optimized global feature matrix of energy consumption types.

[0045] More specifically, in an embodiment of the present application, the feature optimization unit is used to: calculate the variance of a feature set consisting of eigenvalues ​​at all positions of the global feature matrix of the energy consumption type to obtain the eigenvalue variance of the global feature matrix of the energy consumption type; calculate the square of the absolute value of the eigenvalues ​​at each position of the global feature matrix of the energy consumption type and then perform a logarithmic operation with base two to obtain a logarithmic-transformed global feature matrix of the energy consumption type; accumulate and sum all eigenvalues ​​of the logarithmic-transformed global feature matrix of the energy consumption type to obtain the sum of the logarithmic-transformed global feature matrix of the energy consumption type; and divide the sum of the logarithmic-transformed global feature matrix of the energy consumption type by twice the eigenvalue variance of the global feature matrix of the energy consumption type and then multiply it by the eigenvalues ​​at each position of the global feature matrix of the energy consumption type to obtain the optimized global feature matrix of the energy consumption type.

[0046] Preferably, in an embodiment of the present application, the feature optimization unit is configured to perform structural depth stacking based on the feature distribution space on the global feature matrix of the energy consumption type using the following feature optimization formula to obtain the optimized global feature matrix of the energy consumption type; wherein the feature optimization formula is:

[0047]

[0048] Where M represents the global characteristic matrix of the energy consumption type, m i,j is the eigenvalue of the (i, j)th position of the global characteristic matrix of the energy consumption type, σ 2 i,j (m i,j ) represents the eigenvalue set m i,j ∈M, and W and H are the width and height of the global feature matrix M of the energy consumption type, log2 represents the logarithm with base 2, m' i,j is the eigenvalue of the (i, j)th position of the global characteristic matrix of the optimized energy consumption type.

[0049] In particular, here, in order to improve the information aggregation degree of the global feature matrix of the energy consumption type, in the technical solution of the present application, the global feature matrix of the energy consumption type is subjected to structural depth stacking based on the feature distribution space, and the logarithmic value of the absolute phase stacking of the eigenvalues ​​at each position in the global feature matrix of the energy consumption type is used to represent the structural information of the global feature matrix of the energy consumption type at that position, and then a wavelet-like function is used to aggregate the structural information of the eigenvalues ​​at all positions in the global feature matrix of the energy consumption type, and the global feature matrix of the energy consumption type is stacked based on the feature distribution space using the collective variance of the structural information and the feature distribution of the global feature matrix of the energy consumption type. In this way, it helps to maintain the spatial relationship between the features in the global feature matrix of the energy consumption type, thereby restricting the spatial sparsity of the global feature matrix of the energy consumption type to improve the information aggregation degree of the global feature matrix of the energy consumption type.

[0050] Specifically, the link classification unit 152 is configured to input the optimized energy consumption type global feature matrix into a classifier-based energy link discriminator to obtain a classification result for energy-wasting links. In other words, by performing classification processing based on the optimized energy consumption type global feature matrix, energy-wasting links are intelligently identified. In this way, the system can identify patterns and anomalies in energy usage, thereby optimizing energy efficiency during the production process to comply with the green production concept of environmentally friendly antioxidants, while also reducing unnecessary energy consumption and, consequently, lowering production costs.

[0051] It should be understood that the classifier can map the input optimized energy consumption type global feature matrix to a specific category or label, thereby realizing the classification and judgment of the energy waste link, which helps to identify and distinguish different types of energy consumption situations, and then find out the links where energy waste may exist. This can provide support and reference for decision makers, helping them to better identify and understand which links have energy waste problems, and then formulate corresponding improvement and optimization measures. In this way, the classifier-based energy link discriminator can effectively classify and judge the optimized energy consumption type global feature matrix, help identify energy waste links, provide decision support, and realize automatic identification, thereby effectively optimizing energy utilization, reducing energy waste, and improving energy utilization efficiency.

[0052] In summary, the environmentally friendly antioxidant production management system 100 according to the embodiment of the present application is described. It obtains the time series of the consumption of various energy types during the production process of environmentally friendly antioxidants, wherein the energy types include electricity, water, and steam, and uses artificial intelligence technology to perform time series feature analysis and feature extraction on the time series of the consumption of the various energy types, thereby intelligently determining the links of energy waste. In this way, the system can identify patterns and anomalies in energy use, thereby optimizing the energy efficiency of the production process to comply with the green production concept of environmentally friendly antioxidants, while reducing unnecessary energy consumption and thus reducing production costs.

