Regional carbon emission accounting system and method

By identifying and collecting emission source data in the region and using artificial intelligence analysis technology to predict carbon emissions, the problems of low carbon emission accounting accuracy and poor real-time performance in traditional methods are solved, and more efficient carbon emission management is achieved.

CN118377787BActive Publication Date: 2024-11-19BEIJING MUNICIPAL RES INST OF ENVIRONMENT PROTECTION
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410558322.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-08
Publication Date
2024-11-19
Estimated Expiration
2044-05-08

AI Technical Summary

Technical Problem

The traditional regional carbon emission accounting methods rely on statistical data and model calculations, resulting in low carbon emission estimation accuracy and inability to monitor and reflect carbon emission changes in real time, limiting the timely management and regulation of regional carbon emissions.

Method used

By determining the spatial range of carbon emission accounting, identifying emission sources, using sensor networks to collect greenhouse gas concentration and flow velocity data, performing data processing and analysis based on artificial intelligence, generating a greenhouse gas flow timing correlation matrix, extracting feature vectors, and using feature-based first The carbon emission accounting inferencer tests the distribution to make predictions.

Benefits of technology

It improves the accuracy, real-timeness and refinement of carbon emission accounting, and can timely manage and regulate regional carbon emissions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118377787B_ABST
    Figure CN118377787B_ABST
Patent Text Reader

Abstract

The present disclosure relates to a regional carbon emission accounting system and method. The method includes: identifying and listing emission sources that need to be accounted for within the spatial scope of carbon emission accounting; collecting the time series of greenhouse gas concentrations and the time series of greenhouse gas flow rates of each emission source through a sensor network; encoding the time series of greenhouse gas concentrations and the time series of greenhouse gas flow rates of each emission source based on the greenhouse gas concentration-flow rate time series association of the same emission source, and then extracting gas flow time series association features; and passing the set of greenhouse gas flow time series association feature vectors through a carbon emission accounting reasoner based on feature prior distribution to obtain a carbon emission reasoning semantic feature vector, and then determine the annual carbon emission forecast value. In this way, the accuracy, real-time and refinement of carbon emission accounting can be improved, so as to facilitate timely management and regulation of regional carbon emissions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the field of carbon emission technology, and in particular, to a regional carbon emission accounting system and method. Background Art

[0002] As the global climate change problem becomes increasingly serious, it is particularly important to accurately account for and monitor carbon emissions. Regional carbon emissions accounting refers to the process of accurately measuring and evaluating carbon emissions from various emission sources in a specific region. This process is crucial for monitoring and managing carbon emissions in the region, formulating emission reduction policies, and achieving carbon neutrality goals.

[0003] However, traditional regional carbon emission accounting schemes usually rely on statistical data and model extrapolations, which may contain uncertainties and errors, resulting in low accuracy in carbon emission estimates. Specifically, traditional accounting methods often require periodic surveys of historical carbon emission data from various emission sources, which are updated slowly and cannot monitor and reflect changes in carbon emissions in real time. In other words, traditional methods usually require a long data collection and processing cycle, cannot monitor and respond to changes in carbon emissions in real time, and limit the ability to manage and regulate regional carbon emissions in a timely manner.

[0004] Therefore, an optimized regional carbon emissions accounting scheme is desired. Summary of the invention

[0005] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0006] In a first aspect, the present disclosure provides a method for calculating regional carbon emissions, the method comprising:

[0007] Determine the spatial scope of carbon emissions accounting;

[0008] Identify and list the emission sources that need to be accounted for within the spatial scope of the carbon emission accounting;

[0009] Collecting the time series of greenhouse gas concentration and greenhouse gas flow rate of each emission source through a sensor network;

[0010] Performing greenhouse gas concentration-flow rate time series correlation coding based on the same emission source on the time series of greenhouse gas concentration and the time series of greenhouse gas flow rate of each emission source to obtain a set of greenhouse gas flow rate time series correlation matrices;

[0011] Extracting gas flow time series correlation features from each greenhouse gas flow time series correlation matrix in the set of greenhouse gas flow time series correlation matrices to obtain a set of greenhouse gas flow time series correlation feature vectors;

[0012] Passing the set of greenhouse gas flow time series associated feature vectors through a carbon emission accounting reasoner based on feature prior distribution to obtain a carbon emission reasoning semantic feature vector;

[0013] Based on the carbon emission reasoning semantic feature vector, an annual carbon emission prediction value is determined.

[0014] Optionally, the time series of greenhouse gas concentrations and the time series of greenhouse gas flow rates of the various emission sources are coded based on the greenhouse gas concentration-flow rate time series association of the same emission source to obtain a set of greenhouse gas flow time series association matrices, including: arranging the time series of greenhouse gas concentrations and the time series of greenhouse gas flow rates of the various emission sources as input vectors according to the time dimension to obtain a set of {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector}; and performing association coding on each {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector} in the set of {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector} to obtain a set of greenhouse gas flow time series association matrices.

