Cloud edge collaboration based multi-energy form unified carbon metering system and method

Abstract

The application provides a multi-energy form unified carbon metering system and method based on cloud edge cooperation, which comprises an edge computing terminal arranged at an energy consumption site and a cloud accounting platform in asynchronous communication connection with the edge computing terminal. The edge computing terminal comprises a multi-energy metering and collecting module, a semantic feature marking module, a fragmented encryption caching module and an asynchronous uploading module. The multi-energy metering and collecting module is used for collecting energy consumption parameters of multi-energy media in real time. The semantic feature marking module is used for automatically generating multi-dimensional feature marks and binding the feature marks with corresponding energy consumption parameters to form feature-based metering data. The fragmented encryption caching module is used for caching the feature-based metering data and independently fragmenting and encrypting the cached data according to a preset time fragmentation granularity to obtain encrypted fragmented data. The asynchronous uploading module is used for batch uploading the encrypted fragmented data to the cloud accounting platform according to time sequence. The application can realize weak network adaptation of multi-energy carbon metering results.

Description

Technical Field

[0001] This invention relates to the fields of carbon emission accounting, energy Internet of Things and edge computing technology, and in particular to a unified carbon metering system and method based on cloud-edge collaboration for multiple energy forms. Background Technology

[0002] With the deepening implementation of the dual-carbon strategy, accurate carbon emission measurement has become the core foundation for carbon verification, carbon trading, and energy management. Carbon emission accounting technology is gradually evolving from traditional static factor accounting to dynamic time-sharing and region-based accounting, and is beginning to explore a distributed metering architecture that combines cloud and edge computing. However, when facing real-world energy consumption scenarios with multiple energy forms and complex underlying equipment, existing technologies still have significant technical limitations in areas such as semantic integrity of data acquisition, transmission reliability under network fluctuations, standardized matching of accounting models, and retention and traceability of historical data.

[0003] At the underlying data acquisition stage, existing edge computing terminals or metering instruments often only possess basic data reading and reporting functions. The energy consumption values ​​they collect typically only include simple device identification and timestamp information. This type of raw data lacks the necessary semantic environment for carbon accounting, failing to natively characterize the physical flow of energy at the source of collection, nor reflecting the specific technological scenarios in which the energy-consuming equipment operates. This leads to the cloud requiring significant resources for manual configuration or reliance on static ledger mapping when defining accounting boundaries. This not only reduces the efficiency of automated accounting but also easily causes problems such as blurred accounting boundaries, duplicate calculations, or data omissions when the production site topology or process flow undergoes dynamic adjustments.

[0004] In terms of data communication and transmission architecture, existing cloud-edge collaborative carbon metering systems generally tend to build long-term connection links with high real-time performance, which places high demands on the stability of the on-site network environment. In scenarios with weak networks and frequent network outages, such as industrial sites, underground pipelines, or remote stations, the local caching mechanisms of conventional terminals are relatively rigid, lacking flexible adaptive adjustment strategies and fine-grained encryption and isolation methods, making cached data vulnerable to security risks.

[0005] In the cloud-based accounting stage, current energy carbon emission models are mostly fragmented, with source and usage accounting for energy separated, and model calling rules lacking a unified and strong constraint mechanism. Because a strict joint query and matching benchmark has not been established across spatial, temporal, and energy type dimensions, multiple accounting methods may coexist when processing complex underlying data. The same set of measurement data may be applied with different emission factors or accounting formulas at different statistical nodes, leading to ambiguity and weakening the objectivity and standardization of carbon emission accounting results.

[0006] In the retention and traceability of accounting results, conventional data storage solutions mostly only provide single-dimensional anti-tampering protection for the original collected values ​​or the final accounting results. However, carbon footprint accounting has a strong rule dependency, and the emission factor library and accounting benchmark formula are constantly updated over time. Existing technologies fail to structurally and deeply bind the rule version, calculation model, and original feature data at the time of data generation. Once the accounting rules in the cloud are updated, the system often recalculates the old data according to the new rules when performing historical data backtracking, causing historical carbon emission rating results to drift. It is impossible to achieve a true reproduction of historical accounting logic without the current rule library configuration, making it difficult to meet the stringent requirement of the immutability of historical rating standards in carbon verification business. Summary of the Invention

[0007] The purpose of this invention is to provide a unified carbon metering system and method based on cloud-edge collaboration for multiple energy forms, in order to solve the above-mentioned problems and achieve absolute uniqueness of carbon metering results for multiple energy sources, weak network adaptability, and full-chain traceability.

[0008] This invention proposes a unified carbon metering system for multiple energy forms based on cloud-edge collaboration, including an edge computing terminal deployed at the energy consumption site and a cloud accounting platform that establishes an asynchronous communication connection with the edge computing terminal; The edge computing terminal includes a multi-energy metering and acquisition module, a semantic feature tagging module, a fragmented encryption cache module, and an asynchronous upload module; The multi-energy metering and acquisition module is used to collect energy consumption parameters of multiple energy media in real time; The semantic feature tagging module is used to automatically generate multi-dimensional feature tags locally on the edge computing terminal and bind the multi-dimensional feature tags with the corresponding energy consumption parameters to form characteristic metering data. The multi-dimensional feature tags include at least application scenario semantic features and energy flow features used to define the accounting boundary. The sharded encryption cache module is used to cache characteristic measurement data and perform independent sharded encryption processing on the cached data according to the preset time sharding granularity to obtain encrypted sharded data. The encryption key for each time shard is dynamically derived based on the basic key of the edge computing terminal and the time base parameter of the corresponding time shard. The asynchronous upload module is used to upload encrypted fragmented data to the cloud computing platform in batches according to time sequence when preset communication conditions are met; The cloud-based accounting platform includes a time-series verification module and a decoupled accounting module; The timing verification module is used to receive encrypted fragmented data and verify the continuity of the received data based on the fragment timing. The decoupled accounting module is used to decrypt the verified fragmented data to obtain characteristic metering data, extract the application scenario semantic features and energy flow features from the characteristic metering data, and use the extracted features as input parameters to automatically match the corresponding energy carbon emission model for accounting.

[0009] In one embodiment, when the semantic feature tagging module binds multidimensional feature tags to corresponding energy consumption parameters, it is specifically used for: During the data interruption processing stage of the underlying processing unit configured in the edge computing terminal, the multidimensional feature markers are converted into a binary mask with a preset bit width. The binary mask is directly concatenated with the corresponding energy consumption parameter values ​​and encapsulated into an indivisible fixed-length data frame to form characteristic measurement data.