[0053] Figure 5 Flowchart of the environmentally friendly antioxidant production management method according to the embodiment of the present application. Figure 5 As shown, according to the embodiment of the present application, the environmentally friendly antioxidant production management method includes: S110, obtaining the time series of multiple energy type consumption in the environmentally friendly antioxidant production process, wherein the energy types include electricity, water, and steam; S120, arranging the time series of the multiple energy type consumption into multiple energy consumption time series input vectors and then performing energy consumption time series feature extraction to obtain multiple energy consumption multi-scale time series feature vectors; S130, arranging the multiple energy consumption multi-scale time series feature vectors into an energy consumption type time series feature matrix and then performing cross-scale energy consumption feature analysis to obtain an energy consumption type multi-scale feature matrix; S140, capturing energy global features of the energy consumption type multi-scale feature matrix to obtain an energy consumption type global feature matrix; and, S150, determining the energy waste link based on the energy consumption type global feature matrix.

[0054] Here, those skilled in the art will appreciate that the specific operations of each step in the above-mentioned environmentally friendly antioxidant production management method have been described in detail above. Figures 1 to 4 The environmentally friendly antioxidant production management system has been described in detail, and therefore, its repeated description will be omitted.

[0055] According to an exemplary implementation of the present disclosure, a computer-readable storage medium is provided, on which computer-executable instructions are stored, wherein the computer-executable instructions are executed by a processor to implement the method described above. According to an exemplary implementation of the present disclosure, a computer program product is also provided, which is tangibly stored on a non-transitory computer-readable medium and includes computer-executable instructions, and the computer-executable instructions are executed by a processor to implement the method described above. According to an exemplary implementation of the present disclosure, a computer program product is provided, on which a computer program is stored, and when the program is executed by a processor, the method described above is implemented.

[0056] Various aspects of the present disclosure are described herein with reference to flowcharts and / or block diagrams of methods, apparatuses, devices, and computer program products implemented according to the present disclosure. It should be understood that each block of the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer-readable program instructions.

[0057] Furthermore, although operations are depicted in a particular order, this should be understood as requiring that such operations be performed in the particular order shown or in a sequential order, or that all illustrated operations be performed to achieve the desired result. Multitasking and parallel processing may be advantageous under certain circumstances. Similarly, although the above discussion includes several specific implementation details, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of separate embodiments may also be implemented in combination in a single implementation. Conversely, various features described in the context of a single implementation may also be implemented in multiple implementations, either individually or in any suitable subcombination.

[0058] Although the subject matter has been described in language specific to structural features and / or methodological acts, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are merely example forms of implementing the claims.

Claims

1. An environmentally friendly antioxidant production management system, characterized in that: include: The module for acquiring the time series of energy consumption during the production of environmentally friendly antioxidants is used to acquire the time series of consumption of various energy types during the production of environmentally friendly antioxidants, wherein the energy types include electricity, water, and steam; an energy consumption time series feature extraction module, configured to extract energy consumption time series features after arranging the time series of the consumption of the multiple energy types into a plurality of energy consumption time series input vectors to obtain a plurality of energy consumption multi-scale time series feature vectors; a cross-scale energy consumption feature analysis module, configured to arrange the plurality of energy consumption multi-scale time series feature vectors into an energy consumption type time series feature matrix and then perform cross-scale energy consumption feature analysis to obtain an energy consumption type multi-scale feature matrix; An energy global feature capture module, configured to capture the energy global features of the energy consumption type multi-scale feature matrix to obtain an energy consumption type global feature matrix; An energy waste link judgment module, used to determine the energy waste link based on the global characteristic matrix of energy consumption types; The energy waste link judgment module includes: The feature optimization unit is used to: calculate the variance of the feature set composed of the eigenvalues ​​at all positions of the global feature matrix of the energy consumption type to obtain the eigenvalue variance of the global feature matrix of the energy consumption type; calculate the square of the absolute value of the eigenvalue at each position of the global feature matrix of the energy consumption type and then perform a logarithmic operation with base two to obtain a logarithmic transformation of the global feature matrix of the energy consumption type; accumulate and sum all the eigenvalues ​​of the global feature matrix of the logarithmic transformation of the energy consumption type to obtain a sum value of the global feature matrix of the logarithmic transformation of the energy consumption type; divide the sum value of the global feature matrix of the logarithmic transformation of the energy consumption type by twice the eigenvalue variance of the global feature matrix of the energy consumption type and then multiply it by the eigenvalue at each position of the global feature matrix of the energy consumption type to obtain the optimized global feature matrix of the energy consumption type; Link classification unit for inputting the optimized energy consumption type global feature matrix into the energy link discriminator based on the classifier to obtain the classification result of the energy waste link; In the feature optimization unit, the logarithmic value of the absolute phase stacking of the eigenvalues ​​at each position in the global feature matrix of the energy consumption type is used to represent the structural information of the global feature matrix of the energy consumption type at that position, and then a wavelet-like function is used to aggregate the structural information of the eigenvalues ​​at all positions in the global feature matrix of the energy consumption type, and the global feature matrix of the energy consumption type is stacked based on the feature distribution space using the structural information and the collective variance of the feature distribution of the global feature matrix of the energy consumption type. In this way, it helps to maintain the spatial relationship between the features in the global feature matrix of the energy consumption type, thereby limiting the spatial sparsity of the global feature matrix of the energy consumption type to improve the information aggregation degree of the global feature matrix of the energy consumption type.