[0015] Optionally, gas flow timing correlation features are extracted from each greenhouse gas flow timing correlation matrix in the set of greenhouse gas flow timing correlation matrices to obtain a set of greenhouse gas flow timing correlation feature vectors, including: passing the set of greenhouse gas flow timing correlation matrices through a greenhouse gas flow timing correlation feature extractor based on a void convolutional neural network model to obtain a set of greenhouse gas flow timing correlation feature vectors.

[0016] Optionally, the set of greenhouse gas flow time series associated feature vectors is passed through a carbon emission accounting inferencer based on feature prior distribution to obtain a carbon emission reasoning semantic feature vector, including: sorting the set of greenhouse gas flow time series associated feature vectors according to the flow of the emission source from large to small to obtain a sequence of greenhouse gas flow time series associated feature vectors; calculating the similarity between every two greenhouse gas flow time series associated feature vectors in the sequence of greenhouse gas flow time series associated feature vectors to obtain a similarity vector composed of multiple similarities; calculating the number of feature vectors between every two greenhouse gas flow time series associated feature vectors to obtain a quantity counting vector composed of multiple counting quantities; respectively calculating the product of the similarity of each position in the similarity vector and the counting quantity of the corresponding position in the quantity counting vector as a weighting coefficient to obtain a weighting coefficient vector composed of multiple weighting coefficients; using the weighting coefficient of each position in the weighting coefficient vector as a weight, calculating the weighted average sum of multiple greenhouse gas flow time series associated feature vectors in the sequence of greenhouse gas flow time series associated feature vectors to obtain the carbon emission reasoning semantic feature vector.

[0017] Optionally, determining an annual carbon emission prediction value based on the carbon emission reasoning semantic feature vector includes: passing the carbon emission reasoning semantic feature vector through an annual carbon emission predictor based on a decoder to obtain an annual carbon emission prediction value.

[0018] Optionally, a training step is also included: for training the greenhouse gas flow time series correlation feature extractor based on the void convolutional neural network model, the carbon emission accounting reasoner based on feature prior distribution, and the annual carbon emission predictor based on the decoder.

[0019] Optionally, the training step includes: determining the spatial scope of training carbon emission accounting; identifying and listing the training emission sources that need to be accounted for within the spatial scope of the training carbon emission accounting; collecting the training greenhouse gas concentration time series and the training greenhouse gas flow rate time series of each training emission source through the sensor network; performing greenhouse gas concentration-flow rate time series association encoding based on the same emission source on the training greenhouse gas concentration time series and the training greenhouse gas flow rate time series of each training emission source to obtain a set of training greenhouse gas flow time series association matrices; performing gas flow time series association feature encoding on each training greenhouse gas flow time series association matrix in the set of training greenhouse gas flow time series association matrices. Extract to obtain a set of training greenhouse gas flow time series correlation feature vectors; pass the set of training greenhouse gas flow time series correlation feature vectors through the carbon emission accounting reasoner based on feature prior distribution to obtain training carbon emission reasoning semantic feature vectors; pass the training carbon emission reasoning semantic feature vectors through the decoder-based annual carbon emission predictor to obtain a decoding loss function value; based on the decoding loss function value, train the greenhouse gas flow time series correlation feature extractor based on the void convolutional neural network model, the carbon emission accounting reasoner based on feature prior distribution and the decoder-based annual carbon emission predictor, and optimize the training carbon emission reasoning semantic feature vector in each iterative training.

[0020] In a second aspect, the present disclosure provides a regional carbon emission accounting system, the system comprising:

[0021] The spatial scope determination module is used to determine the spatial scope of carbon emission accounting;

[0022] An emission source identification module, used to identify and list emission sources that need to be accounted for within the spatial scope of the carbon emission accounting;

[0023] A greenhouse gas collection module, used to collect the time series of greenhouse gas concentration and greenhouse gas flow rate of each emission source through a sensor network;

[0024] A greenhouse gas concentration-flow rate time series correlation coding module, used for performing greenhouse gas concentration-flow rate time series correlation coding based on the same emission source on the greenhouse gas concentration time series and the greenhouse gas flow rate time series of each emission source to obtain a set of greenhouse gas flow rate time series correlation matrices;

[0025] A gas flow time series correlation feature extraction module, used to extract gas flow time series correlation features from each greenhouse gas flow time series correlation matrix in the set of greenhouse gas flow time series correlation matrices to obtain a set of greenhouse gas flow time series correlation feature vectors;

[0026] A carbon emission accounting reasoning module, used for passing the set of greenhouse gas flow time series associated feature vectors through a carbon emission accounting reasoner based on feature prior distribution to obtain a carbon emission reasoning semantic feature vector;

[0027] The annual carbon emission prediction value determination module is used to determine the annual carbon emission prediction value based on the carbon emission reasoning semantic feature vector.

[0028] Optionally, the greenhouse gas concentration-flow rate time series association coding module is used to: arrange the time series of greenhouse gas concentration and the time series of greenhouse gas flow rate of each emission source as input vectors according to the time dimension to obtain a set of {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector}; and perform association coding on each {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector} in the set of {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector} to obtain a set of greenhouse gas flow time series association matrices.