[0010] In one embodiment, when the semantic feature labeling module automatically generates application scenario semantic features and energy flow direction features locally on the edge computing terminal, it specifically includes: When generating semantic features for application scenarios, at least one of the following methods is used: By reading the operating status register data of the underlying programmable logic controller at the energy consumption site through the industrial standard communication protocol, the equipment start-up and shutdown status and process stage identifiers are extracted and converted into semantic features of the application scenario. Query the local mapping configuration table pre-set on the edge computing terminal, hard-bind the physical port identifier of the current acquisition channel with the corresponding application scenario label, and generate application scenario semantic features; When generating energy flow characteristics, at least one of the following methods can be used: By analyzing the positive and negative metering register data of the bidirectional metering instrument or the positive and negative polarity signals of the underlying acquisition sensor, the physical direction of energy flow is dynamically determined to generate energy flow characteristics. Based on the predefined energy flow topology attributes of the acquisition channel in the preset local mapping configuration table, the energy flow direction characteristics representing the energy inflow or outflow of the current acquisition node are directly mapped and generated.

[0011] In one embodiment, the time base parameter includes the start timestamp of the corresponding time slice; When the segmented encryption cache module dynamically derives and generates encryption keys for each time segment, it specifically uses: a key derivation function based on hash-based message authentication code to take the basic key stored in the edge computing terminal security chip as input key material, and to perform irreversible hashing operation using the start timestamp and the unique device identifier of the edge computing terminal as salt parameters to generate a symmetric encryption key exclusive to the current time segment. When the decoupled accounting module decrypts the verified fragmented data, it specifically derives a decryption key based on the unique device identifier of the edge computing terminal and the start timestamp of the corresponding fragment, and uses the decryption key to decrypt the verified fragmented data.

[0012] In one embodiment, the timing verification module verifies the continuity of received data based on fragment timing, and when a fragment is detected to be missing, it performs the following retransmission interaction mechanism with the asynchronous upload module: After the communication connection is restored, the asynchronous upload module prioritizes sending an index data packet containing all cached time segment identifiers to the cloud computing platform. The timing verification module compares the historical received records in the cloud with the index data packet, generates a missing list containing missing time fragment identifiers through difference operation, and sends the missing list to the asynchronous upload module; The asynchronous upload module re-uploads the corresponding encrypted fragment data according to the received missing list, and clears the corresponding locally uploaded cached data after receiving verification confirmation from the cloud computing platform.

[0013] In one embodiment, the sharded encryption cache module is further configured to dynamically adjust the preset time sharding granularity based on the real-time network quality of the edge computing terminal, specifically for: The real-time network quality parameters between the edge computing terminal and the cloud computing platform are obtained, including network packet loss rate, uplink bandwidth, and round-trip latency. The current network quality level is evaluated based on the real-time network quality parameters, and the target time fragmentation granularity is dynamically determined according to the preset quality level and fragmentation granularity mapping rules. The lower the network quality level, the larger the corresponding target time fragmentation granularity, and the target time fragmentation granularity is not greater than the preset maximum fragmentation granularity upper limit. After the data processing cycle of the current time segment ends, the target time segment granularity is applied, and the granularity identifier representing the target time segment granularity is written into the data frame header of the next segment of data to be synchronized to the cloud computing platform.

[0014] In one embodiment, the multidimensional feature marker further includes spatial location encoding, acquisition timestamp, and energy type encoding; The decoupled accounting module internally constructs a spatiotemporal carbon emission factor library, which constructs a three-dimensional primary key constraint mechanism based on spatial, temporal, and energy type dimensions. When automatically matching the corresponding energy carbon emission model for calculation, the decoupled accounting module uses the extracted spatial location code, collection timestamp, and energy type code as the joint query primary key to match and extract emission factor versions and accounting benchmark formulas with unique correspondences from the spatiotemporal carbon emission factor library.

[0015] In one embodiment, the decoupled accounting module also contains a source accounting model library and a usage accounting model library covering multiple energy media. When the decoupled accounting module automatically matches the corresponding energy carbon emission model for accounting, it is specifically used for: If the extracted energy flow characteristics are the input identifier, the source accounting model corresponding to the energy type code will be automatically called to calculate the transmission carbon emissions covering the energy extraction, conversion to terminal access process; If the extracted energy flow characteristics are used as output identifiers, the extracted application scenario semantic characteristics will be combined to automatically call the usage accounting model corresponding to the application scenario and calculate the terminal carbon emission allocation under the specific energy consumption scenario.

[0016] In one embodiment, the cloud-based accounting platform further includes an evidence storage and traceability module, used to perform three-layer structured chain hash evidence storage on the accounting results obtained by the decoupled accounting module, specifically including: The first layer extracts feature summaries from individual characteristic measurement data and generates raw data hashes. The second layer extracts feature summaries from the emission factor version and corresponding energy carbon emission model called in the current accounting, and concatenates them with the original data hash to generate a rule binding hash. The third layer extracts feature summaries from the accounting results, intermediate parameters of the calculation process, and cloud computing timestamps, and binds them with the rules to perform hash concatenation calculation to generate the final evidence storage hash. All final evidence hashes within the same preset time granularity are chained together in chronological order to form a time-slice hash chain and are persistently stored.

[0017] This invention also proposes a unified carbon metering method for multiple energy forms based on cloud-edge collaboration, applied to the unified carbon metering system for multiple energy forms based on cloud-edge collaboration as described above, including the following steps: Edge computing terminals deployed at energy consumption sites collect energy consumption parameters of multiple energy media in real time; Multidimensional feature tags are automatically generated locally on the edge computing terminal, and the multidimensional feature tags are bound to the corresponding energy consumption parameters to form characteristic metering data. The multidimensional feature tags include at least application scenario semantic features and energy flow features used to define the accounting boundary. The edge computing terminal caches characteristic metering data and performs independent fragment encryption processing on the cached data according to the preset time fragment granularity to obtain encrypted fragment data. The encryption key for each time fragment is dynamically derived based on the edge computing terminal's basic key and the time base parameter of the corresponding time fragment. When the preset communication conditions are met, the edge computing terminal will upload the encrypted fragmented data to the cloud computing platform in batches according to the time sequence. The cloud-based accounting platform receives encrypted fragmented data and verifies the continuity of the received data based on the fragmentation time sequence. The cloud-based accounting platform decrypts the verified data shards to obtain characteristic metering data, extracts application scenario semantic features and energy flow features from the characteristic metering data, and uses the extracted features as input parameters to automatically match the corresponding energy carbon emission model for accounting.

