Emission right transaction management method

Abstract

The invention provides an emission right transaction management method, and the method comprises the steps: enabling a cross-region deployment node to respond to a received emission right transaction request, and obtaining pre-transaction accounting indexes corresponding to a seller enterprise and a buyer enterprise, the pollution discharge right transaction request is used for indicating the pollution discharge right transaction volume to be performed by the seller enterprise and the buyer enterprise and target pollutants; the cross-district deployment node inputs an emission right transaction volume indicated by the emission right transaction request and emission right transaction accounting indexes corresponding to the seller enterprise and the buyer enterprise respectively into an environmental impact evaluation model corresponding to the target pollutant, so as to obtain post-transaction accounting indexes corresponding to the seller enterprise and the buyer enterprise respectively; and the cross-district deployment node determines the capacity values of the target pollutants corresponding to the seller enterprise and the buyer enterprise after the transaction based on the post-transaction accounting indexes corresponding to the seller enterprise and the buyer enterprise so as to determine whether the emission right transaction request is supported or not.

Description

A management method for emissions trading Technical Field

[0001] This application relates to the field of emissions trading technology, and more specifically, to a management method for emissions trading. Background Technology

[0002] The current mainstream emissions trading system is mainly based on direct indicators such as emissions volume and emission concentration of emission sources for trading accounting, and quota transfer is completed through the trading platform.

[0003] In existing technologies, emissions trading accounting systems primarily rely on static information such as emissions monitoring data and trading application materials, mechanically verifying emissions by comparing the remaining emissions allowances of both parties with the trading volume. While this verification method ensures the formal compliance of the trading process, it only considers changes in direct emissions data of both parties and lacks assessment of dynamic factors such as potential changes in regional environmental quality and the impact of pollutant diffusion caused by the trading activities. This leads to significant biases in determining the environmental benefits of the trading. Summary of the Invention

[0004] The purpose of this application is to provide a management method for emissions trading, in order to solve the technical problem that existing emissions trading platforms have mechanical and simplistic assessment methods and lack consideration for environmental impact during the transaction calculation process, resulting in inaccurate assessments.

[0005] In a first aspect, the present invention provides a management method for emissions trading. The method includes an inter-regional allocation node responding to a received emissions trading request by obtaining pre-trade accounting indicators for the seller and buyer companies, respectively. The emissions trading request indicates the emissions trading volume and target pollutant to be traded by the seller and buyer companies. The inter-regional allocation node inputs the emissions trading volume indicated by the emissions trading request and the emissions trading accounting indicators for the seller and buyer companies into an environmental impact assessment model corresponding to the target pollutant to obtain post-trade accounting indicators for the seller and buyer companies. Based on the post-trade accounting indicators for the seller and buyer companies, the inter-regional allocation node determines the capacity value of the target pollutant for each of the seller and buyer companies after the transaction to determine whether to support the emissions trading request.

[0006] In optional implementations, the accounting indicators include at least the regional capacity benchmark value, regional spatial correction coefficient, pollutant spatiotemporal diffusion coefficient, pollutant type correction coefficient, and remaining capacity proportion coefficient. The environmental impact assessment model includes an input layer, a feature encoding layer, a regional interactive attention layer, a physical embedding layer, and an output layer. The input layer normalizes the emission rights trading volume and the emission rights trading accounting indicators corresponding to the seller and buyer companies, outputting a normalized vector. The feature encoding layer performs feature dimensionality upscaling and nonlinear transformation using a multilayer perceptron, outputting a high-dimensional vector. The regional interactive attention layer uses the emission rights trading volume as a query matrix and learns the sub-vectors corresponding to the seller and buyer companies in the high-dimensional vector to output an attention vector. The physical embedding layer determines the physical vector based on the attention vector and the pollutant diffusion sub-model. The output layer extracts the post-trade accounting indicators corresponding to the seller and buyer companies from the physical vector using an activation function.