[0029] Optionally, the gas flow timing correlation feature extraction module is used to: pass the set of greenhouse gas flow timing correlation matrices through a greenhouse gas flow timing correlation feature extractor based on a void convolutional neural network model to obtain a set of greenhouse gas flow timing correlation feature vectors.

[0030] The above technical solution is adopted to identify and list the emission sources that need to be accounted for within the spatial scope of carbon emission accounting; collect the time series of greenhouse gas concentration and the time series of greenhouse gas flow rate of each emission source through the sensor network; perform greenhouse gas concentration-flow rate time series association encoding based on the same emission source for the time series of greenhouse gas concentration and the time series of greenhouse gas flow rate, and then extract the gas flow time series association feature; and pass the set of greenhouse gas flow time series association feature vectors through the carbon emission accounting reasoner based on feature prior distribution to obtain the carbon emission reasoning semantic feature vector, and then determine the annual carbon emission forecast value. In this way, the accuracy, real-time and refinement of carbon emission accounting can be improved, so as to facilitate the timely management and regulation of regional carbon emissions.

[0031] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale. In the drawings:

[0033] Figure 1 The present invention is a flowchart of a method for calculating regional carbon emissions according to an exemplary embodiment.

[0034] Figure 2 The figure is a block diagram of a regional carbon emission accounting system according to an exemplary embodiment.

[0035] Figure 3 It is a block diagram of an electronic device according to an exemplary embodiment.

[0036] Figure 4 It is an application scenario diagram of a regional carbon emission accounting method according to an exemplary embodiment. DETAILED DESCRIPTION

[0037] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0038] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, 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.

[0039] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0040] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0041] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0042] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0043] In order to solve the above problems, the present disclosure provides a regional carbon emission accounting system and method, which identifies and lists the emission sources that need to be accounted for within the spatial scope of carbon emission accounting; collects the time series of greenhouse gas concentration and the time series of greenhouse gas flow rate of each emission source through a sensor network; performs greenhouse gas concentration-flow rate time series association encoding based on the same emission source for the time series of greenhouse gas concentration and the time series of greenhouse gas flow rate of each emission source, and then extracts the gas flow time series association feature; and passes the set of greenhouse gas flow time series association feature vectors through a carbon emission accounting reasoner based on feature prior distribution to obtain a carbon emission reasoning semantic feature vector, and then determines the annual carbon emission forecast value. In this way, the accuracy, real-time and refinement of carbon emission accounting can be improved, so as to facilitate timely management and regulation of regional carbon emissions.

[0044] The specific embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0045] Figure 1 is a flow chart of a method for calculating regional carbon emissions according to an exemplary embodiment. Figure 1 As shown, the method includes:

[0046] Step 101, determining the spatial scope of carbon emission accounting;

[0047] Step 102: Identify and list emission sources that need to be accounted for within the spatial scope of the carbon emission accounting;

[0048] Step 103: collecting the time series of greenhouse gas concentration and the time series of greenhouse gas flow rate of each emission source through a sensor network;

[0049] Step 104: performing greenhouse gas concentration-flow rate time series correlation coding based on the same emission source on the time series of greenhouse gas concentration and the time series of greenhouse gas flow rate of each emission source to obtain a set of greenhouse gas flow rate time series correlation matrices;

[0050] Step 105, extracting gas flow time series correlation features from each greenhouse gas flow time series correlation matrix in the set of greenhouse gas flow time series correlation matrices to obtain a set of greenhouse gas flow time series correlation feature vectors;

[0051] Step 106: Pass the set of greenhouse gas flow time series associated feature vectors through a carbon emission accounting reasoner based on feature prior distribution to obtain a carbon emission reasoning semantic feature vector;

[0052] Step 107: Determine an annual carbon emission prediction value based on the carbon emission reasoning semantic feature vector.

[0053] In response to the above technical problems, the technical concept of this application is to determine the spatial scope of carbon emission accounting and identify various emission sources in the process of regional carbon emission accounting, so as to monitor and collect greenhouse gas concentration and flow rate data of each emission source in real time through a sensor network. In this way, it is possible to introduce artificial intelligence-based data processing and analysis algorithms at the back end to perform time-series collaborative correlation analysis of greenhouse gas concentration and flow rate data of multiple emission sources, so as to predict the carbon emission reasoning semantics in the region, and thus predict the annual carbon emissions of the region. In this way, the annual carbon emissions of the region can be automatically predicted based on the time-series collaborative correlation information of greenhouse gas concentration and flow rate of multiple emission sources in a specific region, thereby improving the accuracy, real-time and refinement of carbon emission accounting, so as to facilitate timely management and regulation of regional carbon emissions.

[0054] Specifically, in the technical solution of the present application, first, the spatial scope of carbon emission accounting is determined, and the emission sources that need to be accounted for within the spatial scope of the carbon emission accounting are identified and listed. Further, the time series of greenhouse gas concentrations and the time series of greenhouse gas flow rates of the respective emission sources are collected through the sensor network. Next, considering that the greenhouse gas concentrations and greenhouse gas flow rates of the respective emission sources have dynamic characteristics of time series in the time dimension, therefore, in order to better capture the trend and law of greenhouse gas carbon emissions changing over time, in the technical solution of the present application, the time series of greenhouse gas concentrations and the time series of greenhouse gas flow rates of the respective emission sources need to be arranged as input vectors according to the time dimension to obtain a set of {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector}. By arranging the time series of greenhouse gas concentrations and flow rates as input vectors according to the time dimension, it is beneficial to retain the time series change information of these data in the time dimension. Doing so can better capture the trend and law of greenhouse gas carbon emissions changing over time, and provide a more accurate data basis for subsequent analysis and prediction.