[0018] In one embodiment, the cloud-edge collaborative multi-energy form unified carbon metering method further includes a historical data backtracking and verification step: After the accounting is completed, the feature summary of the single characteristic measurement data, the energy carbon emission model called, and the accounting result are extracted and spliced ​​layer by layer to generate the final evidence hash. The final evidence hashes within the same preset time segment granularity are then chained together in chronological order to form a time slice hash chain. A unique traceability code is generated for each completed accounting result, and the characteristic measurement data, the energy carbon emission model invoked at that time, and the hash evidence record of the accounting result are bound and stored with the traceability code; Upon receiving a reproduction request carrying the traceability code, the historical energy carbon emission model and historical emission factors corresponding to the historical accounting are retrieved based on the bound evidence records to reproduce the accounting process; The hash value generated by the reproduction is compared with the final evidence hash stored in the corresponding time slice hash chain to verify that the historical carbon emission accounting results and accounting rating standards have not been tampered with.

[0019] Compared with existing technologies, the beneficial effects of the cloud-edge collaborative multi-energy unified carbon metering system and method of the present invention are as follows: 1) This invention, by configuring a semantic feature tagging module on an edge computing terminal, can automatically generate multi-dimensional feature tags containing application scenarios and energy flow directions, convert them and quantitative data into binary masks, and directly concatenate and encapsulate them into indivisible fixed-length data frames. This feature pre-collection mechanism endows the measurement data with a complete carbon accounting semantic environment at the source of data generation, avoiding the boundary ambiguity and resource consumption caused by manual annotation in the cloud. At the same time, it ensures a strong binding between feature tags and quantitative data at the physical level, improving the reliability of the original data and the standardization of automated cloud accounting. In addition, combined with the multi-dimensional feature tagging system, it realizes the automatic definition and full-link coverage of accounting boundaries, avoiding duplicate emission calculations and boundary omissions.

[0020] 2) This invention employs a segmented encryption caching and dynamic key derivation mechanism. For each time segment, a unique symmetric encryption key is dynamically derived using the edge computing terminal's basic key combined with the corresponding segment's start timestamp, achieving secure isolation of "one key per segment" and preventing data risks caused by static key leakage. Simultaneously, the system adaptively and dynamically adjusts the time segment granularity by real-time monitoring of network packet loss rate, uplink bandwidth, and round-trip latency. The asynchronous, non-real-time communication architecture reduces the system's dependence on the real-time network environment. The accompanying time-segmented encryption transmission mechanism is adaptable to scenarios involving network outage caching and breakpoint resumption, ensuring data confidentiality and integrity even under non-real-time transmission conditions. It also ensures complete retention of metering data in scenarios with poor network coverage, making it suitable for remote factories, outdoor stations, underground pipelines, and other weak network environments, expanding the system's applicability and lowering deployment and maintenance thresholds.

[0021] 3) This invention constructs a three-dimensional primary key-constrained spatiotemporal carbon emission factor library and a source and use accounting model library covering various energy media on a cloud-based accounting platform. The decoupled accounting module automatically uses features such as spatial location coding, collection timestamps, energy type coding, and energy flow direction in the characteristic metering data as the primary key for joint queries to match emission factor versions, accounting benchmark formulas, and specific accounting models with unique correspondences. This eliminates the ambiguity of multiple standards caused by manual intervention or misconfiguration at the underlying mechanism, and realizes the automatic and accurate definition of the carbon emission accounting boundaries for different energy forms throughout their entire life cycle, improving the objectivity and standardization of the accounting process.

[0022] 4) This invention introduces a three-layer structured chained hash evidence storage mechanism. It extracts and concatenates the original feature data, the emission factor version and accounting model of the current call, and the final accounting result and intermediate parameters layer by layer. The evidence hashes in the same time slice are concatenated in time sequence to form a time slice hash chain for persistent storage. The original input data, the accounting rule version at a specific time and the calculation result are deeply structured and bound at the cryptographic level. This ensures that even if the official factor library or accounting rules are updated in the future, the system can still reproduce the historical accounting logic based on the traceability code. This prevents unauthorized drift of historical carbon rating results and ensures the long-term traceability and credibility of carbon footprint accounting results. It can directly support application scenarios with high requirements for accounting credibility, such as carbon verification and carbon trading. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the structure of a unified carbon metering system based on cloud-edge collaboration for multiple energy forms according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the module structure of an edge computing terminal according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the module structure of a cloud computing platform according to an embodiment of the present invention; Figure 4 This is a flowchart illustrating a unified carbon metering method for multiple energy forms based on cloud-edge collaboration, according to an embodiment of the present invention. Figure 5 This is a schematic diagram of the three-dimensional primary key constraint logic of a spatiotemporal carbon emission factor library according to an embodiment of the present invention; Figure 6 This is a time-series logic diagram illustrating the binding of the accounting rule version and the metering data time period in an embodiment of the present invention.

[0024] Figure Labels

[0025] 100. Edge computing terminal; 110. Multi-energy metering and acquisition module; 120. Semantic feature tagging module; 130. Segmented encryption cache module; 140. Asynchronous upload module; 200. Cloud accounting platform; 210. Timing verification module; 220. Decoupled accounting module; 230. Evidence storage and traceability module. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention more readily understood, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that while many specific details are set forth in the following description to provide a thorough understanding of the invention, the invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the spirit of the invention; therefore, the invention is not limited to the specific embodiments disclosed below.

[0027] Secondly, the phrase "an embodiment" or "a particular embodiment" in this application refers to a specific feature, structure, or characteristic that can be included in at least one implementation of the present invention. The phrases "in one embodiment" and "a particular embodiment" appearing in different places in this specification do not all refer to the same embodiment, nor are they embodiments that are mutually exclusive, either alone or selectively, with other embodiments. The terms "comprising" or "including" indicate the presence of the claimed feature but do not exclude the presence of one or more other features. The term "and / or" as used in this application includes any and all combinations of one or more of the related listed items.