[0007] In an optional implementation, the step of determining the physical vector based on the fusion vector and the pollutant diffusion sub-model specifically includes: constructing a physical prior vector based on the emissions trading volume and the emissions trading accounting indicators corresponding to the seller and buyer companies; and concatenating the physical prior vector with the attention vector to obtain the concatenated vector. The physical vector is obtained by performing a nonlinear transformation on the spliced ​​vector in the following manner. : ;in, To fuse the weight matrix, This is the bias vector.

[0008] In an optional implementation, the physical prior vector includes at least the pollutant concentration influence coefficient, mass conservation deviation value, attenuation coefficient, and auxiliary characteristic value corresponding to the seller and buyer companies, respectively.

[0009] In an optional implementation, the pollutant concentration influence coefficients corresponding to the seller / buyer are determined by the following method. : ; ; ;in, Emissions per unit time The average wind speed, , These are the diffusion parameters in the horizontal and vertical directions, respectively. For high-efficiency sources, This represents the corresponding spatiotemporal diffusion coefficient of the pollutants. This is the corresponding regional spatial correction coefficient.

[0010] In an optional implementation, the attenuation coefficients corresponding to the seller / buyer are determined by the following method. : ;in, is the attenuation constant.

[0011] In an optional implementation, the mass conservation deviation value is determined by the following method. : ; ; ;in, These are the correction factors for the pollutant types for the seller and buyer companies, respectively. For emissions trading volume, This is the diffusion loss value. This represents the settlement loss value.

[0012] Secondly, the present invention provides a management device for emissions trading, the device comprising: a response module, configured to respond to a received emissions trading request by acquiring pre-trade accounting indicators for the seller and buyer, wherein the emissions trading request indicates the emissions trading volume and target pollutant to be traded by the seller and buyer; a prediction module, configured to input the emissions trading volume indicated by the emissions trading request and the emissions trading accounting indicators for the seller and buyer into an environmental impact assessment model corresponding to the target pollutant, to obtain post-trade accounting indicators for the seller and buyer; and an accounting module, configured to determine the post-trade capacity value of the target pollutant for the seller and buyer based on the post-trade accounting indicators for the seller and buyer, to determine whether the emissions trading request is supported.

[0013] Thirdly, the present invention provides an electronic device, comprising: a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the steps of any of the pollution rights trading management methods described in the foregoing embodiments.

[0014] Fourthly, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of any of the pollution rights trading management methods described in the foregoing embodiments.

[0015] This application provides a management method for emissions trading. The method includes a cross-regional allocation node responding to a received emissions trading request by obtaining pre-trade accounting indicators for both the seller and buyer companies. The emissions trading request indicates the emissions trading volume and target pollutant for both companies. The cross-regional allocation node inputs the emissions trading volume indicated in the request, along with the emissions trading accounting indicators for both companies, into an environmental impact assessment model corresponding to the target pollutant to obtain post-trade accounting indicators for both companies. Based on these post-trade accounting indicators, the cross-regional allocation node determines the post-trade capacity value of the target pollutant for both companies to determine whether to support the emissions trading request. This application utilizes a deep learning model to comprehensively predict pre- and post-trade accounting indicators, thereby aiding in trading decisions. Furthermore, it incorporates an environmental impact assessment dimension into the trading decision-making process, making the emissions trading platform's assessment more accurate and improving the scientific rigor and environmental friendliness of emissions trading decisions. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 is a flowchart of a pollution rights trading management method provided in an embodiment of this application; Figure 2 is a structural schematic diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0018] The technical solutions in the embodiments of this application will now be described with reference to the accompanying drawings.

[0019] Example 1 Figure 1 is a schematic diagram of the structure of a pollution rights trading system provided in an embodiment of this application. As shown in Figure 1, in one embodiment of this application, a pollution rights trading system based on the Internet of Things is provided.

[0020] In one feasible implementation, IoT sensors and high-definition infrared cameras can be deployed at various locations, including sewage outlets, factory production workshops, areas surrounding treatment equipment, and sensitive areas. Regional environmental monitoring stations are established according to a 1km x 1km grid or administrative division to ensure coverage of all key control areas. Edge computing nodes are also configured to process high-frequency real-time monitoring data and reduce data transmission latency.

