Carbon emission management method, system, storage medium and computer program product

Abstract

The application discloses a carbon emission management method, system, storage medium and computer program product, relates to the technical field of carbon emission management, and the method comprises the following steps: acquiring a carbon emission management instruction input by a user; calling a hierarchical digital twin sub-model corresponding to the carbon emission management instruction in a pre-constructed multi-level digital twin model system, wherein the multi-level digital twin model system is divided into a plurality of hierarchical digital twin sub-models in a manner combining administrative region division and project division; if the called hierarchical digital twin sub-model is a project-level digital twin model, collecting project energy consumption data, and calculating annual period carbon emission total amount based on the project energy consumption data; calling a pre-constructed graph generator to extract carbon quota parameters; and generating a carbon quota scheme according to the carbon quota parameters and the annual period carbon emission total amount. The multi-level digital twin model system is adapted to regional policy differences, and supports zero-carbon target quantitative management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of carbon emission management, and in particular to a carbon emission management method, system, storage medium and computer program product. BACKGROUND

[0002] Differentiated policies have been introduced by various regions for carbon emission management, and there are significant regional differences in carbon quota calculation methods, compliance requirements, and emission reduction incentive measures. However, mainstream carbon emission management systems mostly use a single architecture design, which is difficult to adapt to the dynamic adjustment needs of multi-level administrative regions (provincial-city-district), leading to lagging policy parameter extraction, low cross-regional data collaboration efficiency, and inability to provide precise compliance guidance for enterprises. At the same time, the carbon emission management of established parks faces the pain points of "data fragmentation, monitoring lag, and decision-making experience": park equipment energy consumption data, carbon asset information, policy texts, and other data are scattered in different systems, lacking unified data platform support, and it is difficult to achieve "real-time monitoring-trend prediction-scheme optimization" closed-loop management; the whole life cycle management of carbon assets such as enterprise quota, CCER (Certified Voluntary Emission Reduction), and green certificates lacks digital tool support, and "compliance compliance" and "market transactions" are disconnected, making it difficult to reduce emission reduction costs through carbon asset optimization.

[0003] Therefore, how to adapt to regional policy differences and support zero-carbon target quantification management has become a technical problem that needs to be solved by the present application.

[0004] The above content is only used to assist in understanding the technical solutions of the present application and does not represent the acknowledgement of the above content as prior art. SUMMARY

[0005] The main purpose of the present application is to provide a carbon emission management method, system, storage medium and computer program product, which aims to solve the technical problem of how to adapt to regional policy differences and support zero-carbon target quantification management.

[0006] To achieve the above-mentioned purpose, the present application provides a carbon emission management method, which comprises:

[0007] obtaining a carbon emission management instruction input by a user;

[0008] calling a hierarchical digital twin sub-model corresponding to the carbon emission management instruction in a pre-constructed multi-level digital twin model system, wherein the multi-level digital twin model system is divided into a plurality of hierarchical digital twin sub-models in a manner combining administrative region division and project division;

[0009] if the called hierarchical digital twin sub-model is a project-level digital twin model, collecting project energy consumption data and calculating annual period carbon emission total based on the project energy consumption data;

[0010] Call a pre-built atlas generator to extract carbon quota parameters;

[0011] Generate a carbon quota scheme according to the carbon quota parameters and the total annual cycle carbon emissions.

[0012] In an embodiment, the step of calling a hierarchical digital twin sub-model corresponding to the carbon emission management instruction in the pre-built multi-level digital twin model system further comprises:

[0013] Define a double-dimensional division rule according to the administrative region level and the project entity attribute, the double-dimensional division rule including a geographical boundary division rule and a project attribution division rule;

[0014] Collect remote sensing geographical information and rule text based on the geographical boundary division rule, and collect project point cloud data in the form of unmanned aerial oblique photography based on the project attribution division rule;

[0015] Bind the policy text with the remote sensing geographical information, and respectively construct a first-level digital twin model, a second-level digital twin model, and a third-level digital twin model;

[0016] Distributedly deploy a federal learning node in the first-level digital twin model, the second-level digital twin model, and the third-level digital twin model;

[0017] Construct a project-level digital twin model according to the project point cloud data, develop a lightweight data upload interface, connect the federal learning node based on the lightweight data upload interface, and obtain a multi-level digital twin model system.

[0018] In an embodiment, the step of calling a hierarchical digital twin sub-model corresponding to the carbon emission management instruction in the pre-built multi-level digital twin model system comprises:

[0019] Parse the carbon emission management instruction and extract a hierarchical identification parameter;

[0020] Retrieve the multi-level digital twin model system based on the hierarchical identification parameter, and locate a hierarchical digital twin sub-model;

[0021] Send a calling request to the hierarchical digital twin sub-model, call the hierarchical digital twin sub-model, and render a policy text corresponding to the hierarchical identification parameter in a visual interface.

[0022] In an embodiment, if the called hierarchical digital twin sub-model is a project-level digital twin model, the step of collecting project energy consumption data and calculating the total annual cycle carbon emissions based on the project energy consumption data comprises:

[0023] The IoT gateway based on the project-level digital twin model collects project energy consumption data;

[0024] The project energy consumption data is unified to a standard time axis, and Kalman filtering is used to remove outliers from the project energy consumption data to obtain standardized project energy consumption data.

[0025] The annual energy structure basic data is retrieved from the third-level digital twin model, and the energy structure correction coefficients issued by the first-level digital twin model through the federated nodes are obtained.

[0026] The carbon emission factor is obtained by multiplying the annual basic energy structure data and the energy structure correction coefficient.

[0027] By calling historical data from the same period and predicting energy consumption data for the remaining period based on the historical data from the same period and the carbon emission factor;

[0028] By summing the energy consumption data of the standardized project and the energy consumption data of the remaining time period, the total annual carbon emissions are obtained.

[0029] In one embodiment, after the step of calling the hierarchical digital twin sub-model corresponding to the carbon emission management directive in the pre-constructed multi-level digital twin model system, the method further includes:

[0030] If the called hierarchical digital twin sub-model is a first-level digital twin model, then the pre-built graph generator is called to identify the first-level rule text and extract the first instruction parameter set;

[0031] Generate regional energy structure correction coefficients based on the first instruction parameter set;

[0032] Based on federated learning nodes, the industrial structure feature vector and energy intensity feature vector uploaded from the second-level digital twin model are obtained, and an administrative division emission reduction heat map is generated based on the industrial structure feature vector and the energy intensity feature vector.

[0033] If the called hierarchical digital twin sub-model is a second-level digital twin model, then the graph generator is called to identify the second-level rule text and extract the second instruction parameter set;

[0034] Acquire second-level industrial structure data, overlay the second-level industrial structure data and the second-level geographic information layer, and determine the vector boundary of the high-energy-consuming industry cluster area;

[0035] The regional energy structure correction coefficients issued by the first-level digital twin model are invoked, and the regional energy structure correction coefficients are split according to the second-level industrial structure data to generate an industry-specific energy structure correction coefficient matrix.

[0036] If the called hierarchical digital twin sub-model is a third-level digital twin model, the atlas generator is called to identify third-level policy text and extract a third instruction parameter set;

[0037] The project information uploaded by the project-level digital twin model is called, and a visual project list is generated based on the project information.

[0038] In an embodiment, the step of calling the pre-constructed atlas generator to extract the carbon quota parameter includes:

[0039] The pre-constructed atlas generator is called to traverse the rule text to extract the core carbon asset parameter, and the rule text includes a first-level rule text, a second-level rule text, and a third-level rule text.

[0040] The core carbon asset parameter is logically verified and associated with an industry benchmark value of the project to obtain the carbon quota parameter.

[0041] In an embodiment, the step of generating a carbon quota scheme according to the carbon quota parameter and the annual cycle carbon emission total includes:

[0042] The quota gap is calculated based on the annual cycle carbon emission total.

[0043] A multi-objective optimization algorithm is used to calculate the quota gap and the carbon quota parameter to generate an initial emission reduction scheme.

[0044] The initial emission reduction scheme is modified by synchronizing the energy structure correction coefficient through the federal learning node to obtain the carbon quota scheme.

[0045] In addition, to achieve the above-mentioned purpose, the present application also provides a carbon emission management system, which comprises a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the computer program is configured to implement the steps of the carbon emission management method as described above.

[0046] In addition, to achieve the above-mentioned purpose, the present application also provides a storage medium, which is a computer-readable storage medium, and the storage medium stores a computer program, wherein the computer program is executed by a processor to implement the steps of the carbon emission management method as described above.

[0047] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which comprises a computer program, wherein the computer program is executed by a processor to implement the steps of the carbon emission management method as described above.

