Carbon emission real-time monitoring method and system

Abstract

The invention relates to the technical field of carbon emission monitoring, in particular to a carbon emission real-time monitoring method and system. The method comprises the following steps: acquiring a carbon emission data value of each process at each moment in an industrial park in real time; acquiring a plurality of clusters of each process by using an iterative self-organizing clustering algorithm; obtaining a multi-dimensional cluster according to the intersection condition of each moment; calculating the attribution degree of each moment in the target multi-dimensional clustering cluster, wherein the moment is not in the multi-dimensional clustering cluster; according to the affiliation degree, dividing an affiliation multi-dimensional clustering cluster; calculating the abnormal degree of each multi-dimensional cluster; calculating an abnormal value at each moment; and according to the abnormal value, determining a carbon emission condition so as to realize real-time monitoring of carbon emission. According to the method, the problem of inaccurate result caused by direct clustering of a traditional iterative self-organizing clustering algorithm is effectively avoided, the accuracy of carbon emission monitoring is greatly improved, and accurate and real-time monitoring of carbon emission of the industrial park is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission monitoring, and in particular to a method and system for real-time monitoring of carbon emissions. Background Art

[0002] As the global climate change problem becomes increasingly serious, the reduction of carbon emissions has become a common focus of attention for governments and all sectors of society. As one of the main greenhouse gases, excessive carbon dioxide emissions are one of the main causes of climate warming, environmental degradation and ecological imbalance. Therefore, accurate carbon emission monitoring and management have become a key link in environmental protection and climate change response. By applying the iterative self-organizing clustering algorithm to real-time carbon emission monitoring, clustering results can be obtained to ensure the accuracy of subsequent monitoring.

[0003] The patent document with announcement number CN117556245B discloses a method for detecting impurities in the production filtration of tetramethylammonium hydroxide, including: collecting spectral data of a tetramethylammonium hydroxide solution to obtain independent components of the spectral data, obtaining ordered independent components of the spectral data based on the independent components, matching wavelength pairs during DTW matching based on the ordered independent components of the spectral data, the independent components of the spectral data within a cluster, and the independent components of different spectral data within a cluster, obtaining the preference of the cluster to be split into two clusters during iterative self-organizing clustering, and iteratively self-organizing clustering all spectral data based on the preference and preset parameters to obtain the main impurities of the tetramethylammonium hydroxide solution corresponding to the final cluster.

[0004] However, the above patent document is not aimed at carbon emission monitoring, and does not solve the problem that the traditional iterative self-organizing clustering algorithm will directly complete the clustering according to the Euclidean distance between multidimensional data points during the clustering of multidimensional data; and for the carbon emission data of each process in the industrial park, since there are differences in the distribution of data values ​​in different processes at the same time, if all times are clustered directly, the clustering results will be inaccurate, which will affect the accuracy of carbon emission monitoring. Summary of the invention

[0005] In order to solve the problem that the traditional iterative self-organizing clustering algorithm directly completes clustering according to the Euclidean distance between multidimensional data points during the clustering of multidimensional data, and since the carbon emission data values ​​of each process at each moment in the industrial park are different, if all moments are clustered directly, it will lead to inaccurate clustering results, thereby affecting the accuracy of carbon emission monitoring, the present invention provides a real-time monitoring method and system for carbon emissions.