[0021] A data storage module was built, employing blockchain technology to ensure immutable data storage. AI auditing engines, dynamic control models, and other algorithmic models were deployed. Software development and system integration of each functional module were completed to ensure data interoperability between modules.

[0022] Environmental monitoring data and enterprise production emission data are collected in real time through IoT sensors, while the visual acquisition module acquires visual data such as the status of sewage outlets and equipment operation status through cameras. All data is automatically appended with metadata information and transmitted to the preprocessing module at a set frequency.

[0023] The data preprocessing module and the visual preprocessing module perform operations such as cleaning, deduplication, missing value completion, noise reduction, feature extraction, and fusion on the received data. The processed data undergoes a quality assessment; qualified data proceeds to the next stage, while unqualified data is returned for re-collection, triggering an alarm.

[0024] The spatiotemporal data aggregation module completes data aggregation processing according to the process of single-dimensional collection and extraction, single-dimensional fusion, core spatiotemporal correlation calculation, policy correction, result aggregation, and business-oriented output. It generates structured datasets through formula calculations.

[0025] The dynamic analysis module cleans and aligns the aggregated data, employs a composite model to predict total emission trends, extrapolate quota usage dynamics, and assess spatial risk distribution. Based on the comparison between the prediction results and capacity constraints, it generates early warning information and control recommendations.

[0026] The environmental capacity assessment module collects multi-dimensional basic data, and through steps such as single-dimensional fusion and multi-dimensional constraint fusion, it calculates the final environmental capacity of regional conventional pollutants and characteristic pollutants by weighted calculation, and generates an assessment report.

[0027] The emission rights quota verification module decomposes the total regional capacity to each enterprise based on the environmental capacity assessment results and spatiotemporal data aggregation, completes the scientific verification of enterprise quotas, and synchronizes the verification results to the intelligent total control module.

[0028] The intelligent total emission control module calculates the total regional emission volume in real time based on the input data from each module and using the total emission control formulas for conventional and characteristic pollutants. Based on the calculation results and combined with dynamic control coefficients, intelligent control of the total emission volume is achieved. When the quota usage reaches the warning threshold, an early warning is triggered and control recommendations are pushed out.

[0029] The intelligent cross-regional allocation module formulates cross-regional allocation plans based on the environmental capacity and quota usage of each region, combined with cross-regional emission correlation data. While ensuring that regional environmental capacity is not exceeded, it optimizes the allocation of pollution discharge rights, completes the transfer of quota ownership, and records it in the blockchain.

[0030] When a transaction request occurs in the system, the intelligent cross-regional allocation module can collect accounting indicators from both parties to the transaction, including but not limited to regional capacity benchmark value, regional spatial correction coefficient, pollutant spatiotemporal diffusion coefficient, pollutant type correction coefficient, and remaining capacity ratio coefficient.

[0031] Among them, the regional capacity benchmark value, regional spatial correction coefficient, pollutant type correction coefficient, and remaining capacity ratio coefficient can all be obtained from the corresponding environmental capacity assessment module.

[0032] The regional capacity benchmark value is the regional capacity benchmark (tons / year). The regional spatial correction factor can be calculated by multiplying the topographic factor and the ecological factor, with a value ranging from 0.4 to 1. The pollutant type correction factor is determined according to the type of pollutant to be traded; the pollutant type correction factor for conventional pollutants is between 1 and 1.2, and the pollutant type correction factor for characteristic pollutants is between 0.3 and 0.8. The remaining capacity percentage factor ranges from 0 to 1.

[0033] The spatiotemporal diffusion coefficient of pollutants can be obtained from the corresponding spatiotemporal data aggregation module and determined based on real-time meteorological and hydrological information, with a value range between 0.5 and 1.5.

[0034] In one embodiment of this application, an environmental impact assessment model for emissions trading based on physical information neural networks (PINN) is provided to predict the accounting indicators of both parties after the transaction, thereby helping to make transaction decisions.