[0055] In one embodiment of the present disclosure, the time series of greenhouse gas concentrations and the time series of greenhouse gas flow rates of the various emission sources are coded based on the greenhouse gas concentration-flow rate time series association of the same emission source to obtain a set of greenhouse gas flow time series association matrices, including: arranging the time series of greenhouse gas concentrations and the time series of greenhouse gas flow rates of the various emission sources as input vectors according to the time dimension to obtain a set of {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector}; and performing association coding on each {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector} in the set of {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector} to obtain a set of greenhouse gas flow time series association matrices.

[0056] Then, considering that the greenhouse gas concentration and greenhouse gas flow rate of each emission source not only have the time series dynamic change characteristic information, but also there is a time series synergistic correlation between the gas concentration and the flow rate, this correlation plays an important role in inferring the carbon emissions of different emission sources. Based on this, in the technical solution of the present application, in order to better analyze the time series synergistic correlation between greenhouse gas concentration and flow rate, so as to more accurately infer the carbon emissions of different emission sources and the region, in the technical solution of the present application, each {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector} in the set of {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector} is further associated and coded to obtain a set of greenhouse gas flow time series association matrices. By associating the greenhouse gas concentration time series input vector and the greenhouse gas flow rate time series input vector of each emission source, it can help capture the time series dynamic correlation and synergistic relationship between the greenhouse gas concentration and flow rate of each emission source, and provide more useful information for subsequent analysis and carbon emissions prediction, thereby improving the accuracy and effectiveness of regional carbon emissions accounting.

[0057] Furthermore, considering that each emission source contributes when calculating the greenhouse gas carbon emissions in the region. In order to capture the temporal synergistic correlation between the greenhouse gas concentration and flow rate of each emission source, so as to accurately capture the greenhouse gas carbon emission temporal characteristics of each emission source, thereby improving the accuracy of the carbon emissions prediction in the region, in the technical solution of the present application, the set of greenhouse gas flow temporal correlation matrices is passed through a greenhouse gas flow temporal correlation feature extractor based on a void convolutional neural network model to obtain a set of greenhouse gas flow temporal correlation feature vectors. It should be understood that the void convolutional neural network model is a variant of the convolutional neural network model. By introducing holes in the convolution kernel to increase the receptive field, it can effectively capture the long-range dependencies and correlation features in the time series data of greenhouse gas concentration and flow rate, and help to extract the key features in the greenhouse gas flow temporal correlation matrix, so as to better capture the global temporal synergistic correlation feature information between the greenhouse gas concentration and flow rate of each emission source.

[0058] In one embodiment of the present disclosure, gas flow timing correlation features are extracted from each greenhouse gas flow timing correlation matrix in the set of greenhouse gas flow timing correlation matrices to obtain a set of greenhouse gas flow timing correlation feature vectors, including: passing the set of greenhouse gas flow timing correlation matrices through a greenhouse gas flow timing correlation feature extractor based on a void convolutional neural network model to obtain the set of greenhouse gas flow timing correlation feature vectors.

[0059] It should be understood that, since there is a correlation relationship between the greenhouse gas flow time series characteristics of each emission source in the spatial scope of carbon emission accounting, this correlation relationship is of great significance for the prediction of total carbon emissions within the spatial scope of carbon emission accounting. Therefore, in order to better understand the carbon emission information contained in the greenhouse gas emission data of different emission sources, and to more accurately estimate the total carbon emissions based on the carbon emission information and correlation relationship of these emission sources, and realize carbon emission accounting, in the technical solution of the present application, the set of greenhouse gas flow time series associated feature vectors is further passed through a carbon emission accounting reasoner based on feature prior distribution to obtain a carbon emission reasoning semantic feature vector. In particular, the carbon emission accounting reasoner based on feature prior distribution can use the correlation and synergy between the greenhouse gas flow time series characteristics of each emission source as prior knowledge to more accurately perform carbon emission accounting reasoning in the area. That is to say, through the processing of the carbon emission accounting reasoner, the implicit relationship between the greenhouse gas flow time series characteristics of each emission source and the emission source flow characteristics can be captured as prior knowledge to enhance the understanding of the carbon emissions of each emission source and the carbon emission time series correlation characteristics of multiple emission sources, thereby comprehensively characterizing the greenhouse gas carbon emissions in the region, providing more effective support for carbon emission accounting and management, and improving the accuracy and reliability of emission reasoning and prediction in the region.