[0028] This invention proposes a unified carbon metering system based on cloud-edge collaboration and multiple energy forms. See [link to relevant documentation]. Figure 1 The system includes an edge computing terminal 100 deployed at energy consumption sites and a cloud-based accounting platform 200 that establishes an asynchronous communication connection with the edge computing terminal 100. As a concrete manifestation of the system's edge computing terminal layer, the edge computing terminal 100 is deployed at energy input / output nodes (such as power distribution circuits, gas pipelines, heating interfaces, hydrogen filling ports, etc.) at energy consumption sites. It is responsible for real-time acquisition, feature labeling, and local storage of multi-energy data, without relying on a real-time network, and is the core carrier for the feature-based data acquisition of this invention. The edge computing terminal 100 includes a multi-energy metering acquisition module 110, a semantic feature labeling module 120, a sharded encrypted cache module 130, and an asynchronous upload module 140. The cloud-based accounting platform 200 includes a timing verification module 210, a decoupled accounting module 220, and a evidence storage and traceability module 230. Through this overall architecture, the edge computing terminal 100 achieves native multi-dimensional feature labeling and secure offline caching, while the cloud-based accounting platform 200 performs timing verification and unique method mapping based on reported data, thereby constructing an end-to-end closed-loop carbon metering network.

[0029] like Figure 2As shown, the multi-energy metering and acquisition module 110 serves as a data input source, used to collect energy consumption parameters of multiple energy media in real time. In specific implementation scenarios, the multiple energy media cover various forms such as electricity, natural gas, heat (e.g., steam), cold energy, and hydrogen energy. The multi-energy metering and acquisition module 110 is backward compatible with various underlying sensors, such as electricity meters, flow meters, and heat meters, ensuring that the acquisition accuracy meets national metrological standards. For example, in an industrial park scenario, this module can be deployed in the 10kV incoming power distribution room to connect to each production circuit, or deployed at the steam outlet of a centralized heating station and the interface of the main gas pipeline to obtain instantaneous and cumulative energy consumption in real time.

[0030] After data acquisition, the data flows into the semantic feature labeling module 120. The semantic feature labeling module 120 is used to automatically generate multi-dimensional feature labels locally on the edge computing terminal 100, and bind the multi-dimensional feature labels with the corresponding energy consumption parameters to form characteristic metering data. The multi-dimensional feature labels include at least the application scenario semantic features used to define the accounting boundary (such as energy consumption / supply scenario labels such as industrial production, building heating, and green electricity grid connection) and energy flow characteristics (such as input / output identifiers to distinguish whether energy flows into or out of the node). In addition, they also include basic attributes (such as energy type code and unit of measurement), spatiotemporal attributes (such as millisecond-level acquisition timestamp and spatial location code bound to the terminal), and identity attributes (such as unique device ID of the terminal, data verification code, and metering accuracy level). In one embodiment of the present invention, the semantic feature tagging module 120, when binding multi-dimensional feature tags with corresponding energy consumption parameters, specifically performs the following: during the data interrupt processing stage of the underlying processing unit (such as an MCU) configured in the edge computing terminal (e.g., utilizing the low latency characteristics of the microcontroller's underlying interrupt service routine), it converts the multi-dimensional feature tags into a binary mask of a preset bit width (e.g., 64 bits or 128 bits); it directly performs a bit-by-bit concatenation operation between the binary mask and the corresponding energy consumption parameter value data, and encapsulates it into an indivisible fixed-length data frame. This mechanism locks the value with the calculation semantics at the hardware level, avoiding the risk of tampering that may be caused by post-processing in the cloud.

[0031] An embodiment of the present invention discloses a unified carbon metering system for multiple energy forms based on cloud-edge collaboration. In terms of semantic feature generation, it employs a dual-path approach of "dynamic reading + static mapping." To accurately capture accounting semantics, the semantic feature labeling module 120 uses at least one of the following methods when generating application scenario semantic features: The first method is dynamic acquisition, which reads the operating status register data of the underlying programmable logic controller at the energy consumption site through industrial standard communication protocols such as Modbus-RTU or OPC UA, extracts the equipment start / stop status and process stage identifiers, and converts them into application scenario semantic features such as industrial production and building heating; the second method is static mapping, which queries the local mapping configuration table pre-loaded on the edge computing terminal and hard-binds the physical port identifier of the current acquisition channel with the corresponding application scenario label. The local mapping configuration table is a static entry pre-loaded into memory upon power-on, ensuring the normal operation of the feature labeling function in a network-off environment. When generating energy flow characteristics, at least one of the following methods is used: First, by parsing the positive and negative metering register data of bidirectional metering instruments (such as bidirectional electricity meters) or the positive and negative polarity signals of the underlying acquisition sensors, the physical direction of energy flow is dynamically determined to generate energy flow characteristics; Second, based on the predefined energy flow topology attributes of the acquisition channels in the preset local mapping configuration table, the identifier representing the current acquisition node of energy "input" or "output" is directly mapped and generated, thereby clearly defining the boundary of energy flow.

[0032] After the data is marked, it enters the sharding encryption cache module 130. The sharding encryption cache module 130 uses its internally configured non-volatile storage unit to cache the characteristic metering data, and performs independent sharding encryption processing on the cached data according to the preset time sharding granularity to obtain the encrypted sharded data. In the sharding key generation logic of the unified carbon metering system based on cloud-edge collaboration of this invention, each shard has an independent key, and the encryption key for each time shard is dynamically derived. Specifically, the time base parameters include the start timestamp of the corresponding time shard; the sharding encryption cache module 130 uses the Hash Message Authentication Code (HMAC) based key derivation function (HKDF), takes the basic key stored in the edge computing terminal security chip (such as an SE chip or TPM module) as the input key material (this basic key is preset in the terminal security chip and cannot be exported), and uses the start timestamp and the unique device identifier of the edge computing terminal as salt parameters to perform irreversible hash operation (e.g., HMAC-SHA256 irreversible operation) to generate the symmetric encryption key exclusive to the current time shard. Different time slices have independent encryption keys, achieving forward and backward security. The derived slice keys are not persistently stored locally on the edge computing terminal; the key copy in memory is cleared after encryption.

[0033] Considering the fluctuations in network quality in actual deployment environments, the sharded encryption cache module 130 of one embodiment of the present invention is also used to dynamically adjust the time sharding granularity according to the real-time network quality. The specific logic is as follows: Quality parameters such as network packet loss rate, uplink bandwidth, and round-trip latency between the edge computing terminal and the cloud are obtained; the current network quality level is assessed; and the target sharding granularity is determined according to a preset mapping rule between the quality level and the sharding granularity (e.g., sharding granularity is 10 minutes / shard when the network level is high-quality, and expanded to 1-2 hours / shard when the network level is weak), without exceeding the maximum upper limit. The above adjustment only takes effect after the current sharding ends, and the granularity identifier is written into the data frame header of the next shard and synchronized to the cloud, thereby ensuring end-to-end compatibility with encryption, verification, and evidence storage logic. The design of the local non-volatile storage unit supports at least 30 days of network outage recovery, ensuring that data is not lost and time-sequence is not disrupted under extremely weak network conditions.