[0035] The model input vector here can be:

[0036] ; ;in, These are, respectively, the seller's regional capacity benchmark value, regional spatial correction factor, pollutant spatiotemporal diffusion factor, pollutant type correction factor, and remaining capacity percentage factor. These are, respectively, the buyer's regional capacity benchmark value, regional spatial correction factor, pollutant spatiotemporal diffusion factor, pollutant type correction factor, and remaining capacity percentage factor. The target is the volume of emissions trading rights for pollutants.

[0037] The environmental impact assessment model includes an input layer, a feature encoding layer, a regional interaction attention layer, a physical embedding layer, and an output layer.

[0038] The input layer normalizes the emission rights trading volume and the emission rights trading accounting indicators corresponding to the seller and buyer companies, and outputs a normalized vector.

[0039] The input layer here standardizes the input features to form a 13-dimensional vector, which is then passed to the next layer.

[0040] The feature encoding layer performs feature dimensionality upscaling and nonlinear transformation through a multilayer perceptron, outputting a high-dimensional vector.

[0041] The feature encoding layer here can use a multilayer perceptron (MLP) to perform dimensionality upscaling and nonlinear transformations on the features. The first transformation converts the 13-dimensional vector to a 128-dimensional vector, and the second transformation converts the 128-dimensional vector to a 64-dimensional vector. This process can be represented as follows: ;in, It is the ReLU activation function. and These are the weight parameters and bias parameters to be trained.

[0042] The regional interaction attention layer uses the pollution rights trading volume as the query matrix and learns the sub-vectors corresponding to the seller and buyer companies in the high-dimensional vector to output the attention vector.

[0043] The regional interaction attention layer here will input... Split into sub-vectors corresponding to two regions , And learn the intrinsic relationship between the parameters of the two regions. This can be represented as: ; ; ; ; ; ;in, The three-dimensional tensor is formed by stacking the features of two regions. The first dimension is the batch, the second dimension is the region index, and the third dimension is the feature. For stacking operations, For the self-attention output of region A, take All batches, the 0th region, and all feature dimensions in the tensor. For the self-attention output of region B, take All batches, the first region, and all feature dimensions in the tensor. Let be the attention function. It is a flattening function (mapped to a one-dimensional vector). For self-attention output, This is for cross-attention output.

[0044] Here, by using the pollution rights trading volume as the query, cross-attention learning is performed on the seller / buyer sub-vectors to output attention vectors, thereby achieving interpretable modeling of spatial relationships. The attention weights can intuitively reflect the "environmental pressure transmission path of trading volume between the two regions", avoiding the situation where traditional models (such as fully connected networks) simply splice regional features, resulting in the neglect of spatial relationships.

[0045] The physical embedding layer determines the physical vector based on the attention vector and the pollutant diffusion sub-model.

[0046] For example, the steps to determine the physical vector based on the fusion vector and the pollutant diffusion sub-model include: constructing a physical prior vector based on the emission rights trading volume and the emission rights trading accounting indicators corresponding to the seller and buyer companies.

[0047] The physical prior vector includes at least the pollutant concentration influence coefficient, mass conservation deviation value, attenuation coefficient, and auxiliary characteristic value for the seller and buyer companies, respectively.

[0048] For example, the pollutant concentration impact coefficients corresponding to the seller / buyer can be determined in the following ways. : ; ; ;in, Emissions per unit time The average wind speed, , These are the diffusion parameters in the horizontal and vertical directions, respectively. For high-efficiency sources, This represents the corresponding spatiotemporal diffusion coefficient of the pollutants. This is the corresponding regional spatial correction coefficient.

[0049] Here, a simplified Gaussian plume model is used to calculate the dispersion concentration distribution of pollutants from the emission source to the downwind direction. Based on Reverse calculation, based on Reverse calculation. To adapt to neural network calculations, this model focuses on the concentration at 1 km downwind, along the centerline, and at the ground level, taking... (1km downwind) (Centerline) (ground).

[0050] diffusion parameters , Using the Pasquill-Gifford formula: ; Substitute It can be confirmed , .

[0051] The attenuation coefficients for the seller / buyer can be determined in the following ways. : ;in, is the attenuation constant.