[0060] In one embodiment of the present disclosure, the set of greenhouse gas flow time series associated feature vectors is passed through a carbon emission accounting inferencer based on feature prior distribution to obtain a carbon emission reasoning semantic feature vector, including: sorting the set of greenhouse gas flow time series associated feature vectors from large to small according to the flow of the emission source to obtain a sequence of greenhouse gas flow time series associated feature vectors; calculating the similarity between every two greenhouse gas flow time series associated feature vectors in the sequence of greenhouse gas flow time series associated feature vectors to obtain a similarity vector composed of multiple similarities; calculating the number of feature vectors separated between every two greenhouse gas flow time series associated feature vectors to obtain a quantity counting vector composed of multiple counting quantities; respectively calculating the product of the similarity of each position in the similarity vector and the counting quantity of the corresponding position in the quantity counting vector as a weighting coefficient to obtain a weighting coefficient vector composed of multiple weighting coefficients; using the weighting coefficient of each position in the weighting coefficient vector as a weight, calculating the weighted average sum of multiple greenhouse gas flow time series associated feature vectors in the sequence of greenhouse gas flow time series associated feature vectors to obtain the carbon emission reasoning semantic feature vector.

[0061] In particular, in a specific example of the present application, the set of greenhouse gas flow time series associated feature vectors is sorted from large to small according to the flow of the emission source to obtain a sequence of greenhouse gas flow time series associated feature vectors; the sequence of greenhouse gas flow time series associated feature vectors is processed by the carbon emission accounting reasoner based on feature prior distribution with the following carbon emission accounting reasoning formula to obtain the carbon emission reasoning semantic feature vector;

[0062] The carbon emission accounting inference formula is:

[0063]

[0064] Among them, x i and x j represents the i-th and j-th greenhouse gas flow time series correlation feature vectors in the sequence of greenhouse gas flow time series correlation feature vectors, respectively, g(x j ) represents x j With x i The number of eigenvectors between them, f(x i ,x j ) represents x j With x i Similarity between them, M represents the total number of feature vectors in the sequence of the greenhouse gas flow time series associated feature vectors, n is equal to the numerical value of M, and V represents the carbon emission reasoning semantic feature vector.

[0065] Then, the carbon emission reasoning semantic feature vector is passed through the decoder-based annual carbon emission predictor to obtain the annual carbon emission prediction value. In other words, the decoding regression is performed using the comprehensive representation features of the multi-emission source carbon emission reasoning semantics within the spatial range of carbon emission accounting to predict the annual carbon emissions of the region. In this way, the annual carbon emissions of the region can be automatically predicted based on the time-series collaborative correlation information of greenhouse gas concentrations and flow rates of multiple emission sources in a specific area, thereby improving the accuracy, real-time and refinement of carbon emission accounting, so as to facilitate timely management and regulation of regional carbon emissions.

[0066] In one embodiment of the present disclosure, based on the carbon emission reasoning semantic feature vector, determining the annual carbon emission prediction value includes: passing the carbon emission reasoning semantic feature vector through a decoder-based annual carbon emission predictor to obtain the annual carbon emission prediction value.

[0067] Furthermore, in one embodiment of the present disclosure, the regional carbon emission accounting method also includes a training step: for training the greenhouse gas flow time series correlation feature extractor based on the void convolutional neural network model, the carbon emission accounting reasoner based on feature prior distribution, and the decoder-based annual carbon emissions predictor. The training step includes determining the spatial scope of the training carbon emission accounting; identifying and listing the training emission sources that need to be accounted for within the spatial scope of the training carbon emission accounting; collecting the training greenhouse gas concentration time series and the training greenhouse gas flow rate time series of each training emission source through the sensor network; performing greenhouse gas concentration-flow rate time series correlation encoding based on the same emission source on the training greenhouse gas concentration time series and the training greenhouse gas flow rate time series of each training emission source to obtain a set of training greenhouse gas flow time series correlation matrices; performing gas flow time series correlation feature extraction on each training greenhouse gas flow time series correlation matrix in the set of training greenhouse gas flow time series correlation matrices. Obtain a set of training greenhouse gas flow time series correlation feature vectors; pass the set of training greenhouse gas flow time series correlation feature vectors through the carbon emission accounting reasoner based on feature prior distribution to obtain training carbon emission reasoning semantic feature vectors; pass the training carbon emission reasoning semantic feature vectors through the decoder-based annual carbon emission predictor to obtain a decoding loss function value; based on the decoding loss function value, train the greenhouse gas flow time series correlation feature extractor based on the void convolutional neural network model, the carbon emission accounting reasoner based on feature prior distribution and the decoder-based annual carbon emission predictor, and optimize the training carbon emission reasoning semantic feature vectors in each iterative training.