[0034] The asynchronous upload module 140 is used to asynchronously upload encrypted fragmented data in batches to the cloud computing platform in time sequence when preset communication conditions are met (such as network recovery, reaching the set daily scheduled upload period, or receiving an upload command from the cloud). After entering the cloud platform layer, the time sequence verification module 210 receives the fragmented data (each encrypted fragment is accompanied by a fragment sequence number, time slice identifier, and message authentication code) and verifies the continuity of the received data and the legality of the time slice based on the fragment time sequence. In the cloud-edge communication logic, the cloud platform only actively sends non-real-time commands such as accounting rule version updates and terminal configuration parameters, and does not require the terminal to respond immediately. In one embodiment of the present invention, the multi-energy form unified carbon metering system based on cloud-edge collaboration adopts a closed-loop process of "index pre-verification - difference set identification - targeted retransmission" for missing fragment identification and retransmission. When the time sequence verification module 210 detects a missing fragment, it executes the following retransmission interaction mechanism with the asynchronous upload module 140: After the network is restored, the asynchronous upload module 140 prioritizes sending a lightweight index data packet containing all cached time fragment identifiers to the cloud computing platform. The index data packet only contains time fragment identifiers and does not contain the metering data itself. Each fragment is assigned a unique identifier, which consists of "device ID + timestamp + granularity identifier". The cloud compares the local received set history with the index data packet, generates a missing list containing missing time fragment identifiers through difference operation, and sends it down. The terminal retransmits the corresponding encrypted fragment data according to the received list, without retransmitting the entire data, ensuring that the data sequence is not disordered and the content is not tampered with after non-real-time transmission. After receiving the secondary verification confirmation from the cloud computing platform, the terminal clears the corresponding locally uploaded cached data. This mechanism better adapts to the needs of batch interrupted resume transmission without requiring a long connection.

[0035] After the data integrity verification is passed, the decoupling accounting module 220 performs the core decryption and calculation work. In one embodiment of the present invention, the decoupling accounting module 220 uses the terminal device ID and the corresponding fragment start timestamp to derive a decryption key from the cloud through the key management system synchronization parameters. The key is transmitted without passing through the communication link, and the decryption key is used to decrypt the fragment data. Combined with... Figure 3 and Figure 5 As shown, the decoupled accounting module 220 serves as the core accounting and management hub of the system. It internally constructs a spatiotemporal carbon emission factor library module and an energy full-link model library module, and has a built-in unique carbon accounting engine module as the core computing unit, which together support the core capabilities of "unique calculation, complete model, and traceable results".

[0036] like Figure 5 As shown in the three-dimensional coordinate system, the spatiotemporal carbon emission factor library module constructs a three-dimensional primary key constraint mechanism based on spatial dimension (X-axis), time dimension (Y-axis), and energy type dimension (Z-axis) to achieve a strong unique mapping. Specifically, the spatial dimension is divided according to administrative regions or latitude and longitude grids to match the emission characteristics of energy production, transportation, and conversion in different regions; the time dimension is updated in hourly / daily / monthly hierarchical manner to adapt to the time-of-use carbon intensity fluctuations of energy sources such as electricity (e.g., the time-of-use green electricity ratio of the power grid); the energy type covers all categories such as electricity, fossil fuels, heat, cooling, and hydrogen, with each energy type corresponding to a unique accounting formula and factor set. Given a unique combination of "spatiotemporal coordinates + energy type", only a unique emission factor version, accounting formula, and boundary definition can be matched. The decoupled accounting module 220 uses the extracted spatial location code, collection timestamp, and energy type code as the joint query primary key to match and extract emission factor versions and accounting benchmark formulas with unique correspondences in the database, eliminating the ambiguity of "multiple calibers and multiple standards" in the carbon measurement process from the bottom up.

[0037] Meanwhile, the energy full-link model library module establishes complete source and usage models for each type of energy, achieving standardized and full-link accounting boundaries. Each model corresponds one-to-one with terminal feature markers, enabling automatic triggering of model calls. The decoupled accounting module 220 extracts additional semantic features from the terminal for matching: if the energy flow direction feature is an input identifier, it automatically calls the source accounting model corresponding to the energy type code to calculate the transmission carbon emissions covering the energy extraction, processing, transportation, conversion, and terminal access process (e.g., the electricity source model includes the proportion of generation-side structure, transmission and distribution loss emissions, green electricity traceability identifiers, etc.; the natural gas source model includes extraction emissions, pipeline / LNG transportation emissions, etc.). If the extracted energy flow direction feature is an output identifier, it automatically calls the usage accounting model corresponding to the application scenario, combined with the extracted application scenario semantic features, to calculate the terminal carbon emission allocation under a specific energy consumption scenario. The usage models are divided into subcategories such as industrial production, building operation and maintenance, transportation, and public services according to energy consumption scenarios, clearly defining the accounting boundaries, emission allocation rules, and statistical calibers for different uses. The model is bound to energy type and scene label one by one. After the terminal uploads the feature label, the system automatically matches the complete accounting boundary to avoid duplicate calculation or omission.

[0038] The decoupled carbon accounting module 220's built-in unique carbon accounting engine strictly adheres to the principles of "unique input, unique method, and unique result": after receiving raw data with feature tags uploaded by the terminal, it automatically extracts spatiotemporal parameters, energy types, and scenario labels; matches the corresponding version of the spatiotemporal emission factor and the energy full-link model; and calls a unique accounting formula for calculation. This mechanism ensures that repeated calculations under the same batch of raw data and the same version of accounting rules yield completely consistent results, eliminating ambiguity due to multiple interpretations. This fully automatic model matching mechanism requires no manual configuration, guaranteeing a complete closed loop of the accounting boundary. Additionally, edge terminals can selectively integrate lightweight local carbon accounting capabilities, supporting real-time viewing of rough carbon emission data on-site. The cloud remains the sole authoritative accounting entry point, and the final result is based on cloud-based calculations.