[0006] In a first aspect, the present invention provides a method for real-time monitoring of carbon emissions, which adopts the following technical solution: The real-time monitoring method of carbon emissions includes: obtaining the carbon emission data value of each process at each time in an industrial park in real time; clustering the carbon emission data values ​​in each process at all times by using an iterative self-organizing clustering algorithm to obtain multiple clustering clusters for each process and the clustering clusters corresponding to each time in each process; forming multidimensional clustering clusters from the moments that are in the same clustering cluster in the multiple clustering clusters of each process at all times to obtain multiple multidimensional clustering clusters; recording any multidimensional clustering cluster as a target multidimensional clustering cluster, recording the clustering center of the target multidimensional clustering cluster as a target clustering center, and calculating the time that is not in the multidimensional clustering cluster based on the difference from any time that is not in the multidimensional clustering cluster to the target clustering center. The method comprises the following steps: calculating the degree of belonging to the target multidimensional clustering cluster; dividing the multidimensional clustering clusters according to the degree of belonging to obtain all the moments belonging to the target multidimensional clustering cluster; calculating the degree of abnormality of the target multidimensional clustering cluster based on the number of all the moments in the target multidimensional clustering cluster, the average Euclidean distance from all the moments to the target clustering center, and the average dispersion of the Euclidean distance from all the moments to the target clustering center relative to the average Euclidean distance; calculating the abnormal value of each moment in the target multidimensional clustering cluster based on the Euclidean distance from the moment to the target clustering center and the abnormal degree of the average dispersion; determining the carbon emission status according to the size of the abnormal value to realize real-time monitoring of carbon emissions.

[0007] The beneficial effects are: by integrating the clustering results of the carbon emission data values ​​at each moment in the multidimensional data in all processes, the limitations caused by focusing only on the data values ​​in a single process are avoided; by calculating the degree of attribution and outliers at each moment, combined with the discreteness and degree of anomaly within the cluster, abnormal fluctuations in carbon emissions can be effectively discovered; by real-time updating of clusters and cluster centers, it adapts to data changes and ensures that monitoring results are timely and accurate; by comprehensively analyzing the differences between moments and processes, comprehensive monitoring of carbon emissions in industrial parks can be achieved, providing an accurate basis for decision-making.

[0008] Furthermore, the various processes include a smelting process, a power generation process, a steel rolling process and a coking process.

[0009] Furthermore, the carbon emission data value of the cluster center in each process is: the average value of the carbon emission data value in each process at all times in the multidimensional clustering cluster.

[0010] Furthermore, the degree of belonging satisfies the following relationship: ; In the formula, For the The moment that is not in the multidimensional cluster is The degree of belonging to the multidimensional clusters, is the number of processes, For the The time when the cluster is not in the multidimensional clustering cluster to the The difference in cluster centers of multidimensional clusters, For the The moment that is not in the multidimensional cluster is The carbon emission data value in the process, For the The cluster center of the multidimensional cluster is in The carbon emission data value in the process, For all moments that are not in the multidimensional clustering cluster, The average value of carbon emission data in each process, is a natural exponential function.

[0011] The beneficial effects are: by considering the difference between the carbon emission data of each process and the cluster center, and using distance and standardization methods, the degree of belonging to different clusters at each moment can be accurately calculated; by standardizing the carbon emission data, the impact of outliers or unstable data on the clustering results is reduced, and the robustness of the model is improved; using an exponential function as a calculation method for the degree of belonging can strengthen the clustering of data with a close distance, enhance the differentiation of clusters, and help to better distinguish the carbon emission status of different processes.

[0012] Furthermore, the multidimensional clustering cluster is divided to obtain all moments belonging to the target multidimensional clustering cluster, including: in response to the degree of belonging to the target multidimensional clustering cluster of any moment that is not in the multidimensional clustering cluster is greater than a preset belonging threshold, and the degree of belonging is the maximum value among the degrees of belonging of all multidimensional clusters at the moment that is not in the multidimensional clustering cluster, the moment is divided into the target multidimensional clustering cluster; in response to the degree of belonging of all multidimensional clustering clusters of any moment that is not in the multidimensional clustering cluster is not greater than the preset belonging threshold, the moment is recorded as a moment to be divided, and then all the moments to be divided are combined into a multidimensional clustering cluster.

[0013] The beneficial effects are: by setting the attribution threshold, the moments that do not meet the conditions can be effectively eliminated, avoiding the misclassification of noise data or abnormal points into existing multidimensional clustering clusters, ensuring that the multidimensional clustering results are more accurate; the target moments with a lower degree of attribution are classified as "moments to be divided", and new multidimensional clustering clusters are formed subsequently, so that the clustering process can adapt to the dynamically changing pattern, avoiding the forced classification of ambiguous moments into existing multidimensional clustering clusters, and ensuring the system's adaptive ability.