[0052] Furthermore, the mass conservation deviation value can be determined in the following ways. : ; ; ;in, These are the correction factors for the pollutant types for the seller and buyer companies, respectively. For emissions trading volume, This is the diffusion loss value. This represents the settlement loss value.

[0053] Finally, a 16-dimensional vector can be obtained by combining these elements. .

[0054] in, The 11-dimensional auxiliary features can include normalized values ​​of diffusion parameters, wind speed, effective source height, and background concentration. , Settlement loss component, diffusion loss component, and cross-regional impact symmetry index.

[0055] Normalized influence coefficient The calculation method is as follows: .

[0056] The spatiotemporal decay characteristics can be calculated in the following way: .

[0057] in, is the attenuation constant.

[0058] The physics prior vector and the attention vector are concatenated to obtain the concatenated vector. .

[0059] The physical vector is obtained by performing a nonlinear transformation on the concatenated vector using the following method. : ;in, To fuse the weight matrix, This is the bias vector.

[0060] Here, by incorporating a physical model-driven approach and using physical prior vector embedding to overcome the "mechanism silence," environmental science laws are transformed into differentiable and learnable neural network constraints, making the prediction results closer to the actual scenario.

[0061] The output layer extracts the post-transaction accounting indicators for the seller and buyer companies from the physical vectors using activation functions.

[0062] The output layer includes a multi-task output head, specifically a region A prediction head. The calculation process can be represented as follows: ; ; ;

[0063] in, These are the parameters to be trained. for The Middle One portion, For batch processing; the activation function is configured as follows: ; ; .

[0064] The prediction head for region B is similar to that for region A, and can output the corresponding... .

[0065] The output layer also includes a co-exponential prediction header, which can be represented as: ; ; ; .

[0066] Here This is an environmental synergy index after a transaction between two regions; the closer it is to 1, the better the synergy.

[0067] Finally, the model output can be represented as .

[0068] Example 2 Figure 1 is a flowchart of a pollution rights trading management method provided in this application embodiment. As shown in Figure 1, the pollution rights trading management method provided in this application embodiment includes: S1, in response to a received pollution rights trading request, the inter-regional allocation node obtains the pre-trade accounting indicators corresponding to the seller and buyer companies respectively. The pollution rights trading request is used to indicate the pollution rights trading volume and target pollutant to be traded by the seller and buyer companies; S2, the inter-regional allocation node inputs the pollution rights trading volume indicated by the pollution rights trading request and the pollution rights trading accounting indicators corresponding to the seller and buyer companies respectively into the environmental impact assessment model corresponding to the target pollutant to obtain the post-trade accounting indicators corresponding to the seller and buyer companies respectively; S3, based on the post-trade accounting indicators corresponding to the seller and buyer companies respectively, the inter-regional allocation node determines the capacity value of the target pollutant corresponding to the seller and buyer companies respectively after the transaction to determine whether to support the pollution rights trading request.

[0069] Here, the capacity value of the target pollutant can be calculated in the following way. : .

[0070] If a certain area after the transaction If the threshold is exceeded, it may be recommended to terminate the transaction. If the coordination index is low, it may be recommended to optimize the transaction plan. If the remaining capacity in a certain area is insufficient after the transaction, it may be recommended to control the transaction size.

[0071] This application provides a management method for emissions trading, which uses a deep learning model to comprehensively predict accounting indicators before and after the transaction. The accounting dimension is expanded from single-point emissions to dynamic values ​​of regional environmental capacity, thereby helping to make trading decisions. The environmental impact assessment dimension is also added to the trading decision-making process, making the transaction assessment of the emissions trading platform more accurate and improving the scientific nature and environmental friendliness of emissions trading decisions.