[0039] The evidence storage and traceability module 230 takes over the version management of accounting results. Combined with... Figure 6 The rule version and data time period binding logic shown in this invention, in one embodiment of the evidence storage and traceability module, performs a three-layer structured chain hash evidence storage on the accounting results: The first layer uses a single raw energy data with a feature tag as the smallest unit, extracting a feature digest (e.g., using the SHA-256 algorithm) from the "quantity value + full feature tag" to generate the raw data hash; the second layer uses the emission factor version called in the current accounting (e.g., ...) Figure 6The process involves three layers: First, extracting feature summaries from the V1.0 version of the T1 time slot, the corresponding energy carbon emission model, and the accounting formula identifier. These summaries are then concatenated with the original data hash to generate a rule-binding hash. Second, extracting feature summaries from the accounting results, intermediate parameters of the calculation process, and cloud computing timestamps. These summaries are then concatenated with the rule-binding hash to generate the final evidence hash. All final evidence hashes within the same preset time shard granularity are chained together in chronological order to form a time-shard hash chain. The beginning and end of each shard are linked to ensure temporal integrity and are persistently stored. Figure 6 As shown, when the rule version is upgraded to V1.1 in T2, the historical data D1 of the T1 period remains permanently bound to the V1.0 version rule. This evidence storage structure can subsequently upload hash values ​​to the blockchain network, meaning the evidence storage module can combine blockchain technology to put the hash values ​​of the calculation results on the chain, further enhancing the immutability and credibility of the data. The unified model and tagging system of this invention can seamlessly expand to new energy categories and energy consumption scenarios without restructuring the underlying architecture, adapting to the development trend of integrated energy systems and can be widely applied to carbon metering in multiple fields such as industry, construction, and transportation. The system supports backtracking queries by time, region, energy type, equipment ID, and other dimensions, and can reproduce the complete calculation process for any historical period.

[0040] This invention also proposes a unified carbon metering method based on cloud-edge collaboration for multiple energy forms, applied to the aforementioned system, see [link to relevant documentation]. Figure 4 It includes the following steps: S1. Real-time collection of multi-energy quantity data by the terminal: The edge computing terminal deployed at the energy consumption site collects the energy consumption parameters of multiple energy media in real time; S2. Automatically attaching standardized feature tags to the terminal: Automatically generating multi-dimensional feature tags locally on the edge computing terminal and binding the multi-dimensional feature tags with the corresponding energy consumption parameters to form characteristic metering data. Among them, the multi-dimensional feature tags include at least the application scenario semantic features and energy flow features used to define the accounting boundary. S3. Data is written to the local cache storage of the terminal: The edge computing terminal caches the characteristic measurement data and performs independent sharding and encryption processing on the cached data according to the preset time sharding granularity to obtain the encrypted sharded data. The encryption key for each time shard is dynamically derived based on the basic key of the edge computing terminal and the time base parameter of the corresponding time shard. S4. Determine if the communication network is available; S5. Terminal batch asynchronous upload of historical data: When the preset communication conditions are met, the edge computing terminal will upload the encrypted fragmented data to the cloud computing platform in batches according to the time sequence. S6. Cloud-based data reception and integrity verification: The cloud-based accounting platform receives encrypted fragmented data and verifies the continuity of the received data based on the fragmented time sequence. S7. Matching corresponding spatiotemporal emission factors and energy models: The cloud computing platform decrypts the verified fragmented data to obtain characteristic measurement data, extracts the application scenario semantic features and energy flow features from the characteristic measurement data, and uses the extracted features as input parameters to automatically match the corresponding energy carbon emission model. S8. Calculate carbon emissions by calling the unique accounting formula: Call the matched energy carbon emission model and the corresponding unique accounting formula to calculate carbon emissions; S9. Perform result association and archive evidence.

[0041] To ensure auditability throughout the entire lifecycle, one embodiment of the present invention provides a unified carbon metering method for multiple energy forms based on cloud-edge collaboration, which further includes a historical data backtracking and verification step: After completing the calculation, extract the feature summary of a single characteristic metering data, the called energy carbon emission model, and the calculation result, and perform layer-by-layer concatenation calculation to generate a final evidence hash. Then, concatenate the final evidence hashes within the same preset time granularity in chronological order to form a time-slice hash chain. Generate a unique traceability code for each completed calculation result, and bind and store the characteristic metering data, the energy carbon emission model called at that time, and the hash evidence record of the calculation result with the traceability code. When a reproduction request carrying the traceability code is received, retrieve the historical energy carbon emission model and historical emission factor corresponding to the historical calculation based on the bound evidence record to reproduce the calculation process. Compare whether the hash value generated by the reproduction is consistent with the final evidence hash stored in the corresponding time-slice hash chain to verify that the historical carbon emission calculation results and calculation rating standards have not been tampered with.

[0042] Taking a comprehensive industrial park as an example, the end-to-end process is explained as follows: A comprehensive industrial park includes industrial production plants, office buildings, and a centralized heating station, involving three types of energy: electricity, natural gas, and steam. Network signals are unstable in some areas. Within the park, edge computing terminals for electricity, gas, and heat are deployed at the 10kV incoming power distribution room, the natural gas main pipeline interface, and the steam outlet of the centralized heating station. Each terminal collects the instantaneous and cumulative quantities of its corresponding energy source in real time, generating a metering record every 15 seconds and automatically attaching complete feature tags: Specifically, the electricity terminal (set as energy type 01) generates data with an "input" identifier, a millisecond-level timestamp, terminal ID, location code (area A of the park), and "production / office" scenario semantics; the gas terminal generates data with feature tags including "energy type 02, flow direction input, millisecond-level timestamp, terminal ID, location code, and boiler heating scenario"; and the steam terminal (energy type 03) collects data with feature tags including "output flow direction, millisecond-level timestamp, terminal ID, location code, and heating output scenario". The terminal locally performs AES encryption on the above data in 1-hour granularities (each fragment key is derived from the terminal's root key combined with the start timestamp of the time slice, and a fragment message authentication code is generated), and caches it in non-volatile memory. It is configured to automatically send all data from the previous day in batches at 2:00 AM daily when the network is idle; if the network is interrupted, the data continues to be cached locally, and the missing data is automatically retransmitted in chronological order after the network is restored, without manual intervention.

[0043] After receiving the data, the cloud platform's timing verification module checks the integrity and legality of the data packets. Abnormal data triggers retransmission, while qualified encrypted fragmented data is temporarily stored in a cache queue. The fragmented data is then decrypted and restored to its original feature data before being written to the original database. Subsequently, the accounting engine extracts the data's time, location, energy type, and scenario tags, automatically matching the corresponding spatiotemporal emission factors and energy models: for electricity data, it matches the time-of-use grid emission factor for the corresponding time period in the province and calls the electricity source model including power generation structure and transmission and distribution losses; for natural gas data, it matches the emission factor per unit calorific value of natural gas in the region and calls the natural gas source model including extraction and pipeline emissions; for steam data, it matches the boiler combustion emission factor, calls the heat use model, and allocates the emissions to each energy-consuming unit according to the heating area. The engine uses standardized accounting formulas to complete the calculations, ensuring consistent results for multiple calculations of the same input data.