[0048] The one or more technical solutions provided by the present application have at least the following technical effects:

[0049] acquire a carbon emission management instruction input by a user; call a hierarchical digital twin sub-model corresponding to the carbon emission management instruction in a pre-constructed multi-level digital twin model system, wherein the multi-level digital twin model system is divided into a plurality of hierarchical digital twin sub-models in a manner combining administrative region division and project division; if the called hierarchical digital twin sub-model is a project-level digital twin model, collect project energy consumption data, and calculate annual period carbon emission total based on the project energy consumption data; call a pre-constructed graph generator to extract a carbon quota parameter; and generate a carbon quota scheme according to the carbon quota parameter and the annual period carbon emission total. In the present application, a multi-level digital twin model system of "administrative region + project entity" dual dimensions is constructed, so that the carbon emission management instruction input by the user can be directly mapped to the corresponding hierarchical sub-model; when the instruction points to the project-level model, the project energy consumption data is collected in real time and the annual period carbon emission total is calculated, and the graph generator is called to analyze the differentiated carbon quota parameter in the region, to automatically generate a precise carbon quota scheme. Through the closed-loop mechanism of instruction, called model, extracted carbon quota parameter, and formed carbon quota scheme, the dynamic adaptation to the policy differences of various places is realized, and the zero-carbon target is quantified into a predictable, adjustable, and tradable quota path, thereby supporting the quantitative management and landing of the whole-process zero-carbon target of the park. BRIEF DESCRIPTION OF DRAWINGS

[0050] The accompanying drawings, which are incorporated herein and form a part of the specification, illustrate embodiments consistent with the present application and, together with the description, further serve to explain the principles of the application.

[0051] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced hereinafter. Obviously, for those of ordinary skill in the art, other drawings can also be obtained based on these drawings without any creative effort.

[0052] Figure 1 a flowchart provided for a first embodiment of the carbon emission management method of the present application;

[0053] Figure 2 a flowchart provided for a second embodiment of the carbon emission management method of the present application;

[0054] Figure 3 a flowchart provided for a fourth embodiment of the carbon emission management method of the present application;

[0055] Figure 4 a module structure diagram of the carbon emission management device according to the embodiments of the present application;

[0056] Figure 5A device structure schematic diagram of a hardware operating environment involved in a carbon emission management method in embodiments of the present application.

[0057] The object implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0058] It should be understood that the specific embodiments described herein are merely intended to explain the technical solutions of the present application, and are not intended to limit the present application.

[0059] In order to better understand the technical solutions of the present application, the specific embodiments will be described in detail below with reference to the drawings and the accompanying drawings.

[0060] The main solution of the embodiments of the present application is: obtaining a carbon emission management instruction input by a user; calling a hierarchical digital twin sub-model corresponding to the carbon emission management instruction in a pre-constructed multi-level digital twin model system, wherein the multi-level digital twin model system is divided into a plurality of hierarchical digital twin sub-models in a manner combining administrative region division and project division; if the called hierarchical digital twin sub-model is a project-level digital twin model, collecting project energy consumption data and calculating annual period carbon emission total amount based on the project energy consumption data; calling a pre-constructed graph generator to extract carbon quota parameters; and generating a carbon quota scheme according to the carbon quota parameters and the annual period carbon emission total amount.

[0061] The embodiments of the present application take into account that: different policies are introduced by each region for carbon emission management, and there are significant regional differences in carbon quota calculation methods, compliance requirements, and emission reduction incentive measures. However, existing carbon emission management systems mostly adopt a single architecture design, which is difficult to adapt to the dynamic adjustment needs of multi-level administrative regions (provincial-city-district) policies, resulting in lagging policy parameter extraction, low cross-regional data collaboration efficiency, and inability to provide accurate compliance guidance for enterprises. At the same time, the carbon emission management of the built park faces the pain points of “data fragmentation, monitoring lag, and decision experience”: park equipment energy consumption data, carbon asset information, policy texts, etc. are scattered in different systems, lack of unified data platform support, and it is difficult to realize the closed-loop management of “real-time monitoring-trend prediction-scheme optimization”; the whole life cycle management of carbon assets such as enterprise quota, CCER (Certified Emission Reduction, Certified Emission Reduction), green certificate, etc. lacks digital tool support, and “compliance compliance” and “market transaction” are disconnected, making it difficult to reduce emission reduction costs through carbon asset optimization.

[0062] Therefore, the solution provided by the application can make the carbon emission management instruction input by the user directly mapped to the sub-model of the corresponding level by constructing a multi-level digital twin model system with two dimensions of "administrative region + project entity", when the instruction points to the project level model, the project energy consumption data is collected and the annual carbon emission total is calculated, and the graph generator is called to analyze the differentiated carbon quota parameters in the region to automatically generate a precise carbon quota scheme. Through the closed-loop mechanism of instruction, calling model, extracting carbon quota parameters, and forming a carbon quota scheme, the dynamic adaptation to the policy differences of each region is realized, and the zero-carbon target is quantified into a predictable, adjustable, and tradable quota path, thereby supporting the quantitative management and landing of the whole-process zero-carbon target of the park.

[0063] It should be noted that the execution subject of the embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device, a carbon emission management system, a carbon emission management platform, etc. capable of realizing the above functions. Hereinafter, the carbon emission management system will be taken as an example to describe the embodiment and the following embodiments.

[0064] Based on this, the embodiment of the application provides a carbon emission management method, referring to Figure 1 , Figure 1 The flowchart of the first embodiment of the carbon emission management method of the application is shown in the figure.

[0065] In the embodiment, the carbon emission management method includes steps S10-S50:

[0066] Step S10, obtaining a carbon emission management instruction input by a user;

[0067] Obtaining the carbon emission management instruction input by the user means receiving the instruction data from the government end, the park end, the enterprise end or the third-party platform through the man-machine interaction interface, the API interface or the Internet of Things gateway. The instruction data is carried in the form of JSON (JavaScript Object Notation), XML (Extensible Markup Language) or MQTT (Message Queuing Telemetry Transport) message, and contains the user identity, the target administrative level identifier, the project unique code, the management type (query / simulation / transaction / compliance) and the time range field.

[0068] The purpose of obtaining the carbon emission management instruction is to trigger the accurate calling of the subsequent multi-level digital twin model and the carbon asset closed-loop management, so as to decompose the macro zero-carbon target into executable and quantifiable operation tasks layer by layer, and avoid the errors and time lag caused by traditional manual decomposition.

[0069] In one possible implementation, the carbon emission management system provides a graphical instruction builder, and the user can automatically generate instructions by dragging and dropping administrative region components and project nodes into a canvas; in another possible implementation, the system opens a RESTful (Representational State Transfer) interface to allow enterprise ERP (Enterprise Resource Planning) or management platforms to directly push instructions.

[0070] Step S20, calling a hierarchical digital twin sub-model corresponding to the carbon emission management instruction in a pre-constructed multi-level digital twin model system, wherein the multi-level digital twin model system is divided into a plurality of hierarchical digital twin sub-models in a manner combining administrative region division and project division;

[0071] The multi-level digital twin model system refers to a four-layer tree model cluster composed of an "administrative region dimension" and a "project entity dimension" orthogonal combination, including a first-level digital twin model, a second-level digital twin model, a third-level digital twin model, and a project-level digital twin model, wherein the project-level digital twin model is the model with the smallest granularity, each model has independent spatial resolution, time resolution and data granularity, and realizes cross-layer parameter synchronization through a federal learning node.

[0072] Exemplarily, the multi-level digital twin model system can be a four-layer model cluster divided in two dimensions of "administrative region + project entity": a provincial digital twin model, a municipal digital twin model, a district digital twin model, and a project-level digital twin model.

[0073] Administrative region division rule: according to the latest statistical division code and urban-rural division code defined provincial, municipal and county level boundaries.

[0074] Project division rule: taking the project unique code in the project record certificate or the environmental impact assessment reply document as the identifier, mapping the physical project to the digital twin node.

[0075] In addition, it should be noted that the purpose of calling the hierarchical digital twin sub-model is to automatically route the user instruction to the model with corresponding policy authority, data granularity and computing capacity, and reduce the adaptation cost across regions and projects.