[0068] In the technical solution of the present application, each training greenhouse gas flow time series association feature vector in the set of training greenhouse gas flow time series association feature vectors respectively represents the semantic association features of the training temperature gas flow rate-training temperature gas concentration of each training emission source in the local time domain space determined based on the convolution kernel. Considering that each training greenhouse gas flow time series association matrix in the set of training greenhouse gas flow time series association matrices has significant differences in the time series distribution in the data source domain, each training greenhouse gas flow time series association feature vector in the set of training greenhouse gas flow time series association feature vectors has significant characteristic distribution distinctiveness. In the process of passing the set of training greenhouse gas flow time series correlation feature vectors through the carbon emission accounting reasoner based on feature prior distribution, the carbon emission accounting reasoner based on feature prior distribution uses the prior distribution law of training greenhouse gas flow of different training emission sources in the time domain space to conduct feature distribution conduction so as to transfer the time series correlation characteristics of training greenhouse gas flow of different training emission sources to the carbon emission reasoning semantics. The training carbon emission reasoning semantic feature vector obtained in this way has a complex time series semantic feature correlation representation dimension of the interaction of time domain characteristics of training greenhouse gas flow of multiple training emission sources within the spatial scope of carbon emission accounting. Therefore, it is also expected to improve the feature expression effect under the complex time series semantic feature correlation representation dimension during the model iteration process.

[0069] Based on this, the applicant of the present application optimizes the training carbon emission reasoning semantic feature vector, for example, V, during each iteration of the model, which is expressed as: in each iteration training, the training carbon emission reasoning semantic feature vector is optimized using the following optimization formula to obtain the optimized training carbon emission reasoning semantic feature vector; wherein the optimization formula is:

[0070]

[0071] Among them, μ, σ and v max They are the feature sets v i ∈V, log is the logarithmic function with base 2, and α is the weight hyperparameter, V is the training carbon emission reasoning semantic feature vector, V′ is the optimized training carbon emission reasoning semantic feature vector, v max -1 is the feature set v i ∈V is the reciprocal of the maximum value, ⊕ represents positional addition, θ represents positional subtraction, and ⊙ represents positional dot multiplication.

[0072] Here, we use an interactive representation containing probability statistics features represented by mean μ and standard value σ and v as a latent variable feature. i and v maxThe short sequence of the distribution interaction representation is used as the sub-manifold potential motif under the complex manifold network of the training carbon emission reasoning semantic feature vector V, and the obtained potential motif feature information pattern and feature distribution pattern of the training carbon emission reasoning semantic feature vector V are used as its global structure inference unit, so as to reconstruct the complex manifold structure of the training carbon emission reasoning semantic feature vector V in the form of a global structure potential motif dictionary based on connection, so as to improve the generation and evolution understanding ability of the model in the iterative process for the manifold structure corresponding to the image semantic association feature of the dimension represented by the complex temporal semantic feature association, and improve the feature expression effect under the dimension represented by the complex temporal semantic feature association in the model iteration process, so as to improve the accuracy of the annual carbon emission prediction value obtained by the training carbon emission reasoning semantic feature vector through the decoder-based annual carbon emission predictor. In this way, the annual carbon emission prediction within the carbon emission accounting space can be more accurately carried out, thereby improving the accuracy, real-time and refinement of carbon emission accounting, so as to facilitate the timely management and regulation of regional carbon emissions.

[0073] In summary, the above scheme is adopted. In the process of regional carbon emission accounting, the spatial scope of carbon emission accounting is determined, and various emission sources are identified, so that the greenhouse gas concentration and flow rate data of each emission source can be collected in real time through the sensor network. In this way, the data processing and analysis algorithm based on artificial intelligence can be introduced in the back end to perform the time series collaborative correlation analysis of greenhouse gas concentration and flow rate data of multiple emission sources, so as to predict the carbon emission reasoning semantics in the region, and thus predict the annual carbon emissions of the region. In this way, the annual carbon emissions of the region can be automatically predicted based on the time series collaborative correlation information of greenhouse gas concentration and flow rate of multiple emission sources in a specific region, thereby improving the accuracy, real-time and refinement of carbon emission accounting, so as to facilitate the timely management and regulation of regional carbon emissions.

[0074] Figure 2 FIG. 1 is a block diagram of a regional carbon emission accounting system according to an exemplary embodiment. Figure 2 As shown, the system 200 includes:

[0075] A spatial scope determination module 201 is used to determine the spatial scope of carbon emission accounting;

[0076] An emission source identification module 202 is used to identify and list emission sources that need to be accounted for within the spatial scope of the carbon emission accounting;

[0077] A greenhouse gas collection module 203 is used to collect the time series of greenhouse gas concentration and the time series of greenhouse gas flow rate of each emission source through a sensor network;

[0078] A greenhouse gas concentration-flow rate time series correlation coding module 204 is used to perform greenhouse gas concentration-flow rate time series correlation coding based on the same emission source on the greenhouse gas concentration time series and the greenhouse gas flow rate time series of each emission source to obtain a set of greenhouse gas flow rate time series correlation matrices;

[0079] The gas flow time series correlation feature extraction module 205 is used to extract the gas flow time series correlation feature of each greenhouse gas flow time series correlation matrix in the set of greenhouse gas flow time series correlation matrices to obtain a set of greenhouse gas flow time series correlation feature vectors;

[0080] A carbon emission accounting reasoning module 206, configured to pass the set of greenhouse gas flow time series associated feature vectors through a carbon emission accounting reasoner based on feature prior distribution to obtain a carbon emission reasoning semantic feature vector;

[0081] The annual carbon emission prediction value determination module 207 is used to determine the annual carbon emission prediction value based on the carbon emission reasoning semantic feature vector.