[0014] Furthermore, the abnormality degree satisfies the following relationship: ; In the formula, For the The abnormality of the multidimensional clusters, For the The number of moments within a multidimensional clustering cluster, For the All the time in the multidimensional clustering clusters to the The average Euclidean distance between the cluster centers of multidimensional clusters, For the All the time in the multidimensional clustering clusters to the The Euclidean distance of the cluster centers of multidimensional clusters relative to The average dispersion of For the Multidimensional clustering cluster The moment to The Euclidean distance between the cluster centers of multidimensional clusters, is the natural exponential function, is the standard normalization function.

[0015] The beneficial effects are: by calculating the abnormality degree of the multidimensional clustering cluster, the abnormal data points within the multidimensional clustering cluster can be effectively identified; the abnormality degree combines the number of moments, the average distance of the cluster center and the discreteness of the data, and can more comprehensively measure the abnormality of the multidimensional clustering cluster; it can dynamically adapt to different multidimensional clustering clusters, taking into account not only the number of moments of each cluster, but also the discreteness of the data points, and can cope with clusters with different data distributions and sizes.

[0016] Furthermore, the abnormal value satisfies the following relationship: ; In the formula, For the Multidimensional clustering cluster The outlier value at a moment, For the Multidimensional clustering cluster The moment to The Euclidean distance between the cluster centers of multidimensional clusters, For the All the time in the multidimensional clustering clusters to the The Euclidean distance of the cluster centers of multidimensional clusters relative to The average dispersion of For the The number of moments within a multidimensional clustering cluster, For the All the time in the multidimensional clustering clusters to the The average Euclidean distance between the cluster centers of multidimensional clusters, For the The abnormality of the multidimensional clusters, is the standard normalization function.

[0017] The beneficial effects are as follows: by comprehensively considering the distance, discreteness and degree of abnormality within the multidimensional clustering cluster, potential abnormal moments can be monitored in more complex situations; by combining Euclidean distance, discreteness and standardized processing, outliers are made more flexible and can adapt to different types of clustering clusters and data sets; the introduction of discreteness takes into account the changes within the clustering cluster, avoiding the misjudgment of naturally fluctuating data points within the cluster as abnormalities, thereby improving the accuracy of anomaly monitoring; by introducing the degree of abnormality of the cluster, the sensitivity of the outliers can be dynamically adjusted according to the nature of the cluster, making anomaly monitoring of different clusters more accurate.

[0018] Furthermore, the real-time monitoring of carbon emissions includes: in response to the abnormal value at the current moment being greater than a preset abnormal threshold, determining that the carbon emission status at the current moment is abnormal, and issuing an early warning prompt to complete the real-time monitoring of carbon emissions.

[0019] In a second aspect, the present invention provides a real-time carbon emission monitoring system, which adopts the following technical solution: The carbon emission real-time monitoring system comprises: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned carbon emission real-time monitoring method is implemented.

[0020] By adopting the above technical solution, the above carbon emission real-time monitoring method is generated into a computer program and stored in a memory so as to be loaded and executed by a processor, thereby making a terminal device based on the memory and the processor for easy use.

[0021] The present invention has the following technical effects: By rationally using the iterative self-organizing clustering algorithm and taking full account of the differences in carbon emission data values ​​of each process in the industrial park, the carbon emission data values ​​of each process are first clustered to obtain multiple clustering clusters, and then multidimensional clustering clusters are further constructed. The degree of attribution of the target time is accurately calculated based on the Euclidean distance and reasonably divided. In addition, the abnormal degree of the multidimensional clustering cluster and the abnormal values ​​of each time are calculated based on various factors. Finally, the carbon emission status is determined based on the abnormal values. This effectively avoids the problem of inaccurate results caused by direct clustering in traditional algorithms, greatly improves the accuracy of carbon emission monitoring, and realizes accurate and real-time monitoring of carbon emissions in industrial parks. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-restrictive manner, and the same or corresponding numbers are the same or corresponding parts.