[0076] In one possible implementation, the carbon emission management system adopts a distributed service bus + microservice registry center architecture to complete the call through hierarchical keyword matching + load balancing. Specifically, after the user instruction is parsed, the hierarchical identification parameter (such as "provincial power industry") is extracted, the microservice registry center retrieves the service metadata (including address, version, and load state) of the corresponding hierarchical digital twin sub-model according to the parameter. Then, the distributed service bus combines the hierarchical keyword matching algorithm (such as administrative region code + industry label) to filter the optimal node from the available service instances, and distributes the call request through the weighted round-robin load balancing strategy. During the call process, the service bus monitors the node health status in real time, automatically removes abnormal instances and supplements standby nodes. Finally, after the target sub-model executes the instruction, the rule text and three-dimensional scene data are returned through the service bus, and dynamic rendering is completed in the visualization interface.

[0077] In another possible implementation, the carbon emission management system based on a blockchain smart contract automatically triggers the cold start of the corresponding model image after obtaining the user digital signature. Specifically, when the user initiates a carbon emission management instruction, the user first generates a digital signature through an asymmetric encryption algorithm, and submits the instruction content and the signature to the blockchain node. After the smart contract verifies the legality of the signature and the user's authority, it automatically matches the hierarchical identification parameter (such as "first-level model") in the instruction, and calls the hash address of the corresponding model image from the off-chain storage. Subsequently, the contract triggers the container orchestration engine to pull the image and allocate computing resources, loads the policy parameters and federated learning node configuration through the predefined initialization script, and completes the model cold start. During the startup process, the node real-time uploads the image verification value, resource occupation, and other metadata to the chain for evidence, ensuring traceability. After the model is ready, the smart contract returns the call result to the user and triggers the subsequent data collection and carbon quota calculation process through the event notification mechanism, realizing the entire process of decentralized and automated trusted execution.

[0078] For example, in a specific implementation, when the instruction carries the "A province-B city-C chemical project" field, the system first activates the A province digital twin model to obtain the provincial energy structure correction coefficient, then cascades to activate the B city digital twin model to extract the city-level industry benchmark value, and finally wakes up the project-level digital twin model to load the 3D grid model generated by the unmanned aerial vehicle oblique photography.

[0079] In step S30, if the called hierarchical digital twin sub-model is a project-level digital twin model, project energy consumption data is collected, and the annual period carbon emission total is calculated based on the project energy consumption data;

[0080] Project-level digital twin model: A high-fidelity digital twin with a single construction or operation project as the smallest management unit. Its three-dimensional visual shell is obtained by Mesh reconstruction of point cloud data collected by unmanned aerial vehicle oblique photography. It is embedded with Internet of Things gateways, energy consumption monitoring terminals, and edge computing nodes. Mesh reconstruction is the process of converting point cloud data collected by unmanned aerial vehicle oblique photography into continuous polygon mesh models through three-dimensional grid generation algorithms.

[0081] Project energy consumption data: cumulative, instantaneous, and state variables uploaded by intelligent electric meters, water meters, gas meters, steam flow meters, photovoltaic inverters, etc. at preset time intervals.

[0082] Annual total carbon emissions: Under the framework of ISO 14064-1, energy consumption data is converted into annual cumulative CO2 equivalent values through emission factor method or mass balance method. ISO 14064-1 framework refers to the greenhouse gas quantification framework published by the International Organization for Standardization.

[0083] In addition, it should be noted that the purpose of collecting project energy consumption data and calculating annual total carbon emissions is to establish a "auditable, traceable, and predictable" carbon emission benchmark for the project, support accurate calculation of quota gap and later carbon asset development, and avoid compliance risks caused by data loss.

[0084] In one possible implementation, the Internet of Things gateway uploads data through an encrypted channel; in another possible implementation, the edge computing node first performs local data cleaning and then returns the data to the cloud time series database in batch mode through a lightweight interface.

[0085] For example, in a specific implementation, the data center project-level digital twin model collects power and flow data from 312 measurement points of UPS (Uninterruptible Power Supply), water chillers, and precision air conditioners in real time through Modbus-TCP (Modbus Transmission Control Protocol) protocol. After removing abnormal spikes through Kalman filtering, combined with the regional power grid emission factor 0.6299tCO2 / MWh issued by the provincial model, the annual total carbon emissions in 2024 are calculated to be 15,788tCO2.

[0086] Step S40, call the pre-constructed graph generator to extract carbon quota parameters;

[0087] Graph generator: An automated rule analysis engine based on natural language processing and knowledge graph technology, with input of carbon emission policy texts, quota allocation schemes, and industry benchmark values published by national, provincial, municipal, and district authorities.

[0088] Carbon quota parameters: five-tuple of "unit product carbon emission benchmark value, free quota proportion, pre-borrowing proportion, compliance period, and CCER offset upper limit" corresponding to the industry to which the project belongs.

[0089] In addition, it should be noted that the purpose of extracting carbon quota parameters is to convert unstructured policy provisions into calculable numerical rules, thereby eliminating subjective differences in manual interpretation of policies and ensuring that subsequent quota schemes can achieve the goal of zero carbon emission with the lowest cost and minimal changes.

[0090] In one possible implementation, the graph generator extracts entities and relationships using a BERT+CRF model and completes logical verification through a rule engine. BERT (Bidirectional Encoder Representations from Transformers) is a pre-trained language model based on the Transformer architecture, which can understand text semantics through bidirectional context; CRF (Conditional Random Field) is a sequence labeling model that can optimize entity boundary prediction by combining context features. The bidirectional context understanding capability of BERT solves the recognition error of traditional models for ambiguous text (such as "carbon sink" and "carbon trading"), and CRF improves the recognition accuracy of regular text by optimizing label prediction through sequence dependency.

[0091] Specifically, the carbon emission-related policy text input data is segmented, tagged with parts of speech, and converted into a word vector sequence that can be recognized by the BERT model. The BERT model encodes the text bidirectionally, outputting a context semantic vector for each character; the CRF layer predicts entity boundaries (such as "enterprise name" and "carbon emissions") and relationship types (such as "belongs to" and "emission") based on semantic vectors and label transition probabilities. The extracted results are passed into the rule engine, which verifies the matching of entity types and the logic of relationships through pre-set rules (such as "carbon quota" entities must be associated with "region" attributes, and "emission relationships" must include timestamps) to filter contradictory or invalid triples. Entities and relationships that pass the verification are stored in the knowledge graph, forming the graph generator.

[0092] Step S50, generating a carbon quota scheme according to the carbon quota parameters and the total annual carbon emissions.

[0093] Carbon quota scheme: taking the total carbon emissions and carbon quota parameters as inputs in an annual cycle, the "quota gap, quota trading strategy, CCER development scale, emission reduction technology investment list, and compliance schedule" combination scheme is calculated by a multi-objective optimization algorithm (NSGA-III or improved particle swarm).

[0094] It should be noted that NSGA-III (Non-dominated Sorting Genetic Algorithm III) is a multi-objective optimization algorithm that guides population evolution by introducing uniformly distributed reference points, balances convergence and diversity of solutions by combining non-dominated sorting and adaptive normalization strategies, and is suitable for high-dimensional optimization problems with more than three objectives.

[0095] The improved particle swarm algorithm is an improvement of the standard particle swarm optimization, which enhances the global search ability and avoids premature convergence by dynamically adjusting the inertia weight, learning factor or fusing genetic operators, simulated annealing and other strategies, and improves the solution accuracy and efficiency of complex optimization problems.

[0096] By using NSGA-III or improved particle swarm for carbon quota scheme generation, multiple objectives such as emission reduction cost and carbon asset income can be optimized simultaneously.

[0097] In addition, it should be noted that the purpose of generating a carbon quota scheme is to convert the macro zero-carbon target into an annual action blueprint that can be executed by enterprises, and to balance compliance, minimize emission reduction costs, and maximize carbon asset income simultaneously, avoiding passive high-cost quota purchase or excessive emission reduction waste in the future.

[0098] In one possible implementation, the system targets "economic cost, emission reduction, and investment payback period", and calls the improved NSGA-III algorithm to generate a Pareto frontier, which is selected by the enterprise itself; in another possible implementation, the system introduces a federal learning node to synchronize real-time energy structure correction coefficients and dynamically adjust the scheme to respond to policy mutations.

[0099] The embodiment provides a carbon emission management method, which constructs a multi-level digital twin model system in two dimensions of "administrative region + project entity", so that the carbon emission management instructions input by the user can be directly mapped to the sub-model of the corresponding level; when the instruction points to the project level model, the project energy consumption data is collected and the annual total carbon emissions are calculated, and the graph generator is called to analyze the differentiated carbon quota parameters in the region, and a precise carbon quota scheme is automatically generated. Through the closed-loop mechanism of instructions, calling models, extracting carbon quota parameters, and forming carbon quota schemes, not only can the dynamic adaptation to policy differences in different places be realized, but also the zero-carbon target can be quantified into a predictable, adjustable, and tradable quota path, thereby supporting the quantitative management and landing of the whole-process zero-carbon target of the park.