2. The environmentally friendly antioxidant production management system according to claim 1, characterized in that: The energy consumption time series feature extraction module includes: an energy consumption time sample arrangement unit, configured to arrange the time series of the consumption of the plurality of energy types into the plurality of energy consumption time series input vectors according to a time dimension and a sample dimension; The time series feature extraction unit is used to input the multiple energy consumption time series input vectors into the energy consumption time series feature extraction network based on 1D-CNN to obtain the multiple energy consumption multi-scale time series feature vectors.

3. The environmentally friendly antioxidant production management system according to claim 2, characterized in that: The cross-scale energy consumption characteristic analysis module is used to: The multiple energy consumption multi-scale time series feature vectors are arranged into the energy consumption type time series feature matrix and then input into a cross-scale energy consumption feature analysis network to obtain the energy consumption type multi-scale feature matrix.

4. The environmentally friendly antioxidant production management system according to claim 3, characterized in that: The cross-scale energy consumption characteristic analysis network includes: A first convolutional layer, a second convolutional layer, and a cascade layer, wherein the first convolutional layer is parallel to the second convolutional layer, the cascade layer is connected to the first convolutional layer and the second convolutional layer, and the first convolutional layer and the second convolutional layer use convolution kernels of different scales.

5. The environmentally friendly antioxidant production management system according to claim 4, characterized in that: The energy global feature capture module is used to: The multi-scale feature matrix of energy consumption type is input into an energy global feature capture module based on a non-local neural network to obtain the global feature matrix of energy consumption type.

6. A production management method for environmentally friendly antioxidants, characterized in that: include: Obtaining a time series of consumption of various energy types during the production process of an environmentally friendly antioxidant, wherein the energy types include electricity, water, and steam; Arranging the time series of the consumption of the multiple energy types into a plurality of energy consumption time series input vectors and then extracting energy consumption time series features to obtain a plurality of energy consumption multi-scale time series feature vectors; Arranging the plurality of energy consumption multi-scale time series feature vectors into an energy consumption type time series feature matrix and then performing cross-scale energy consumption feature analysis to obtain an energy consumption type multi-scale feature matrix; Capturing global energy features on the multi-scale feature matrix of energy consumption types to obtain a global feature matrix of energy consumption types; Based on the global characteristic matrix of energy consumption types, determining the link of energy waste; Wherein, based on the global characteristic matrix of energy consumption types, determining the link of energy waste includes: Calculating the variance of a feature set composed of eigenvalues ​​at all positions of the global feature matrix of the energy consumption type to obtain the eigenvalue variance of the global feature matrix of the energy consumption type; calculating the square of the absolute value of the eigenvalue at each position of the global feature matrix of the energy consumption type and then performing a logarithmic operation with base two to obtain a logarithmic transformation of the global feature matrix of the energy consumption type; accumulating and summing all eigenvalues ​​of the global feature matrix of the logarithmic transformation of the energy consumption type to obtain a sum of the global feature matrix of the logarithmic transformation of the energy consumption type; dividing the sum of the global feature matrix of the logarithmic transformation of the energy consumption type by twice the eigenvalue variance of the global feature matrix of the energy consumption type and then multiplying it by the eigenvalue at each position of the global feature matrix of the energy consumption type to obtain the optimized global feature matrix of the energy consumption type; Inputting the optimized energy consumption type global feature matrix into the energy link discriminator based on the classifier to obtain the classification result of the energy waste link; In the feature optimization unit, the logarithmic value of the absolute phase stacking of the eigenvalues ​​at each position in the global feature matrix of the energy consumption type is used to represent the structural information of the global feature matrix of the energy consumption type at that position, and then a wavelet-like function is used to aggregate the structural information of the eigenvalues ​​at all positions in the global feature matrix of the energy consumption type, and the global feature matrix of the energy consumption type is stacked based on the feature distribution space using the structural information and the collective variance of the feature distribution of the global feature matrix of the energy consumption type. In this way, it helps to maintain the spatial relationship between the features in the global feature matrix of the energy consumption type, thereby limiting the spatial sparsity of the global feature matrix of the energy consumption type to improve the information aggregation degree of the global feature matrix of the energy consumption type.

7. The environmentally friendly antioxidant production management method according to claim 6, characterized in that: Arranging the time series of the consumption of the multiple energy types into multiple energy consumption time series input vectors and then performing energy consumption time series feature extraction to obtain multiple energy consumption multi-scale time series feature vectors, including: Arranging the time series of consumption of the multiple energy types into the multiple energy consumption time series input vectors according to the time dimension and the sample dimension; The multiple energy consumption time series input vectors are input into an energy consumption time series feature extraction network based on 1D-CNN to obtain the multiple energy consumption multi-scale time series feature vectors.

Images