[0072] Example 3, based on the same inventive concept, also provides a management device for emissions trading. The device includes: a response module, used to respond to a received emissions trading request and obtain the pre-trade accounting indicators corresponding to the seller and buyer, wherein the emissions trading request indicates the amount of emissions trading to be conducted by the seller and buyer and the target pollutant; a prediction module, used to input the emissions trading amount indicated by the emissions trading request and the emissions trading accounting indicators corresponding to the seller and buyer into an environmental impact assessment model corresponding to the target pollutant to obtain the post-trade accounting indicators corresponding to the seller and buyer; and an accounting module, used to determine the capacity value of the target pollutant corresponding to the seller and buyer after the transaction based on the post-trade accounting indicators corresponding to the seller and buyer, to determine whether to support the emissions trading request.

[0073] In a preferred embodiment, the accounting indicators include at least the regional capacity benchmark value, the regional spatial correction coefficient, the spatiotemporal diffusion coefficient of pollutants, the pollutant type correction coefficient, and the remaining capacity proportion coefficient. The environmental impact assessment model includes an input layer, a feature encoding layer, a regional interactive attention layer, a physical embedding layer, and an output layer. The input layer normalizes the emission rights trading volume and the emission rights trading accounting indicators corresponding to the seller and buyer companies, outputting a normalized vector. The feature encoding layer performs feature dimensionality upscaling and nonlinear transformation using a multilayer perceptron, outputting a high-dimensional vector. The regional interactive attention layer uses the emission rights trading volume as a query matrix and learns the sub-vectors corresponding to the seller and buyer companies in the high-dimensional vector to output an attention vector. The physical embedding layer determines the physical vector based on the attention vector and the pollutant diffusion sub-model. The output layer extracts the post-trade accounting indicators corresponding to the seller and buyer companies from the physical vector using an activation function.

[0074] In a preferred embodiment, the step of determining the physical vector based on the fusion vector and the pollutant diffusion sub-model specifically includes: constructing a physical prior vector based on the pollution rights trading volume and the pollution rights trading accounting indicators corresponding to the seller and buyer companies; and concatenating the physical prior vector with the attention vector to obtain the concatenated vector. The physical vector is obtained by performing a nonlinear transformation on the spliced ​​vector in the following manner. : ;in, To fuse the weight matrix, This is the bias vector.

[0075] In a preferred embodiment, the physical prior vector includes at least the pollutant concentration influence coefficient, mass conservation deviation value, attenuation coefficient, and auxiliary characteristic value corresponding to the seller and buyer respectively.

[0076] In a preferred embodiment, the pollutant concentration influence coefficients corresponding to the seller / buyer are determined by the following method. : ; ; ;in, Emissions per unit time The average wind speed, , These are the diffusion parameters in the horizontal and vertical directions, respectively. For high-efficiency sources, This represents the corresponding spatiotemporal diffusion coefficient of the pollutants. This is the corresponding regional spatial correction coefficient.

[0077] In a preferred embodiment, the attenuation coefficients corresponding to the seller / buyer are determined by the following method. : ;in, is the attenuation constant.

[0078] In a preferred embodiment, the mass conservation deviation value is determined by the following method. : ; ; ;in, These are the correction factors for the pollutant types for the seller and buyer companies, respectively. For emissions trading volume, This is the diffusion loss value. This represents the settlement loss value.

[0079] Please refer to Figure 2, which is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. As shown in Figure 2, the electronic device 200 includes a processor 210, a memory 220, and a bus 230.

[0080] The memory 220 stores machine-readable instructions that can be executed by the processor 210. When the electronic device 200 is running, the processor 210 and the memory 220 communicate via the bus 230. When the machine-readable instructions are executed by the processor 210, the steps of the pollution rights trading management method as described in the above method embodiment can be executed. For specific implementation details, please refer to the method embodiment, which will not be repeated here.

[0081] This application also provides a computer-readable storage medium storing a computer program. When the computer program is run by a processor, it can execute the steps of the pollution rights trading management method as described in the above method embodiments. For specific implementation details, please refer to the method embodiments, which will not be repeated here.