[0082] In one embodiment of the present disclosure, the greenhouse gas concentration-flow rate time series association coding module is used to: arrange the time series of greenhouse gas concentration and the time series of greenhouse gas flow rate of each emission source as input vectors according to the time dimension to obtain a set of {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector}; and perform association coding on each {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector} in the set of {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector} to obtain a set of greenhouse gas flow time series association matrices.

[0083] In one embodiment of the present disclosure, the gas flow time series correlation feature extraction module is used to: pass the set of the greenhouse gas flow time series correlation matrices through a greenhouse gas flow time series correlation feature extractor based on a void convolutional neural network model to obtain a set of the greenhouse gas flow time series correlation feature vectors.

[0084] Reference below Figure 3 , which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing the embodiment of the present disclosure. The terminal device in the embodiment of the present disclosure may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle terminals (such as vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0085] like Figure 3 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0086] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0087] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0088] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0089] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0090] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0091] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0092] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0093] The modules involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of a module does not limit the module itself in some cases. For example, a test parameter acquisition module may also be described as a "module for acquiring device test parameters corresponding to a target device".

[0094] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0095] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0096] Figure 4 FIG. 1 is an application scenario diagram of a regional carbon emission accounting method according to an exemplary embodiment. Figure 4 As shown in the application scenario, first, the spatial scope of carbon emission accounting is determined (for example, Figure 4 Identify and list the emission sources that need to be accounted for within the spatial scope of the carbon emission accounting (e.g. Figure 4 Then, the obtained spatial range and emission sources are input into a server where a regional carbon emission accounting algorithm is deployed (for example, Figure 4 In S) shown in , the server is capable of processing the spatial range and the emission source based on a regional carbon emission accounting algorithm to determine an annual carbon emission forecast value.

[0097] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0098] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0099] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims. Regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.

Claims

1. A regional carbon emissions accounting method, characterized in that: include: Determine the spatial scope of carbon emissions accounting; Identify and list the emission sources that need to be accounted for within the spatial scope of the carbon emission accounting; Collect the time series of greenhouse gas concentration and greenhouse gas flow rate of each emission source through the sensor network; Performing greenhouse gas concentration-flow rate time series correlation coding based on the same emission source on the time series of greenhouse gas concentration and the time series of greenhouse gas flow rate of each emission source to obtain a set of greenhouse gas flow rate time series correlation matrices; Extracting gas flow time series correlation features from each greenhouse gas flow time series correlation matrix in the set of greenhouse gas flow time series correlation matrices to obtain a set of greenhouse gas flow time series correlation feature vectors; Passing the set of greenhouse gas flow time series associated feature vectors through a carbon emission accounting reasoner based on feature prior distribution to obtain a carbon emission reasoning semantic feature vector; Determining an annual carbon emission prediction value based on the carbon emission reasoning semantic feature vector; The set of greenhouse gas flow time series associated feature vectors is passed through a carbon emission accounting reasoner based on feature prior distribution to obtain a carbon emission reasoning semantic feature vector, including: Sorting the set of greenhouse gas flow time series correlation feature vectors according to the flow of emission sources from large to small to obtain a sequence of greenhouse gas flow time series correlation feature vectors; Calculating the similarity between every two greenhouse gas flow time series correlation feature vectors in the sequence of greenhouse gas flow time series correlation feature vectors to obtain a similarity vector composed of multiple similarities; Calculating the number of characteristic vectors between each two greenhouse gas flow time series associated characteristic vectors to obtain a quantity counting vector composed of a plurality of counting quantities; Calculating the product of the similarity of each position in the similarity vector and the count quantity of the corresponding position in the quantity count vector respectively as a weighting coefficient to obtain a weighting coefficient vector composed of a plurality of weighting coefficients; The weighted coefficients of each position in the weighted coefficient vector are used as weights to calculate the weighted average sum of multiple greenhouse gas flow time series associated feature vectors in the sequence of greenhouse gas flow time series associated feature vectors to obtain the carbon emission reasoning semantic feature vector.

2. The regional carbon emission accounting method according to claim 1, characterized in that: The greenhouse gas concentration time series and the greenhouse gas flow rate time series of each emission source are coded based on the greenhouse gas concentration-flow rate time series association of the same emission source to obtain a set of greenhouse gas flow time series association matrices, including: Arranging the time series of greenhouse gas concentration and the time series of greenhouse gas flow rate of each emission source as input vectors according to the time dimension to obtain a set of {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector}; Association coding is performed on each {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector} in the set of {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector} to obtain the set of greenhouse gas flow time series association matrices.

3. The regional carbon emission accounting method according to claim 2 is characterized in that: Gas flow time series correlation features are extracted from each greenhouse gas flow time series correlation matrix in the set of greenhouse gas flow time series correlation matrices to obtain a set of greenhouse gas flow time series correlation feature vectors, including: passing the set of greenhouse gas flow time series correlation matrices through a greenhouse gas flow time series correlation feature extractor based on a void convolutional neural network model to obtain a set of greenhouse gas flow time series correlation feature vectors.