[0023] Figure 1It is a flow chart of the method in the real-time monitoring method of carbon emissions according to an embodiment of the present invention.

[0024] Figure 2 It is a schematic diagram of the moments of forming multidimensional clusters in the real-time monitoring method for carbon emissions according to an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0026] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0027] The present invention discloses a method for real-time monitoring of carbon emissions. Figure 1 , comprising steps S1 to S8: S1: Obtain the carbon emission data value of each process at each time in the industrial park in real time.

[0028] It should be noted that carbon emission data refers to the measurement of carbon dioxide (CO2) released into the atmosphere by a region or enterprise within a specific period of time. ) and other greenhouse gases, such as methane and nitrogen oxides. These data are usually closely related to the combustion of fossil fuels, industrial production, transportation, energy consumption, agriculture and other activities. Since carbon emissions are one of the main drivers of global climate change. Monitoring carbon emissions data helps assess the impact of human activities on the climate system, and then helps formulate policies and measures to address climate change.

[0029] Specifically, the various processes include a smelting process, a power generation process, a steel rolling process and a coking process.

[0030] Implementers can set the collection time interval according to the specific implementation situation, for example, 15 minutes / time.

[0031] S2: Use iterative self-organizing clustering algorithm to obtain multiple clusters for each process.

[0032] It should be noted that in industrial production, carbon emission anomalies are usually related to equipment failure, production anomalies or operational errors. By clustering the carbon emission data of each process separately, it is easier to find the abnormal pattern of a certain process. Therefore, the present invention obtains the clustering cluster of each process by clustering the data of any process.

[0033] The iterative self-organizing clustering algorithm is used to cluster the carbon emission data values ​​in each process at all times, and multiple clusters for each process and the corresponding clusters in each process at each time are obtained.

[0034] S3: Obtain multidimensional clusters according to the intersection at each moment.

[0035] It should be noted that in most industrial processes, the production links between different processes are closely related. For example, in an industrial park, the smelting process and the power generation process are closely related. A large amount of electricity is required in the smelting process, and the carbon emissions of the power generation process are affected by the quality of energy supply. If the carbon emission data of the power generation process at a certain moment on a certain day increases significantly, and the carbon emission data of the smelting process also has the same fluctuation, then they are likely to be affected by the same external factors (such as fuel quality problems, power fluctuations, etc.). By clustering the carbon emission data of these processes at the same time, it can help staff better obtain abnormalities in the monitored carbon emission data.

[0036] All the moments that are in the same cluster in multiple clusters of each process are combined into multidimensional clusters to obtain multiple multidimensional clusters, such as Figure 2 As shown, respectively Moment and The data of all processes collected at the time is recorded as , , and , , ;like and In process In the same cluster of and In process In the same cluster of and In process In any cluster of Moment and At any moment, a multidimensional cluster is formed.

[0037] S4: Calculate the degree of belonging to the target multidimensional clustering cluster at each moment that is not in the multidimensional clustering cluster.

[0038] It should be noted that, in the above steps, multidimensional data points with a greater degree of similarity (i.e., each moment) are placed in the same cluster to form a multidimensional cluster with data values ​​of multiple dimensions (i.e., processes). As for the multidimensional data points that are not in the multidimensional cluster, since there is a correlation between the carbon emission data of different processes, the present invention calculates the degree of belonging of any multidimensional data point that is not in the multidimensional cluster relative to any multidimensional cluster, so as to facilitate subsequent classification according to the degree of belonging, which helps to improve the overall clustering effect.

[0039] Any multidimensional clustering cluster is recorded as a target multidimensional clustering cluster, and the cluster center of the target multidimensional clustering cluster is recorded as the target cluster center.

[0040] Specifically, the carbon emission data value of the cluster center in each process is: the average value of the carbon emission data value in each process at all times in the multidimensional clustering cluster.