[0100] In a possible implementation, step S20 can include steps S21-S23:

[0101] In step S21, the carbon emission management instruction is parsed to extract a hierarchy identification parameter.

[0102] The original message of the carbon emission management instruction is subjected to joint syntax-semantics analysis by using a natural language processing engine to extract a hierarchy identification parameter. The hierarchy identification parameter indicates a string or an enumeration value extracted from the instruction for uniquely identifying the hierarchy of the four-level model of the province, city, district and project.

[0103] In a possible implementation, the system extracts the hierarchy identification parameter by using regular expressions and dictionary matching. Specifically, the system first parses the original message of the carbon emission management instruction, constructs a regular expression rule library based on a preset administrative division keyword dictionary (such as “province”, “city”, “district” and “project”), and locates the hierarchy identification field (such as “A province-B city-C project”) in the instruction by using a greedy matching algorithm. Then, the system calls an administrative division code dictionary to verify the legality of the matching result and filters invalid or ambiguous matching items. Finally, the system extracts standardized hierarchy identification parameters (such as the province-level code, city-level code and project unique code) and outputs them to the model calling module.

[0104] In another possible implementation, the system automatically identifies and outputs the hierarchy label by using a BERT+CRF sequence labeling model. Specifically, the system tokenizes the instruction text and converts it into a word vector, inputs the pre-trained BERT model for bidirectional context encoding to generate a character-level feature vector containing context semantics, and predicts the hierarchy label of each character based on the feature vector and a label transition probability matrix (such as the transition weight from the “project level” to the “province level”). The system decodes the optimal label sequence by using the Viterbi algorithm, extracts the hierarchy identification parameter corresponding to the continuous label, and outputs it after verification by a rule engine.

[0105] In step S22, the multi-level digital twin model system is retrieved based on the hierarchy identification parameter to locate a hierarchy digital twin submodel.

[0106] The hierarchy identification parameter is used to perform node matching query to retrieve the matching degree of each hierarchy digital twin submodel and the hierarchy identification parameter in the multi-level digital twin model system, and the metadata of the target submodel, including the model version, calculation accuracy, data update time and service address, are returned according to the matching result. Thus, the global model space is reduced to a single target model, and the performance loss caused by full-scan is avoided.

[0107] In a possible implementation, the system adopts Redis key-value cache and Bloom filter to accelerate positioning. Specifically, the level identification parameters of the multi-level digital twin model system are preloaded into the Bloom filter. After receiving the level identification parameters, it is first determined by the Bloom filter whether the parameters exist or not, and invalid queries are filtered. For the parameters determined to exist, the model metadata (service address, version, etc.) is obtained by querying the Redis cache. If the cache is not hit, the database is searched and the results are written into Redis to realize model positioning. By adopting the double filtering mechanism of Redis key-value cache and Bloom filter, invalid database access is reduced and positioning time is compressed.

[0108] Step S23, send a call request to the hierarchical digital twin sub-model, call the hierarchical digital twin sub-model, and render the rule text corresponding to the hierarchical identification parameter in the visualization interface.

[0109] The RESTful call message based on HTTPS (HyperText Transfer Protocol Secure) carries user identity tokens, model instance IDs, time ranges, and output format parameters, and sends a call request to the hierarchical digital twin sub-model to call the hierarchical digital twin sub-model.

[0110] The visualization interface specifically refers to a three-dimensional digital twin browser that supports spatial roaming, attribute query, and timeline playback. The policy text refers to the carbon emission policy clauses corresponding to the target level after structured analysis, which are superimposed on the administrative boundaries or project red lines in the three-dimensional scene in the form of highlighted cards.

[0111] In addition, it should be noted that sending a call request and rendering a policy text can enable users to intuitively view the correspondence between policy requirements and spatial objects in an immersive three-dimensional scene, significantly reducing the policy understanding threshold and improving decision-making efficiency.

[0112] In a possible implementation, step S40 can include steps S41-S42:

[0113] Step S41, call the pre-constructed graph generator to traverse the rule text to extract core carbon asset parameters. The rule text includes first-level rule text, second-level rule text, and third-level rule text.

[0114] The pre-constructed atlas generator is called to scan the paragraphs, chapters, and articles of the rule text according to the provincial, municipal, and district rule texts, to obtain a set of numerical fields directly related to quotas, emission reduction amounts, transactions, and performance from the rule text, and to obtain core carbon asset parameters. Exemplarily, the core carbon asset parameters include an industry carbon emission benchmark value, a free quota proportion, a pre-borrowing proportion, a performance period, a CCER offset upper limit, and a green electricity deduction coefficient.

[0115] In step S42, the core carbon asset parameters are logically checked, and an industry benchmark value to which the project belongs is associated to obtain carbon quota parameters.

[0116] Logical checking: According to a preset business rule, the legality, rationality, and consistency of the core carbon asset parameters are automatically verified, including numerical interval checking, unit consistency checking, and time logic checking.

[0117] Industry benchmark value to which the project belongs: A unit product or unit value carbon emission benchmark line of the industry to which the project belongs, which is released by an official.

[0118] Carbon quota parameters: A final available parameter set after being checked and associated with the industry benchmark value, which is used for subsequent quota gap calculation and emission reduction scheme generation.

[0119] The purpose of logical checking and associating the benchmark value is to eliminate policy text ambiguity, numerical errors, and version conflicts, to ensure the uniqueness and authority of the parameters, and to ensure the accuracy of the subsequent carbon quota scheme.

[0120] Based on the first embodiment of the present application, the second embodiment of the present application is proposed. In the second embodiment of the present application, the same or similar contents as the above first embodiment can be referred to the above introduction, and will not be described in detail.

[0121] On this basis, please refer to Figure 2 , Figure 2 The second embodiment process schematic diagram provided for the present application. As Figure 2 shown, before step S10, the carbon emission management method further includes steps S01-S05:

[0122] In step S01, a double-dimension division rule is defined according to an administrative region level and a project entity attribute, the double-dimension division rule including a geographical boundary division rule and a project attribution division rule.

[0123] Project entity attribute: A set of attributes including a project name, a project unique code, an industry category, a construction site, a legal entity, a construction scale, a production process, and the like, which are specified in a project approval, an environmental impact assessment approval, or a record certificate.

[0124] Two-dimensional division rule: The division criterion is obtained by orthogonally combining the geographical dimension of "administrative region" and the engineering dimension of "project entity".

[0125] Geographic boundary delineation rules: Based on the vector data of administrative division boundaries that can be identified by remote sensing imagery, natural features such as river systems and transportation arteries are superimposed as auxiliary constraints to form closed polygonal boundaries.

[0126] Project attribution rules: Using the coordinates of the construction location recorded on the project filing certificate as the anchor point, determine which geographical boundary polygon the coordinates fall into, and thus determine which administrative region the project belongs to.

[0127] In addition, it should be noted that the purpose of defining the two-dimensional division rules is to decompose the regional carbon management tasks step by step, so that the subsequent digital twin model has both policy consistency and project feasibility; when policies are adjusted or projects are added or removed, the system only needs to update the rules to automatically rearrange the model hierarchy, which significantly reduces maintenance costs.

[0128] In one possible implementation, the geographic boundary delineation rule uses an API interface to obtain the latest boundary vector in real time; in another possible implementation, the project affiliation delineation rule automatically completes the affiliation matching by spatially determining the spatial inclusion of project coordinates and administrative divisions.

[0129] Step S02: Collect remote sensing geographic information and rule text based on the geographic boundary division rules, and collect project point cloud data by UAV oblique photography based on the project affiliation division rules.

[0130] Remote sensing geographic information: Multispectral, hyperspectral, and synthetic aperture radar (SAR) images acquired by satellite or airborne remote sensing platforms, and their derived land cover classification, vegetation index, and building density raster data.

[0131] The rule text refers to documents publicly released by national, provincial, municipal, and district-level departments or authoritative organizations regarding carbon emission management policies, quota allocation schemes, and industry energy consumption limit standards.

[0132] Drone oblique photography: Using a multi-rotor drone equipped with a five-lens oblique camera, the project site is photographed from multiple angles along a "grid" shaped flight path to obtain a high-resolution image sequence;

[0133] Project point cloud data: A set of three-dimensional point coordinates generated by encrypting and densely matching oblique images using the structure-of-motion motion reconstruction algorithm. It includes X, Y, Z three-dimensional coordinates and RGB color information.