[0044] After the system completes the calculation, it generates a chained evidence hash according to a three-layer structure of "original feature data - accounting rule version - accounting result," generating a carbon emission record with a unique traceability code. This record is then associated with the original data ID, factor version V1.2, model version V2.0, and the three-layer evidence hash value and stored in the evidence repository. The hashes of data from the same day are concatenated by time slices to form a hash chain for that day, ensuring the integrity of the time sequence. Users can query the complete calculation chain through the traceability code, reproduce the accounting process, and verify that the data and rules have not been tampered with through hash verification. Even if the emission factor is subsequently updated to version V1.3, the historical data is still bound to the rules of version V1.2 for traceability, and the rating result remains unchanged. Compared with the traditional static factor accounting scheme, this process achieves the unification of multi-energy accounting standards, eliminating the result deviation caused by different value standards; there is no data loss or time sequence disorder during network interruptions; historical carbon emission data can be completely traced back through the traceability code, meeting the standardization requirements of local carbon verification and corporate carbon asset management.

[0045] Meanwhile, this system can be expanded to adapt to carbon emission reduction metering for distributed green electricity such as photovoltaic and wind power. Specifically, green electricity source feature markers can be added to the terminal, and the corresponding green electricity emission reduction accounting model can be updated in the cloud.

[0046] Although the methods described above are illustrated and depicted as a series of actions for the sake of simplicity, it should be understood and appreciated that these methods are not limited by the order of the actions, as some actions may occur in a different order and / or concurrently with other actions from the illustrations and descriptions herein or not illustrated and described herein but which may be understood by those skilled in the art, according to one or more embodiments.

[0047] The constructions and arrangements of this application shown in several different exemplary embodiments are merely illustrative. Although only a few embodiments are described in detail in this disclosure, those who consult this disclosure will readily understand that many modifications are possible (e.g., changes in size, structure, shape, and proportions, as well as parameter values, installation arrangements, use of materials, color, orientation, etc.) without substantially departing from the novel teachings and advantages of the subject matter described in this application. For example, an element shown as integrally formed may be composed of multiple parts or elements, the position of the elements may be inverted or otherwise changed, and the nature or number or position of discrete elements may be altered or changed. Therefore, all such modifications are intended to be included within the scope of this invention. The order or sequence of any process or method steps may be changed or rearranged according to alternative embodiments. In the claims, any "device plus function" clause is intended to cover the structure described herein that performs the function, and not only structurally equivalent but also equivalent in structure. Various corresponding modifications and variations can be made by those skilled in the art according to this invention without departing from the spirit and essence of the invention, but such corresponding modifications and variations should fall within the protection scope of this invention.

Claims

1. A unified carbon metering system for multiple energy forms based on cloud-edge collaboration, characterized in that, This includes edge computing terminals deployed at energy consumption sites and cloud-based accounting platforms that establish asynchronous communication connections with the edge computing terminals; The edge computing terminal includes a multi-energy metering and acquisition module, a semantic feature tagging module, a fragmented encryption cache module, and an asynchronous upload module; The multi-energy metering and acquisition module is used to collect energy consumption parameters of multiple energy media in real time; The semantic feature tagging module is used to automatically generate multi-dimensional feature tags locally on the edge computing terminal and bind the multi-dimensional feature tags with the corresponding energy consumption parameters to form characteristic metering data. The multi-dimensional feature tags include at least application scenario semantic features and energy flow features used to define the accounting boundary. The sharded encryption cache module is used to cache characteristic measurement data and perform independent sharded encryption processing on the cached data according to the preset time sharding granularity to obtain encrypted sharded data. The encryption key for each time shard is dynamically derived based on the basic key of the edge computing terminal and the time base parameter of the corresponding time shard. The asynchronous upload module is used to upload encrypted fragmented data to the cloud computing platform in batches according to time sequence when preset communication conditions are met; The cloud-based accounting platform includes a time-series verification module and a decoupled accounting module; The timing verification module is used to receive encrypted fragmented data and verify the continuity of the received data based on the fragment timing. The decoupled accounting module is used to decrypt the verified fragmented data to obtain characteristic metering data, extract the application scenario semantic features and energy flow features from the characteristic metering data, and use the extracted features as input parameters to automatically match the corresponding energy carbon emission model for accounting.

2. The unified carbon metering system based on cloud-edge collaboration for multiple energy forms according to claim 1, characterized in that, The semantic feature labeling module, when binding multidimensional feature labels with corresponding energy consumption parameters, is specifically used for: During the data interruption processing stage of the underlying processing unit configured in the edge computing terminal, the multidimensional feature markers are converted into a binary mask with a preset bit width. The binary mask is directly concatenated with the corresponding energy consumption parameter values ​​and encapsulated into an indivisible fixed-length data frame to form characteristic measurement data.

3. The unified carbon metering system based on cloud-edge collaboration for multiple energy forms according to claim 1, characterized in that, When the semantic feature labeling module automatically generates application scenario semantic features and energy flow direction features locally on the edge computing terminal, it specifically includes: When generating semantic features for application scenarios, at least one of the following methods is used: By reading the operating status register data of the underlying programmable logic controller at the energy consumption site through the industrial standard communication protocol, the equipment start-up and shutdown status and process stage identifiers are extracted and converted into semantic features of the application scenario. Query the local mapping configuration table pre-set on the edge computing terminal, hard-bind the physical port identifier of the current acquisition channel with the corresponding application scenario label, and generate application scenario semantic features; When generating energy flow characteristics, at least one of the following methods can be used: By analyzing the positive and negative metering register data of the bidirectional metering instrument or the positive and negative polarity signals of the underlying acquisition sensor, the physical direction of energy flow is dynamically determined to generate energy flow characteristics. Based on the predefined energy flow topology attributes of the acquisition channel in the preset local mapping configuration table, the energy flow direction characteristics representing the energy inflow or outflow of the current acquisition node are directly mapped and generated.

4. The unified carbon metering system based on cloud-edge collaboration for multiple energy forms according to claim 1, characterized in that, The time base parameters include the start timestamp of the corresponding time slice; When the segmented encryption cache module dynamically derives and generates encryption keys for each time segment, it specifically uses: a key derivation function based on hash-based message authentication code to take the basic key stored in the edge computing terminal security chip as input key material, and to perform irreversible hashing operation using the start timestamp and the unique device identifier of the edge computing terminal as salt parameters to generate a symmetric encryption key exclusive to the current time segment. When the decoupled accounting module decrypts the verified fragmented data, it specifically derives a decryption key based on the unique device identifier of the edge computing terminal and the start timestamp of the corresponding fragment, and uses the decryption key to decrypt the verified fragmented data.