4. The regional carbon emission accounting method according to claim 3 is characterized in that: Based on the carbon emission reasoning semantic feature vector, determining an annual carbon emission prediction value includes: passing the carbon emission reasoning semantic feature vector through an annual carbon emission predictor based on a decoder to obtain an annual carbon emission prediction value.

5. The regional carbon emission accounting method according to claim 4 is characterized in that: It also includes a training step: for training the greenhouse gas flow time series correlation feature extractor based on the void convolutional neural network model, the carbon emission accounting reasoner based on feature prior distribution, and the annual carbon emission predictor based on the decoder.

6. The regional carbon emission accounting method according to claim 5, characterized in that: The training step comprises: Determine the spatial scope of training carbon emission accounting; Identify and list the training emission sources that need to be accounted for within the spatial scope of the training carbon emission accounting; Collecting the time series of training greenhouse gas concentration and training greenhouse gas flow rate of each training emission source through the sensor network; Performing greenhouse gas concentration-flow rate time series association coding based on the same emission source on the training greenhouse gas concentration time series and the training greenhouse gas flow rate time series of each training emission source to obtain a set of training greenhouse gas flow rate time series association matrices; Extracting gas flow time series correlation features from each training greenhouse gas flow time series correlation matrix in the set of training greenhouse gas flow time series correlation matrices to obtain a set of training greenhouse gas flow time series correlation feature vectors; Passing the set of training greenhouse gas flow time series associated feature vectors through the carbon emission accounting reasoner based on feature prior distribution to obtain a training carbon emission reasoning semantic feature vector; Passing the training carbon emission reasoning semantic feature vector through the decoder-based annual carbon emission predictor to obtain a decoding loss function value; Based on the decoding loss function value, the greenhouse gas flow time series correlation feature extractor based on the void convolutional neural network model, the carbon emission accounting reasoner based on feature prior distribution and the annual carbon emission predictor based on the decoder are trained, and in each iterative training, the training carbon emission reasoning semantic feature vector is optimized.

7. A regional carbon emission accounting system, characterized in that: include: space The scope determination module is used to determine the spatial scope of carbon emission accounting; An emission source identification module, used to identify and list emission sources that need to be accounted for within the spatial scope of the carbon emission accounting; A greenhouse gas collection module is used to collect the time series of greenhouse gas concentration and greenhouse gas flow rate of each emission source through a sensor network; A greenhouse gas concentration-flow rate time series correlation coding module, used for performing greenhouse gas concentration-flow rate time series correlation coding based on the same emission source on the greenhouse gas concentration time series and the greenhouse gas flow rate time series of each emission source to obtain a set of greenhouse gas flow rate time series correlation matrices; A gas flow time series correlation feature extraction module, used to extract gas flow time series correlation features from each greenhouse gas flow time series correlation matrix in the set of greenhouse gas flow time series correlation matrices to obtain a set of greenhouse gas flow time series correlation feature vectors; A carbon emission accounting reasoning module, used for passing the set of greenhouse gas flow time series associated feature vectors through a carbon emission accounting reasoner based on feature prior distribution to obtain a carbon emission reasoning semantic feature vector; An annual carbon emission prediction value determination module is used to determine the annual carbon emission prediction value based on the carbon emission reasoning semantic feature vector; The carbon emission accounting reasoning module is used to: Sorting the set of greenhouse gas flow time series correlation feature vectors according to the flow of emission sources from large to small to obtain a sequence of greenhouse gas flow time series correlation feature vectors; Calculating the similarity between every two greenhouse gas flow time series correlation feature vectors in the sequence of greenhouse gas flow time series correlation feature vectors to obtain a similarity vector composed of multiple similarities; Calculating the number of characteristic vectors between each two greenhouse gas flow time series associated characteristic vectors to obtain a quantity counting vector composed of a plurality of counting quantities; Calculating the product of the similarity of each position in the similarity vector and the count quantity of the corresponding position in the quantity count vector respectively as a weighting coefficient to obtain a weighting coefficient vector composed of a plurality of weighting coefficients; The weighted coefficients of each position in the weighted coefficient vector are used as weights to calculate the weighted average sum of multiple greenhouse gas flow time series associated feature vectors in the sequence of greenhouse gas flow time series associated feature vectors to obtain the carbon emission reasoning semantic feature vector.

8. The regional carbon emission accounting system according to claim 7, characterized in that: The greenhouse gas concentration-flow rate time series correlation coding module is used to: Arranging the time series of greenhouse gas concentration and the time series of greenhouse gas flow rate of each emission source as input vectors according to the time dimension to obtain a set of {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector}; Association coding is performed on each {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector} in the set of {greenhouse gas concentration time series input vector, greenhouse gas flow rate time series input vector} to obtain the set of greenhouse gas flow time series association matrices.

9. The regional carbon emission accounting system according to claim 8, characterized in that: The gas flow time series correlation feature extraction module is used to: pass the set of the greenhouse gas flow time series correlation matrices through a greenhouse gas flow time series correlation feature extractor based on a void convolutional neural network model to obtain a set of the greenhouse gas flow time series correlation feature vectors.