[0134] Step S03: Bind the rule text to the remote sensing geographic information to construct a first-level digital twin model, a second-level digital twin model, and a third-level digital twin model respectively;

[0135] The administrative division field in the rule text is semantically and spatially aligned with the vector boundary in the remote sensing geographic information, and is stored in a "province-city-district" node relationship in a graph database to construct a first-level digital twin model, a second-level digital twin model, and a third-level digital twin model, i.e., a province-city-district digital twin model.

[0136] The province-level digital twin model refers to a carbon emission grid model with the whole province as a unit, including multi-dimensional attributes such as energy structure, industrial structure, traffic flow, and population density. The city-level digital twin model refers to a carbon emission grid model with the whole city as a unit, which inherits the correction coefficient of the province-level model and superimposes the city-level industrial agglomeration zone boundary. The district-level digital twin model refers to a carbon emission grid model with the whole district as a unit, which further superimposes the district-level key project vector boundary and real-time air quality monitoring station data.

[0137] The purpose of constructing the province, city, and district three-level models is to form a continuous downscaling calculation framework. The upper-level model provides boundary conditions and initial fields for the lower-level model, and the lower-level model provides verification and feedback for the upper-level model.

[0138] In one possible implementation, the province-level digital twin model adopts a WRF-Chem (Weather Research and Forecasting model coupled to Chemistry) and LEAP (Long-range Energy Alternatives Planning system) coupled energy-atmosphere interaction. Specifically, WRF-Chem is an online coupled meteorology-chemistry model that can simulate regional-scale atmospheric pollutant transport (such as PM2.5, ozone), meteorological conditions (such as wind speed, temperature) on carbon emission diffusion, and aerosol-radiation feedback (such as the effect of greenhouse gas concentration on regional climate). LEAP is a bottom-up energy planning tool that predicts the impact of different energy policies (such as coal power replacement, renewable energy proportion increase) on provincial carbon emission total through scenario analysis, outputting carbon emission lists by industry and fuel type. The energy carbon emission data generated by LEAP is input as a human emission source to drive atmospheric pollution simulation. The meteorological correction parameters output by WRF-Chem are fed back to LEAP to optimize the dynamic adjustment of carbon emission factors.

[0139] The coupling of WRF-Chem and LEAP can realize the simulation of regional atmospheric pollutant transport, the influence of meteorological conditions on carbon emission diffusion, and the feedback of aerosol-radiation, while LEAP can predict the influence of energy policy on the total carbon emission of the province and output the carbon emission list of different industries and fuel types. LEAP drives the WRF-Chem atmospheric pollution simulation, and the output of the meteorological correction parameter feeds back to the LEAP optimization of the dynamic adjustment of the carbon emission factor, improving the simulation accuracy and policy adaptability.

[0140] Step S04, the federated learning nodes are distributedly deployed in the first-level digital twin model, the second-level digital twin model, and the third-level digital twin model;

[0141] The federated learning node is a lightweight computing agent embedded in the digital twin model, which has the functions of local model training, gradient encryption, and parameter uploading.

[0142] Specifically, the federated learning nodes are deployed in the form of Docker containers in the three-level digital twin models, forming a collaborative computing network that is horizontally cross-domain and vertically cross-level. Thus, the model parameters are shared without domain data, realizing the dynamic collaboration of cross-regional policy parameters and energy structure correction coefficients.

[0143] In a possible implementation, the nodes use personalized federated learning to retain local characteristic parameters for different administrative levels. Specifically, after initializing the globally shared model, each administrative level node splits the parameters into a shared layer (such as the basic carbon emission factor) and a local characteristic layer (such as the regional industrial structure coefficient); the shared layer is frozen during local training, and only the characteristic layer parameters are updated; only the shared layer parameters are uploaded during the global aggregation stage, and a new shared model is generated after weighted averaging by the FedAvg algorithm; the node combines the updated shared layer and local characteristic layer parameters, and adapts to the local data distribution through gradient fine-tuning, realizing the model optimization of "global collaboration + level personalization". The FedAvg algorithm (Federated Averaging Algorithm) generates a global model by weighted averaging the local model parameters of each node.

[0144] Step S05, constructing a project-level digital twin model according to the project point cloud data, developing a lightweight data uploading interface, connecting the federated learning nodes based on the lightweight data uploading interface, and obtaining a multi-level digital twin model system.

[0145] The project-level digital twin model refers to a high-fidelity digital twin body with a single project land red line as a spatial range and a single device as a minimum management unit. The geometric shell of the digital twin body is obtained by Mesh reconstruction of project point cloud data, and the internal association is BIM (Building Information Modeling), IoT (Internet of Things) sensors, and video monitoring streams.

[0146] The lightweight data upload interface is a narrowband transmission interface. The project-level model uploads the locally calculated carbon emission factor and device energy efficiency curve to the district-level node through the lightweight interface, and receives the correction coefficient and policy parameter issued by the provincial, municipal, and district nodes, so as to complete the closed loop of the multi-level digital twin model system.

[0147] In a possible implementation, the lightweight interface is used for ultra-low power consumption sensors based on CoAP (Constrained Application Protocol); in another possible implementation, the interface supports WebRTC (Web Real-Time Communication) point-to-point transmission to realize real-time linkage of monitoring video and the digital twin model.

[0148] In the embodiment, four-level management boundaries of province, city, district, and project are divided according to the double-dimension rule. Remote sensing geographic information, policy texts, and unmanned aerial vehicle oblique photography point cloud data are used to complete data collection. Policy semantics and spatial boundaries are bound to construct a digital twin model. Cross-domain parameter collaboration is realized by using a distributed federated learning node. Finally, the project-level digital twin model is generated based on project point cloud data and is connected to the federated network through the lightweight interface. The four-level carbon emission digital twin system is closed and connected on the premise that data does not leave the domain. The carbon emission quantification management system that meets regional policy differences and is accurate to the project is finally formed. The zero-carbon goal can be decomposed, simulated, verified, optimized, and significantly reduced in the planning, construction, and operation life cycle. The cost of model reconstruction and data re-sampling caused by policy changes or project increases or decreases is significantly reduced. Finally, the government and enterprises can achieve the zero-carbon commitment at the lowest compliance cost.

[0149] Based on the first embodiment and / or the second embodiment of the present application, the third embodiment of the present application is proposed. In the third embodiment of the present application, the same or similar contents as the above embodiments can be referred to the above introduction, and will not be described hereinafter.

[0150] In the embodiment, if the called hierarchical digital twin sub-model is a project-level digital twin model, the step S30 of collecting project energy consumption data and calculating the total annual carbon emission based on the project energy consumption data can include steps S31-S36:

[0151] Step S31, the Internet of Things gateway based on the project level digital twin model collects project energy consumption data;

[0152] The Internet of Things gateway refers to an industrial-grade edge computing device deployed on the project site and supporting multi-protocol communication. Through the Internet of Things gateway, multi-dimensional data such as cumulative energy consumption, instantaneous power, flow, and state quantity uploaded in real time by terminals such as smart meters, water meters, gas meters, steam flow meters, and photovoltaic inverters are collected, and then project energy consumption data is obtained.

[0153] Step S32, the project energy consumption data is unified to a standard time axis, and Kalman filtering is used to remove outliers of the project energy consumption data, to obtain standardized project energy consumption data;

[0154] The project energy consumption data is unified to a standard time axis for time alignment. The standard time axis refers to a unified time series based on UTC. After time alignment, Kalman filtering is used to remove outliers of the project energy consumption data, and finally a high-quality energy consumption sequence after time axis alignment and outlier removal is obtained, i.e., standardized project energy consumption data.

[0155] Step S33, annual energy structure basic data is called from the third level digital twin model, and an energy structure correction coefficient issued by the first level digital twin model through a federal node is obtained;

[0156] Annual energy structure basic data: annual consumption and proportion by energy type (coal, oil, gas, electricity, heat, and renewable energy) stored in the district-level model.

[0157] Energy structure correction coefficient: a dimensionless coefficient calculated by the provincial model according to policy targets, external power, and renewable energy increment, used to correct the national default emission factor.

[0158] In addition, it should be noted that the purpose of calling district-level basic data and obtaining provincial correction coefficients is to achieve the same standard and same caliber of "regional-project" two-level data, and to avoid confusion caused by multiple sets of factors in parallel.

[0159] In one possible implementation, the correction coefficient is pushed daily through an HTTPS REST interface. The HTTPS REST interface is a RESTful style interface based on the HTTPS protocol, which realizes data interaction between the client and the server through HTTP methods. Specifically, at a fixed time every day (such as 2 a.m.), a daily correction coefficient data package is pushed to each node through the HTTPS REST interface; after receiving, the node verifies the data integrity, replaces the local historical coefficient, and records the update log.