[0041] Based on the difference from any moment not in the multidimensional clustering cluster to the target clustering center, the degree of belonging of the moment not in the multidimensional clustering cluster to the target multidimensional clustering cluster is calculated.

[0042] Specifically, the degree of belonging satisfies the following relationship: ; In the formula, For the The moment that is not in the multidimensional cluster is The degree of belonging to the multidimensional clusters, is the number of processes, For the The time when the cluster is not in the multidimensional clustering cluster to the The difference in cluster centers of multidimensional clusters, For the The moment that is not in the multidimensional cluster is The carbon emission data value in the process, For the The cluster center of the multidimensional cluster is in The carbon emission data value in the process, For all moments that are not in the multidimensional clustering cluster, The average value of carbon emission data in each process, is a natural exponential function.

[0043] in, Indicates The time when the cluster is not in the multidimensional clustering cluster to the The smaller the difference of the cluster centers of the multidimensional clusters, the greater the degree of belonging to the multidimensional cluster at that moment. right Standardization operations are performed to eliminate the impact of data distribution between different dimensions, making the calculation results of the degree of attribution more accurate.

[0044] S5: Divide the multidimensional clusters according to the degree of belonging.

[0045] It should be noted that if a moment that is not in a multidimensional cluster has a higher degree of belonging to a multidimensional cluster, it is more likely to be classified into the multidimensional cluster. At the same time, if a moment has a lower degree of belonging to all multidimensional clusters, it means that the moment is very different from all multidimensional clusters, and this type of moment should be treated as a separate multidimensional cluster.

[0046] According to the degree of belonging, the multidimensional clusters are divided to obtain all the moments belonging to the target multidimensional cluster.

[0047] Specifically, the dividing of the multidimensional clusters to obtain all moments belonging to the target multidimensional cluster includes: In response to the belonging degree of any moment not in the multidimensional clustering cluster in the target multidimensional clustering cluster being greater than a preset belonging threshold, and the belonging degree at the moment not in the multidimensional clustering cluster is the maximum among the belonging degrees of all multidimensional clustering clusters, the moment is classified into the target multidimensional clustering cluster; In response to the fact that the degree of belonging of any moment not in the multidimensional clustering cluster to all multidimensional clustering clusters is not greater than the preset belonging threshold, the moment is recorded as the moment to be divided, and then all the moments to be divided are combined into a multidimensional clustering cluster.

[0048] Implementers can set the attribution threshold according to specific implementation circumstances, for example, 0.5.

[0049] S6: Calculate the abnormality degree of each multidimensional cluster.

[0050] It should be noted that, for any multidimensional clustering cluster, the fewer the number of moments in the multidimensional clustering cluster and the more discrete the distribution of the moments in the multidimensional clustering cluster, the higher the abnormality of the multidimensional clustering cluster.

[0051] The abnormality degree of the target multidimensional clustering cluster is calculated based on the number of moments in the target multidimensional clustering cluster, the average Euclidean distance from all moments in the target multidimensional clustering cluster to the target cluster center, and the average dispersion of the Euclidean distance from all moments in the target multidimensional clustering cluster to the target cluster center relative to the average Euclidean distance.

[0052] Specifically, the abnormality degree satisfies the following relationship: ; In the formula, For the The abnormality of the multidimensional clusters, For the The number of moments within a multidimensional clustering cluster, For the All the time in the multidimensional clustering clusters to the The average Euclidean distance between the cluster centers of multidimensional clusters, For the All the time in the multidimensional clustering clusters to the The Euclidean distance of the cluster centers of multidimensional clusters relative to The average dispersion of For the Multidimensional clustering cluster The moment to The Euclidean distance between the cluster centers of multidimensional clusters, is the natural exponential function, is the standard normalization function.