5. The unified carbon metering system based on cloud-edge collaboration for multiple energy forms according to claim 1, characterized in that, The timing verification module verifies the continuity of received data based on fragment timing, and when a fragment is detected to be missing, it performs the following retransmission interaction mechanism with the asynchronous upload module: After the communication connection is restored, the asynchronous upload module prioritizes sending an index data packet containing all cached time segment identifiers to the cloud computing platform. The timing verification module compares the historical received records in the cloud with the index data packet, generates a missing list containing missing time fragment identifiers through difference operation, and sends the missing list to the asynchronous upload module; The asynchronous upload module re-uploads the corresponding encrypted fragment data according to the received missing list, and clears the corresponding locally uploaded cached data after receiving verification confirmation from the cloud computing platform.

6. The unified carbon metering system based on cloud-edge collaboration for multiple energy forms according to claim 1, characterized in that, The sharded encryption cache module is also used to dynamically adjust the preset time sharding granularity according to the real-time network quality of the edge computing terminal, specifically for: The real-time network quality parameters between the edge computing terminal and the cloud computing platform are obtained, including network packet loss rate, uplink bandwidth, and round-trip latency. The current network quality level is evaluated based on the real-time network quality parameters, and the target time fragmentation granularity is dynamically determined according to the preset quality level and fragmentation granularity mapping rules. The lower the network quality level, the larger the corresponding target time fragmentation granularity, and the target time fragmentation granularity is not greater than the preset maximum fragmentation granularity upper limit. After the data processing cycle of the current time segment ends, the target time segment granularity is applied, and the granularity identifier representing the target time segment granularity is written into the data frame header of the next segment of data to be synchronized to the cloud computing platform.

7. The unified carbon metering system based on cloud-edge collaboration for multiple energy forms according to claim 1, characterized in that, The multidimensional feature marker also includes spatial location encoding, acquisition timestamp, and energy type encoding; The decoupled accounting module internally constructs a spatiotemporal carbon emission factor library, which constructs a three-dimensional primary key constraint mechanism based on spatial, temporal, and energy type dimensions. When automatically matching the corresponding energy carbon emission model for calculation, the decoupled accounting module uses the extracted spatial location code, collection timestamp, and energy type code as the joint query primary key to match and extract emission factor versions and accounting benchmark formulas with unique correspondences from the spatiotemporal carbon emission factor library.

8. The unified carbon metering system based on cloud-edge collaboration for multiple energy forms according to claim 7, characterized in that, The decoupled accounting module also contains a source accounting model library and a usage accounting model library covering a variety of energy media. When the decoupled accounting module automatically matches the corresponding energy carbon emission model for accounting, it is specifically used for: If the extracted energy flow characteristics are the input identifier, the source accounting model corresponding to the energy type code will be automatically called to calculate the transmission carbon emissions covering the energy extraction, conversion to terminal access process; If the extracted energy flow characteristics are used as output identifiers, the extracted application scenario semantic characteristics will be combined to automatically call the usage accounting model corresponding to the application scenario and calculate the terminal carbon emission allocation under the specific energy consumption scenario.

9. The unified carbon metering system for multiple energy forms based on cloud-edge collaboration according to claim 1, characterized in that, The cloud-based accounting platform also includes an evidence storage and traceability module, used to perform three-layer structured chain hash evidence storage on the accounting results obtained by the decoupled accounting module, specifically including: The first layer extracts feature summaries from individual characteristic measurement data and generates raw data hashes. The second layer extracts feature summaries from the emission factor version and corresponding energy carbon emission model called in the current accounting, and concatenates them with the original data hash to generate a rule binding hash. The third layer extracts feature summaries from the accounting results, intermediate parameters of the calculation process, and cloud computing timestamps, and binds them with the rules to perform hash concatenation calculation to generate the final evidence storage hash. All final evidence hashes within the same preset time granularity are chained together in chronological order to form a time-slice hash chain and are persistently stored.

10. A unified carbon measurement method for multiple energy forms based on cloud-edge collaboration, characterized in that, The application to the unified carbon metering system for multiple energy forms based on cloud-edge collaboration as described in any one of claims 1-9 includes the following steps: Edge computing terminals deployed at energy consumption sites collect energy consumption parameters of multiple energy media in real time; Multidimensional feature tags are automatically generated locally on the edge computing terminal, and the multidimensional feature tags are bound to the corresponding energy consumption parameters to form characteristic metering data. The multidimensional feature tags include at least application scenario semantic features and energy flow features used to define the accounting boundary. The edge computing terminal caches characteristic metering data and performs independent fragment encryption processing on the cached data according to the preset time fragment granularity to obtain encrypted fragment data. The encryption key for each time fragment is dynamically derived based on the edge computing terminal's basic key and the time base parameter of the corresponding time fragment. When the preset communication conditions are met, the edge computing terminal will upload the encrypted fragmented data to the cloud computing platform in batches according to the time sequence. The cloud-based accounting platform receives encrypted fragmented data and verifies the continuity of the received data based on the fragmentation time sequence. The cloud-based accounting platform decrypts the verified data shards to obtain characteristic metering data, extracts application scenario semantic features and energy flow features from the characteristic metering data, and uses the extracted features as input parameters to automatically match the corresponding energy carbon emission model for accounting.

11. The unified carbon measurement method for multiple energy forms based on cloud-edge collaboration according to claim 10, characterized in that, It also includes historical data backtracking and verification steps: After the accounting is completed, the feature summary of the single characteristic measurement data, the energy carbon emission model called, and the accounting result are extracted and spliced ​​layer by layer to generate the final evidence hash. The final evidence hashes within the same preset time segment granularity are then chained together in chronological order to form a time slice hash chain. A unique traceability code is generated for each completed accounting result, and the characteristic measurement data, the energy carbon emission model invoked at that time, and the hash evidence record of the accounting result are bound and stored with the traceability code; Upon receiving a reproduction request carrying the traceability code, the historical energy carbon emission model and historical emission factors corresponding to the historical accounting are retrieved based on the bound evidence records to reproduce the accounting process; The hash value generated by the reproduction is compared with the final evidence hash stored in the corresponding time slice hash chain to verify that the historical carbon emission accounting results and accounting rating standards have not been tampered with.

Images