[0160] In another possible implementation, real-time synchronization is used using gRPC (gRPC Bidirectional Streaming Real-time Synchronization). The node establishes a gRPC long connection with the server, the server monitors the coefficient changes in real time and pushes the updates through streaming; the node applies the updates immediately after receiving, and simultaneously feeds back the synchronization status, to realize dynamic synchronization of the correction coefficient.

[0161] Step S34, multiplying the annual energy structure basic data and the energy structure correction coefficient to obtain a carbon emission factor;

[0162] The carbon emission factor refers to a key conversion coefficient for converting project energy consumption data into carbon dioxide equivalent. Multiplication refers to multiplying each energy variety consumption in the annual energy structure basic data by the corresponding correction coefficient and then weighting and summing to obtain a localized comprehensive emission factor.

[0163] In addition, it should be noted that the purpose of generating the localized carbon emission factor is to make the project carbon emission accounting results truly reflect the differences in local power grid, heat, and fuel structure, thereby significantly improving the accuracy of the annual total carbon emission.

[0164] Step S35, calling historical same-period data, and predicting remaining period energy consumption data based on the historical same-period data and the carbon emission factor;

[0165] The historical same-period data refers to the energy consumption and carbon emission time series of the same project in the same calendar period in the historical period. The remaining period energy consumption data refers to the completed data obtained by using machine learning or statistical models to predict the energy consumption of the future unfinished period based on historical rules, production plans, weather forecasts, market orders, and other multi-dimensional factors.

[0166] In addition, it should be noted that the purpose of calling historical data and predicting the remaining period energy consumption is to solve the problem of incomplete real-time data for some projects that have not yet ended in the year, and to realize complete estimation of the annual total carbon emission; its effect is to lock the quota gap in advance under the premise of ensuring accuracy, to win the time for quota transaction or emission reduction measures for the enterprise.

[0167] In one possible implementation, the system uses a Prophet time series prediction model. The Prophet time series prediction model is a time series prediction algorithm suitable for medium and long term prediction, and is good at processing data containing seasonality, trend and outliers.

[0168] Specifically, input historical carbon emission data (such as monthly / quarterly emissions), automatically identify trend items (long-term growth / downturn), seasonal items (annual / quarterly cycles) and holiday effects in time series; the model generates prediction intervals through Bayesian inference, supports custom trend mutation points (such as policy implementation dates); output future 1-3 years of total carbon emission prediction curve, and provide trend decomposition chart (trend + season + residual), to assist in identifying key impact cycles.

[0169] In another possible implementation, the system adopts an XGBoost regression model, and the input includes production scheduling, air temperature, holidays and other features. The XGBoost regression model is an ensemble learning model based on gradient boosting trees, which improves prediction accuracy by optimizing the objective function and regularization, and supports multiple feature inputs.

[0170] Specifically, integrate production scheduling (such as industrial output value), meteorological data (daily average temperature), holidays (whether it is a working day) and historical carbon emission data; divide the training / test set, optimize parameters such as tree depth and learning rate through grid search, train the model to fit the nonlinear relationship between features and carbon emission values; input real-time feature data (such as daily production planning, weather forecast), output short-term (daily / weekly) carbon emission prediction values, and identify key driving factors (such as the impact of high-temperature weather on air conditioner energy consumption) through feature importance ranking.

[0171] Step S36, accumulate the standardized project energy consumption data and the remaining period energy consumption data to obtain the total annual cycle carbon emission.

[0172] After splicing the occurred standardized project energy consumption data and the predicted remaining period energy consumption data in time sequence, multiplying by the localized carbon emission factor and summing, the total annual cycle carbon emission is obtained, and the total carbon dioxide equivalent emissions of the project in the complete natural year is determined.

[0173] In this embodiment, the Internet of Things gateway collects multi-source energy consumption data of the project site in real time and losslessly, then uses Kalman filtering and a standard time axis to complete data cleaning and time sequence unification, ensuring data quality and comparability; then the provincial model is issued through the federal learning node. The regional energy structure correction coefficient and the annual energy structure basic data provided by the regional model are fused to dynamically generate the localized carbon emission factor, and then the remaining period energy consumption is predicted with the help of historical same period data, and finally the total annual cycle carbon emission is accumulated to ensure high accuracy of nucleic acid results.

[0174] Based on the above-mentioned embodiments of the application, the fourth embodiment of the application is proposed. In the fourth embodiment of the application, the same or similar contents as the above-mentioned embodiments can be referred to the above introduction, and will not be described in detail hereinafter.

[0175] On this basis, please refer to Figure 3 , Figure 3 The flowchart provided by the fourth embodiment of the present application. As shown in Figure 3 After step S20 of calling the hierarchical digital twin sub-model corresponding to the carbon emission management instruction in the pre-constructed multi-level digital twin model system, steps S231-S238 are further included:

[0176] Step S231, if the called hierarchical digital twin sub-model is a first-level digital twin model, a pre-constructed graph generator is called to identify the first-level rule text and extract a first instruction parameter set;

[0177] The first instruction parameter set refers to a set of structured fields extracted from the provincial policy text and can be used to drive the model. The purpose of identifying the provincial policy text and extracting the provincial policy instruction parameter set is to convert the unstructured policy language into numerical instructions that can be directly read by the model.

[0178] In one possible implementation, the graph generator uses a rule template and a deep learning fusion strategy to first lock the chapter by the rule template and then extract entities by the BERT model.

[0179] Step S232, generating a regional energy structure correction coefficient based on the first instruction parameter set;

[0180] The regional energy structure correction coefficient refers to a dimensionless correction coefficient generated by considering factors such as primary energy consumption structure, power import and export, renewable energy proportion, and purchased power emission factor based on the provincial administrative region, which is used to adjust the national default emission factor to a localized factor that conforms to the actual situation of the province, so that the carbon emission results output by the provincial digital twin model reflect both the national unified standard and the local differences, and the accuracy of quota allocation is improved.

[0181] Step S233, based on the federated learning node, obtaining an industrial structure feature vector and an energy intensity feature vector uploaded from a second-level digital twin model, and generating an administrative division emission reduction heat map based on the industrial structure feature vector and the energy intensity feature vector;

[0182] The industrial structure feature vector refers to a high-dimensional vector output by the municipal digital twin model, which reflects the proportion of the value of each industry, the proportion of employment, and the proportion of energy consumption in the city; the energy intensity feature vector refers to a high-dimensional vector output by the municipal digital twin model, which reflects the energy consumption per unit of GDP and the energy consumption per unit of product in the city; and the administrative division emission reduction heat map refers to a visualization layer that takes the provincial administrative region as the base map, takes the grid or city as the unit, and uses color depth to represent the carbon emission reduction potential or the emission reduction urgency.

[0183] In addition, it should be noted that the purpose of obtaining the municipal feature vector and generating the provincial emission reduction thermal map is to realize cross-level knowledge sharing by using federated learning, avoid the out-of-domain of raw data, and at the same time, display the spatial distribution of the provincial emission reduction key points in an intuitive graphical manner.

[0184] In step S234, if the called hierarchical digital twin sub-model is a second-level digital twin model, the atlas generator is called to identify the second-level rule text and extract a second instruction parameter set.

[0185] The municipal policy text is identified and the municipal policy instruction parameter set is extracted, and the municipal differentiated management and control requirements are injected into the model. In one possible implementation, the system directly interfaces with the municipal government OA interface to obtain structured policy data, uses the atlas generator to identify the municipal policy text, and extracts the municipal policy instruction parameter set.

[0186] In step S235, second-level industrial structure data is obtained, the second-level industrial structure data and a second-level geographic information layer are superimposed, and a high-energy-consuming industrial agglomeration area vector boundary is determined.

[0187] The second-level industrial structure data refers to a statistical index of a kind of industry value, energy consumption, employment, and tax in a municipal administrative region; the second-level geographic information layer, i.e., the municipal geographic information layer, refers to a vector layer containing administrative division, land use, road, water system, and ecological red line; and the high-energy-consuming industrial agglomeration area vector boundary refers to a polygon boundary formed by aggregating industrial plots with energy consumption density exceeding the standard through a spatial clustering algorithm.

[0188] In step S236, the district energy structure correction coefficient issued by the first-level digital twin model is called, the district energy structure correction coefficient is split according to the second-level industrial structure data, and a split industry energy structure correction coefficient matrix is generated.

[0189] The provincial district energy structure correction coefficient is decomposed according to the municipal industrial structure weight to obtain a two-dimensional matrix with industry as the row and energy variety as the column, and each element represents the correction coefficient of the industry and energy variety relative to the provincial average level. The provincial policy intensity is accurately transmitted to each industry at the municipal level to improve the fairness of the allocation quota.