[0053] in, Indicates All the time in the multidimensional clustering clusters to the The average Euclidean distance of the cluster center of a multidimensional cluster. The larger the value, the farther the distance between all multidimensional data points and the cluster center. Indicates All the time in the multidimensional clustering clusters to the The Euclidean distance of the cluster centers of multidimensional clusters relative to The smaller the average dispersion, the more discrete the distribution of all multidimensional data points in the multidimensional clustering cluster is. At this time, the higher the abnormality of the multidimensional clustering cluster is.

[0054] S7: Calculate the outlier value at each moment.

[0055] It should be noted that, for any moment, the closer the distance between the moment and the cluster center is, the more the abnormality degree of the multidimensional cluster can be used to represent the abnormal value of the moment.

[0056] For each moment in the target multidimensional cluster, the outlier value at that moment is calculated based on the Euclidean distance from that moment to the target cluster center, as well as the average dispersion and the degree of outlier.

[0057] Specifically, the abnormal value satisfies the following relationship: ; In the formula, For the Multidimensional clustering cluster The outlier value at a moment, For the Multidimensional clustering cluster The moment to The Euclidean distance between the cluster centers of multidimensional clusters, For the All the time in the multidimensional clustering clusters to the The Euclidean distance of the cluster centers of multidimensional clusters relative to The average dispersion of For the The number of moments within a multidimensional clustering cluster, For the All the time in the multidimensional clustering clusters to the The average Euclidean distance between the cluster centers of multidimensional clusters, For the The abnormality of the multidimensional clusters, is the standard normalization function.

[0058] in, Indicates Multidimensional clustering cluster The moment to The difference between the Euclidean distances of the cluster centers of the clusters and the average dispersion, the smaller the value is, the more the abnormal degree of the multidimensional cluster can be used to represent the abnormal value of the moment; if the value is larger, it means that the difference between the moment and the multidimensional cluster is greater, and in the above-mentioned process of dividing the multidimensional clusters and assigning multidimensional clusters to all moments that are not in the initial clusters, all moments have been divided into the multidimensional cluster cluster that is most similar to the moment. Therefore, at this time, if the difference is greater, it means that the abnormal value of the moment is higher.

[0059] S8: Determine the carbon emission status according to the size of the abnormal value to achieve real-time monitoring of carbon emissions.

[0060] Specifically, the real-time monitoring of carbon emissions includes: In response to the abnormal value at the current moment being greater than the preset abnormal threshold, the carbon emission status at the current moment is determined to be abnormal, and an early warning prompt is issued to complete the real-time monitoring of carbon emissions.

[0061] Implementers can set the abnormal threshold according to specific implementation circumstances, for example, 0.8.

[0062] An embodiment of the present invention further discloses a carbon emission real-time monitoring system, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a carbon emission real-time monitoring method according to the present invention is implemented.

[0063] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0064] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, or any other medium that can be used to store the required information and can be accessed by an application, module, or both. Any such computer storage medium may be part of the device or accessible or connectable to the device.

[0065] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0066] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for real-time monitoring of carbon emissions, characterized in that: include: Obtain the carbon emission data values ​​of each process at each moment in the industrial park in real time; use the iterative self-organizing clustering algorithm to cluster the carbon emission data values ​​in each process at all moments, and obtain multiple clustering clusters for each process and the clustering clusters corresponding to each moment in each process; form multidimensional clustering clusters from all moments that are in the same clustering cluster in multiple clustering clusters of each process, and obtain multiple multidimensional clustering clusters; record any multidimensional clustering cluster as the target multidimensional clustering cluster, record the clustering center of the target multidimensional clustering cluster as the target clustering center, and calculate the degree of belonging of any moment that is not in the multidimensional clustering cluster to the target multidimensional clustering cluster based on the difference from any moment that is not in the multidimensional clustering cluster to the target clustering center; According to the degree of belonging, the multidimensional clustering clusters are divided to obtain all the moments belonging to the target multidimensional clustering cluster; based on the number of all the moments in the target multidimensional clustering cluster, the average Euclidean distance from all the moments to the target clustering center, and the average dispersion of the Euclidean distance from all the moments to the target clustering center relative to the average Euclidean distance, the abnormality degree of the target multidimensional clustering cluster is calculated; For each moment in the target multidimensional cluster, based on the Euclidean distance from the moment to the target cluster center and the abnormal degree of the average dispersion, calculate the abnormal value of the moment; The carbon emission status is determined according to the size of the abnormal value to achieve real-time monitoring of carbon emissions.