[0082] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0083] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0084] Furthermore, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0085] Furthermore, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0086] It should be noted that if the function is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0087] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0088] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A management method for emissions trading, characterized in that, The method includes: in response to a received emissions trading request, the cross-regional allocation node obtains the pre-trade accounting indicators for the seller and buyer companies, respectively. The emissions trading request indicates the emissions trading volume and target pollutant to be traded by the seller and buyer companies. The cross-regional allocation node inputs the emissions trading volume indicated by the emissions trading request and the emissions trading accounting indicators for the seller and buyer companies into an environmental impact assessment model corresponding to the target pollutant to obtain the post-trade accounting indicators for the seller and buyer companies. Based on the post-trade accounting indicators for the seller and buyer companies, the cross-regional allocation node determines the capacity value of the target pollutant for each of the seller and buyer companies after the trade, to determine whether to support the emissions trading request.

2. The method according to claim 1, characterized in that, The accounting indicators include at least the regional capacity benchmark value, regional spatial correction coefficient, pollutant spatiotemporal diffusion coefficient, pollutant type correction coefficient, and remaining capacity proportion coefficient. The environmental impact assessment model includes an input layer, a feature encoding layer, a regional interactive attention layer, a physical embedding layer, and an output layer. The input layer normalizes the pollution rights trading volume and the corresponding pollution rights trading accounting indicators for the seller and buyer companies, outputting a normalized vector. The feature encoding layer performs feature dimensionality upscaling and nonlinear transformation using a multilayer perceptron, outputting a high-dimensional vector. The regional interactive attention layer uses the pollution rights trading volume as a query matrix and learns the sub-vectors corresponding to the seller and buyer companies in the high-dimensional vector to output an attention vector. The physical embedding layer determines the physical vector based on the attention vector and the pollutant diffusion sub-model. The output layer extracts the post-trade accounting indicators for the seller and buyer companies from the physical vector using an activation function.

3. The method according to claim 2, characterized in that, The steps for determining the physical vector based on the fusion vector and pollutant diffusion sub-model specifically include: constructing a physical prior vector based on the pollution rights trading volume and the pollution rights trading accounting indicators corresponding to the seller and buyer companies; and concatenating the physical prior vector with the attention vector to obtain the concatenated vector. The physical vector is obtained by performing a nonlinear transformation on the spliced ​​vector in the following manner. ： ;in, To fuse the weight matrix, This is the bias vector.

4. The method according to claim 3, characterized in that, The physical prior vector includes at least the pollutant concentration influence coefficient, mass conservation deviation value, attenuation coefficient, and auxiliary characteristic value for the seller and buyer companies, respectively.

5. The method according to claim 4, characterized in that, The pollutant concentration impact coefficients for the seller / buyer companies are determined using the following method. ： ； ； ;in, Emissions per unit time The average wind speed, 、 These are the diffusion parameters in the horizontal and vertical directions, respectively. For high-efficiency sources, This represents the corresponding spatiotemporal diffusion coefficient of the pollutants. This is the corresponding regional spatial correction coefficient.

6. The method according to claim 4, characterized in that, The attenuation coefficients corresponding to the seller / buyer companies are determined using the following methods. ： ;in, is the attenuation constant.

7. The method according to claim 4, characterized in that, The mass conservation deviation value was determined using the following method. ： ； ； ;in, These are the correction factors for the pollutant types for the seller and buyer companies, respectively. For emissions trading volume, This is the diffusion loss value. This represents the settlement loss value.

8. A management device for emissions trading, characterized in that, The device includes: a response module, configured to respond to a received emissions trading request by acquiring pre-trade accounting indicators for both the seller and buyer companies, wherein the emissions trading request indicates the emissions trading volume and target pollutant to be traded by the seller and buyer companies; a prediction module, configured to input the emissions trading volume indicated by the emissions trading request and the emissions trading accounting indicators for both the seller and buyer companies into an environmental impact assessment model corresponding to the target pollutant, to obtain post-trade accounting indicators for both the seller and buyer companies; and an accounting module, configured to determine the post-trade capacity value of the target pollutant for both the seller and buyer companies based on the post-trade accounting indicators, to determine whether the emissions trading request is supported.

9. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the steps of the management method for emissions trading as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the management method for emissions trading as described in any one of claims 1 to 7.

Images