[0190] In one possible implementation, the system uses the entropy weight method to calculate the industrial structure weight. Specifically, first, the industrial structure index data (such as the proportion of the value of each industry and energy intensity) is standardized to eliminate dimensional differences; then the information entropy value of each index is calculated, and the smaller the entropy value, the higher the index dispersion degree and the greater the information quantity; then the difference coefficient (1-entropy value) is derived from the entropy value, and the difference coefficient is positively correlated with the weight; finally, the difference coefficient is normalized to obtain the objective weight of each industrial structure index, and the total weight is 1.

[0191] Step S237, if the called hierarchical digital twin sub-model is a third-level digital twin model, the atlas generator is called to identify third-level rule text and extract a third instruction parameter set;

[0192] The system periodically captures policy updates on the district government portal website through a crawler, calls the atlas generator to identify district-level policy text, and extracts a third instruction parameter set, i.e., a district-level policy instruction parameter set.

[0193] It can be understood that the difference between the provincial, municipal, and district policy instruction parameter sets lies in which level of policy text is extracted.

[0194] Step S238, the project information uploaded by the project-level digital twin model is called to generate a visual project list based on the project information.

[0195] The project information refers to the structured data of the project name, project code, construction status, design capacity, real-time energy consumption, real-time carbon emission, allocated quota, remaining quota, and emission reduction progress uploaded by the project-level digital twin model in real time. The visual project list refers to the project list displayed in the form of a card or table on the front end of the district-level digital twin model, which supports multi-dimensional filtering and sorting according to industry, energy consumption, and emission reduction progress.

[0196] In this embodiment, the provincial, municipal, and district policy instruction parameter sets parsed by the atlas generator are converted into district energy structure correction coefficients, industry-specific correction matrices, and visual project lists at different levels, and rely on federal learning nodes to realize cross-level data encryption sharing and spatial heat map dynamic generation. This enables macro policy objectives to be transmitted to project sites, ensuring that zero-carbon targets are quantified synchronously between administrative and market subjects at all levels, ultimately significantly reducing compliance costs and emission reduction risks caused by policy differences and information lags.

[0197] Based on the above embodiments of the application, a fifth embodiment of the application is proposed. In the fifth embodiment of the application, the same or similar contents as the above embodiments can be referred to the above introduction, and will not be described in detail hereinafter.

[0198] In this embodiment, the step S50 of generating a carbon quota scheme according to the carbon quota parameter and the total annual period carbon emission can include steps S51-S53:

[0199] Step S51, calculating the quota gap based on the total annual period carbon emission;

[0200] The quota gap refers to the difference between the total annual carbon emissions and the free quota approved by the competent department. A positive value indicates that the gap needs to be purchased or reduced, and a negative value indicates that the surplus can be sold or carried over. The difference between the total annual carbon emissions and the free quota is calculated by simple difference or weighted difference algorithm, which quantifies the enterprise's compliance pressure and provides clear targets for subsequent emission reduction scheme development.

[0201] In one possible implementation, the system uses a real-time gap dashboard for daily updates. In another possible implementation, the system uses scenario simulation to show the gap changes under different emission reduction efforts.

[0202] Step S52, using a multi-objective optimization algorithm to calculate the quota gap and the carbon quota parameters to generate an initial emission reduction scheme;

[0203] The multi-objective optimization algorithm refers to an improved particle swarm algorithm that targets "minimum economic cost, maximum emission reduction, and shortest investment payback period". The initial emission reduction scheme refers to a set of Pareto optimal solutions that include technical emission reduction, CCER development, quota trading, and green electricity substitution.

[0204] The purpose of using a multi-objective optimization algorithm is to find the optimal or near-optimal compliance path for enterprises under multiple constraints. The effect is to achieve the best balance of cost-benefit-risk while ensuring compliance.

[0205] In one possible implementation, the system uses the NSGA-III algorithm to generate 50 Pareto solutions for enterprises to choose from. Specifically, 50 random solutions containing carbon emissions, cost, energy consumption, and other target functions are initialized, and the non-dominated sorting is used to divide the dominance level and calculate the crowding distance to maintain the diversity of solutions. The reference point is introduced to guide the evolution of the population, and the simulated binary crossover and polynomial mutation are used to generate offspring. The parent and child populations are merged, and the optimal individual is selected based on the reference point correlation. After iteration to convergence, 50 evenly distributed Pareto solutions are output, and enterprises can choose a scheme based on their actual needs (such as emission reduction priority).

[0206] Step S53, modify the initial emission reduction scheme by synchronizing the energy structure correction coefficient through the federal learning node to obtain the carbon quota scheme.

[0207] The latest energy structure correction coefficient is encrypted and pushed to the project-level model through the federal learning node. Based on the correction coefficient changes, the carbon emission factor is recalculated, the marginal cost of emission reduction measures is reevaluated, and the emission reduction scheme is updated. Finally, the final executable scheme is obtained after federal correction, including the carbon quota scheme, technical list, investment budget, time node, transaction plan, and compliance list.

[0208] It can be understood that the synchronization and modification by the federal learning node can ensure that the scheme responds to the provincial power grid structure, renewable energy output and policy adjustment in real time, realizes dynamic optimization, and avoids compliance risks or economic losses caused by outdated schemes.

[0209] In the embodiment, first, the total annual carbon emissions and the accurate quota are calculated to quantify the abstract zero-carbon target into a directly operable carbon asset gap value; then, a multi-objective optimization algorithm is called to generate a Pareto optimal initial emission reduction scheme under multiple constraints; subsequently, the provincial energy structure correction coefficient is synchronized in real time through the federal learning node, the scheme is dynamically corrected, and the final executable carbon quota scheme is output. It not only guarantees compliance, but also minimizes economic cost and policy lag risk, significantly improving the efficiency of carbon asset operation and the agility of zero-carbon target landing.

[0210] The application also provides a carbon emission management device, please refer to Figure 4 , the carbon emission management device comprises:

[0211] The acquisition module 10 is used for acquiring the carbon emission management instruction input by the user;

[0212] The multi-level digital twin model module 20 is used for calling a hierarchical digital twin sub-model corresponding to the carbon emission management instruction in a pre-constructed multi-level digital twin model system, wherein the multi-level digital twin model system is divided into a plurality of hierarchical digital twin sub-models in a manner combining administrative region division and project division;

[0213] The carbon quota scheme generation module 30 is used for collecting project energy consumption data if the called hierarchical digital twin sub-model is a project-level digital twin model, and calculating the total annual carbon emissions based on the project energy consumption data; calling a pre-constructed graph generator to extract carbon quota parameters; and generating a carbon quota scheme according to the carbon quota parameters and the total annual carbon emissions.

[0214] The carbon emission management device provided by the application adopts the carbon emission management method in the above embodiment, and can solve the technical problem of carbon emission management. Compared with the prior art, the carbon emission management device provided by the application has the same beneficial effects as the carbon emission management method provided by the above embodiment, and other technical features in the carbon emission management device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0215] The application provides a carbon emission management device, comprising: at least one processor; and a memory connected with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the carbon emission management method in the above embodiment one.

[0216] Reference will now be made to the following description Figure 5 which shows a structural diagram of a carbon emission management device suitable for implementing embodiments of the application. The carbon emission management device in the embodiments of the application can include, but is not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (Personal Digital Assistant), PADs (Portable Application Description), PMPs (Portable Media Player), vehicle-mounted terminals (for example, vehicle-mounted navigation terminals), and the like, and fixed terminals such as digital TVs, desktop computers, and the like. Figure 5 The carbon emission management device shown is merely an example and should not impose any limitation on the functions and use range of the embodiments of the application.

[0217] As shown in Figure 5 , the carbon emission management device can include a processing apparatus 1001 (for example, a central processing unit, a graphic processing unit, or the like) that can perform various appropriate actions and processes according to programs stored in a read-only memory 1002 or loaded from a storage apparatus 1003 into a random access memory 1004. Various programs and data required for operation of the carbon emission management device are also stored in the random access memory 1004. The processing apparatus 1001, the read-only memory 1002, and the random access memory 1004 are connected to each other through a bus 1005. An input / output interface 1006 is also connected to the bus. Generally, the following systems can be connected to the input / output interface 1006: input apparatuses 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, and the like; output apparatuses 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, and the like; the storage apparatus 1003 including, for example, a magnetic tape, a hard disk, and the like; and a communication apparatus 1009. The communication apparatus 1009 can allow the carbon emission management device to perform wireless or wired communication with other devices to exchange data. Although the carbon emission management device having various systems is shown in the figure, it should be understood that all the systems shown are not required to be implemented or provided. More or fewer systems can be alternatively implemented or provided.