2. The method for real-time monitoring of carbon emissions according to claim 1, characterized in that: The various processes include a smelting process, a power generation process, a steel rolling process and a coking process.

3. The method for real-time monitoring of carbon emissions according to claim 1, characterized in that: The carbon emission data value of the cluster center in each process is: the average value of the carbon emission data value in each process at all times in the multidimensional clustering cluster.

4. The method for real-time monitoring of carbon emissions according to claim 3, characterized in that: The degree of belonging satisfies the following relationship: ； In the formula, For the The moment that is not in the multidimensional cluster is The degree of belonging to the multidimensional clusters, is the number of processes, For the The time when the cluster is not in the multidimensional clustering cluster to the The difference in cluster centers of multidimensional clusters, For the The moment that is not in the multidimensional cluster is The carbon emission data value in the process, For the The cluster center of the multidimensional cluster is in The carbon emission data value in the process, For all moments that are not in the multidimensional clustering cluster, The average value of carbon emission data in each process, is a natural exponential function.

5. The method for real-time monitoring of carbon emissions according to claim 1, characterized in that: The dividing of the multidimensional clustering clusters to obtain all the moments belonging to the target multidimensional clustering cluster includes: In response to the belonging degree of any moment not in the multidimensional clustering cluster in the target multidimensional clustering cluster being greater than a preset belonging threshold, and the belonging degree at the moment not in the multidimensional clustering cluster is the maximum among the belonging degrees of all multidimensional clustering clusters, the moment is classified into the target multidimensional clustering cluster; In response to the fact that the degree of belonging of any moment not in the multidimensional clustering cluster to all multidimensional clustering clusters is not greater than the preset belonging threshold, the moment is recorded as the moment to be divided, and then all the moments to be divided are combined into a multidimensional clustering cluster.

6. The method for real-time monitoring of carbon emissions according to claim 1, characterized in that: The abnormality degree satisfies the following relationship: ； In the formula, For the The abnormality of the multidimensional clusters, For the The number of moments within a multidimensional clustering cluster, For the All the time in the multidimensional clustering clusters to the The average Euclidean distance between the cluster centers of multidimensional clusters, For the All the time in the multidimensional clustering clusters to the The Euclidean distance of the cluster centers of multidimensional clusters relative to The average dispersion of For the Multidimensional clustering cluster The moment to The Euclidean distance between the cluster centers of multidimensional clusters, is the natural exponential function, is the standard normalization function.

7. The method for real-time monitoring of carbon emissions according to claim 1, characterized in that: The abnormal value satisfies the following relationship: ； In the formula, For the Multidimensional clustering cluster The outlier value at a moment, For the Multidimensional clustering cluster The moment to The Euclidean distance between the cluster centers of multidimensional clusters, For the All the time in the multidimensional clustering clusters to the The Euclidean distance of the cluster centers of multidimensional clusters relative to The average dispersion of For the The number of moments within a multidimensional clustering cluster, For the All the time in the multidimensional clustering clusters to the The average Euclidean distance between the cluster centers of multidimensional clusters, For the The abnormality of the multidimensional clusters, is the standard normalization function.

8. The method for real-time monitoring of carbon emissions according to claim 1, characterized in that: The real-time monitoring of carbon emissions is achieved by: In response to the abnormal value at the current moment being greater than the preset abnormal threshold, the carbon emission status at the current moment is determined to be abnormal, and an early warning prompt is issued to complete the real-time monitoring of carbon emissions.

9. Carbon emission real-time monitoring system, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the real-time carbon emission monitoring method according to any one of claims 1 to 8 is implemented.

Images