[0218] In particular, according to embodiments of the present disclosure, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer readable medium, the computer program comprising program code for performing the methods illustrated by the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device 1003, or installed from a read-only memory 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the methods of the embodiments of the present disclosure are performed.

[0219] The carbon emission management device provided by the present application adopts the carbon emission management method in the above-mentioned embodiments, and can solve the technical problem of carbon emission management. Compared with the prior art, the carbon emission management device provided by the present application has the same beneficial effects as the carbon emission management method provided by the above-mentioned embodiments, and other technical features in the carbon emission management device are the same as the features disclosed in the previous embodiment method, which will not be repeated here.

[0220] It should be understood that various parts of the present disclosure can be realized by hardware, software, firmware or a combination thereof. In the description of the above-mentioned embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0221] The above is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0222] The present application provides a computer readable storage medium having stored thereon computer readable program instructions (i.e. computer program) for performing the carbon emission management method in the above-mentioned embodiments.

[0223] The computer readable storage medium provided in the application may be, for example, a U disk, but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system or device, or any combination of the above. More specific examples of the computer readable storage medium may include, but are not limited to, an electrical connection with one or more conductive 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 embodiment, the computer readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system or device. The program code contained on the computer readable storage medium can be transmitted by any suitable medium, including but not limited to an electrical wire, an optical cable, an RF (Radio Frequency), and the like, or any suitable combination of the above.

[0224] The above computer readable storage medium may be contained in the carbon emission management device, or may exist separately without being assembled into the carbon emission management device.

[0225] The above computer readable storage medium carries one or more programs, which, when executed by the carbon emission management device, cause the carbon emission management device to: acquire a carbon emission management instruction input by a user; call a hierarchical digital twin sub-model corresponding to the carbon emission management instruction in a pre-constructed multi-level digital twin model system, wherein the multi-level digital twin model system is divided into a plurality of hierarchical digital twin sub-models in a manner combining administrative region division and project division; if the called hierarchical digital twin sub-model is a project-level digital twin model, collect project energy consumption data and calculate annual period carbon emission total based on the project energy consumption data; call a pre-constructed graph generator to extract a carbon quota parameter; and generate a carbon quota scheme according to the carbon quota parameter and the annual period carbon emission total.

[0226] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can 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 the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0227] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0228] The modules involved in the embodiments of the present application can be implemented in software or hardware. In some cases, the name of the module does not constitute a limitation on the module itself.

[0229] The readable storage medium provided by the present application is a computer readable storage medium, which stores computer readable program instructions (i.e. computer programs) for executing the above carbon emission management method, and can solve the technical problem of carbon emission management. Compared with the prior art, the computer readable storage medium provided by the present application has the same beneficial effects as the carbon emission management method provided by the above embodiments, which will not be repeated here.

[0230] The application further provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the carbon emission management method as described above.

[0231] The computer program product provided by the application can solve the technical problem of carbon emission management. Compared with the prior art, the beneficial effects of the computer program product provided by the application are the same as those of the carbon emission management method provided by the above-mentioned embodiments, and are not described here.

[0232] The above only describes some embodiments of the application, and does not limit the patent scope of the application. Any equivalent structural transformation made by using the content of the specification and drawings, or direct / indirect application in other related technical fields under the technical concept of the application is included in the patent protection scope of the application.

Claims

1. A carbon emission management method, characterized in that, The carbon emission management method includes: Obtain user-inputted carbon emission management instructions; A two-dimensional classification rule is defined based on the administrative region level and the project entity attribute. The two-dimensional classification rule includes a geographical boundary classification rule and a project affiliation classification rule. Based on the geographic boundary delineation rules, remote sensing geographic information and rule text are collected, and project point cloud data are collected by UAV oblique photography based on the project affiliation delineation rules. The rule text is bound to the remote sensing geographic information to construct a first-level digital twin model, a second-level digital twin model, and a third-level digital twin model, respectively. Federated learning nodes are deployed in a distributed manner in the first-level digital twin model, the second-level digital twin model, and the third-level digital twin model; A project-level digital twin model is constructed based on the project point cloud data. A lightweight data upload interface is developed. The federated learning node is connected based on the lightweight data upload interface to obtain a multi-level digital twin model system. In a pre-constructed multi-level digital twin model system, the hierarchical digital twin sub-model corresponding to the carbon emission management instruction is invoked. The multi-level digital twin model system is divided into multiple hierarchical digital twin sub-models by combining administrative region division and project division. If the called hierarchical digital twin sub-model is a project-level digital twin model, then project energy consumption data is collected, and the annual cycle carbon emissions are calculated based on the project energy consumption data. The carbon quota parameters are extracted by calling a pre-built graph generator. This step includes: calling the pre-built graph generator to traverse the rule text and extract the core carbon asset parameters, wherein the rule text includes first-level rule text, second-level rule text and third-level rule text; performing logical verification on the core carbon asset parameters and associating them with the industry benchmark value to which the project belongs to obtain the carbon quota parameters. A carbon quota scheme is generated based on the carbon quota parameters and the annual cycle carbon emission total. The step of calling the hierarchical digital twin sub-model corresponding to the carbon emission management directive in the pre-constructed multi-level digital twin model system includes: Parse the carbon emission management instructions and extract the hierarchical identification parameters; Based on the hierarchical identifier parameters, the multi-level digital twin model system is retrieved, and the hierarchical digital twin sub-models are located. A call request is sent to the hierarchical digital twin sub-model to invoke the hierarchical digital twin sub-model, and the rule text corresponding to the hierarchical identifier parameter is rendered in the visualization interface.

2. The carbon emission management method as described in claim 1, characterized in that, If the hierarchical digital twin sub-model invoked is a project-level digital twin model, the steps of collecting project energy consumption data and calculating the annual cycle carbon emissions based on the project energy consumption data include: The IoT gateway based on the project-level digital twin model collects project energy consumption data; The project energy consumption data is unified to a standard time axis, and Kalman filtering is used to remove outliers from the project energy consumption data to obtain standardized project energy consumption data. The annual energy structure basic data is retrieved from the third-level digital twin model, and the energy structure correction coefficients issued by the first-level digital twin model through the federated nodes are obtained. The carbon emission factor is obtained by multiplying the annual basic energy structure data and the energy structure correction coefficient. By calling historical data from the same period and predicting energy consumption data for the remaining period based on the historical data from the same period and the carbon emission factor; By summing the energy consumption data of the standardized project and the energy consumption data of the remaining time period, the total annual carbon emissions are obtained.

3. The carbon emission management method as described in claim 1, characterized in that, The step of calling the hierarchical digital twin sub-model corresponding to the carbon emission management directive within the pre-constructed multi-level digital twin model system further includes: If the called hierarchical digital twin sub-model is a first-level digital twin model, then the pre-built graph generator is called to identify the first-level rule text and extract the first instruction parameter set; Generate regional energy structure correction coefficients based on the first instruction parameter set; Based on federated learning nodes, the industrial structure feature vector and energy intensity feature vector uploaded from the second-level digital twin model are obtained, and an administrative division emission reduction heat map is generated based on the industrial structure feature vector and the energy intensity feature vector. If the called hierarchical digital twin sub-model is a second-level digital twin model, then the graph generator is called to identify the second-level rule text and extract the second instruction parameter set; Acquire second-level industrial structure data, overlay the second-level industrial structure data and the second-level geographic information layer, and determine the vector boundary of the high-energy-consuming industry cluster area; The regional energy structure correction coefficients issued by the first-level digital twin model are invoked, and the regional energy structure correction coefficients are split according to the second-level industrial structure data to generate an industry-specific energy structure correction coefficient matrix. If the called hierarchical digital twin sub-model is a third-level digital twin model, then the graph generator is called to identify the third-level rule text and extract the third instruction parameter set; The system retrieves project information uploaded from a project-level digital twin model and generates a visual project list based on that information.

4. The carbon emission management method as described in claim 1, characterized in that, The step of generating a carbon quota scheme based on the carbon quota parameters and the annual cycle total carbon emissions includes: The quota gap is calculated based on the total annual carbon emissions. A multi-objective optimization algorithm is used to calculate the quota gap and the carbon quota parameters to generate an initial emission reduction plan; By synchronizing the energy structure correction coefficients through federated learning nodes, the initial emission reduction scheme is modified to obtain a carbon quota scheme.

5. A carbon emission management system, characterized in that, The system includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the carbon emission management method as described in any one of claims 1 to 4.

6. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the carbon emission management method as described in any one of claims 1 to 4.

7. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the carbon emission management method as described in any one of claims 1 to 4.

Images