Cross-network segment device carbon flow metering and regulation method based on industrial control communication data analysis

Abstract

This application discloses a cross-network segment device carbon flow metering and control method based on industrial control communication data parsing. It proposes a universal parsing rule for cross-network segment multi-protocol data, a keyword hash field mapping template, and a quantitative data cleaning standard, along with a supporting solidified algorithm. This effectively solves the problems of inconsistent aggregation and poor adaptability of heterogeneous cross-network segment data. A coupled calculation model of industrial control communication, energy consumption, and equipment carbon emissions is constructed, combined with a four-level dynamic carbon factor real-time correction method, effectively improving the accuracy of carbon emission accounting and reducing calculation errors. A carbon flow time-series calibration, equipment carbon emission numerical anomaly classification judgment, and quantitative low-carbon collaborative control system are established, combining carbon metering with low-carbon management and control, solving the problems of traditional carbon metering and management being disconnected and unable to proactively reduce emissions. Through full-process algorithm solidification, combined with data processing, a complete closed loop of "metering and accounting - anomaly judgment - low-carbon control - data archiving" is formed, solving the pain points of poor implementation and weak universality of cross-network segment carbon flow metering.

Description

Technical Field

[0001] This invention belongs to the field of industrial low-carbon management and carbon metering technology, specifically involving a method for cross-network segment device carbon flow metering and control based on industrial control communication data parsing.

[0002] Terminology Definition

[0003] This invention relates to two types of coupled data, defined as follows:

[0004] 1. Carbon Flow Timing Data: Communication messages, heartbeat timing data, and energy flow link data of industrial control equipment across network segments are the underlying basic data for carbon metering and are used for data aggregation, clock synchronization, and transmission anomaly identification.

[0005] 2. Equipment carbon emission values: Based on the carbon flow time-series data after cleaning and calibration, the total carbon emissions per equipment cycle are calculated through the coupled calculation model of industrial control communication-energy consumption-equipment carbon emission. This serves as the core quantitative indicator for equipment emission classification, low-carbon regulation, and carbon ledger archiving. Background Technology

[0006] Against the backdrop of the comprehensive advancement of the "dual carbon" goals, the industrial sector, as a core source of carbon emissions, faces the critical challenge of achieving precise carbon metering and low-carbon management at the equipment and supply chain levels to fulfill emission reduction targets. Currently, industrial sites commonly employ cross-network segment layouts, with the field control layer and upper-level monitoring layer network segments being separated—a conventional deployment method. Carbon flow management for such cross-network segment industrial control equipment suffers from numerous shortcomings that existing technologies cannot address, specifically as follows:

[0007] Firstly, traditional carbon metering relies heavily on dedicated energy consumption sensors and gas collection equipment, requiring additional hardware installation, on-site wiring modifications, and the development of supporting data collection and carbon accounting systems. Hardware procurement and system development are costly and time-consuming, making it extremely unsuitable for existing old industrial sites and scenarios without dedicated renovation budgets, and extremely difficult to implement.

[0008] Secondly, cross-network segment industrial control scenarios often adopt a multi-protocol coexistence mode such as TCP / IP, Modbus-RTU, Modbus-TCP, Profinet, and S7 series. The data formats, field definitions, and storage volumes of different manufacturers and models of equipment vary significantly. Furthermore, cross-network segment data is physically isolated, and there is no universal cross-protocol parsing and data aggregation method in the industry. Multi-source heterogeneous data cannot be normalized, resulting in carbon metering being able to achieve only a general accounting of the entire plant area, and unable to break down the carbon flow of individual devices or single links.

[0009] Third, there is no universal data cleaning standard in the industry that is suitable for carbon metering. There is no unified basis for filtering invalid and abnormal data, resulting in poor data processing reliability. Furthermore, there is no lightweight solution for storing massive amounts of historical data, which can easily lead to data redundancy and low processing efficiency in the long run.

[0010] Fourth, existing carbon metering mostly uses fixed carbon factors and does not combine equipment operating conditions, energy structure, and regional green electricity ratio for real-time correction, resulting in low metering accuracy. Furthermore, carbon metering is disconnected from low-carbon management and control, making it impossible to achieve proactive emission reduction. Network fluctuations across different grid segments can also easily lead to data interruptions, further affecting the accuracy of accounting.

[0011] In summary, the general carbon accounting factor system in the industrial sector has become relatively complete, and the intrinsic correlation between energy consumption and carbon emissions of industrial control equipment has been clarified. However, there is a lack of lightweight algorithm coupling solutions that can adapt to cross-network segment, multi-protocol, and heterogeneous data scenarios. Furthermore, existing technologies have not solved the practical problems of cross-network segment data aggregation, general cleaning, and dynamic factor correction. As a result, carbon metering of cross-network segment industrial control equipment has always suffered from insufficient accuracy, poor practicality, and weak versatility, which are industry pain points. Summary of the Invention

[0012] The purpose of this application is to address the technical shortcomings of existing cross-network segment industrial equipment carbon metering, such as high implementation costs, difficulty in transformation, difficulty in interoperability of multi-source heterogeneous data, insufficient metering accuracy, poor dynamic adaptability of carbon factors, and disconnect between carbon metering and low-carbon management. This application provides a cross-network segment equipment carbon flow metering and control method based on industrial control communication data parsing to solve the practical problems of cross-network segment data aggregation, general cleaning, and dynamic factor correction.

[0013] To address the aforementioned technical problems, this application provides a method for carbon flow metering and control of cross-network segment devices based on industrial control communication data parsing, comprising:

[0014] Construct a cross-network segment industrial control data general parsing and aggregation system to achieve seamless aggregation and normalization of heterogeneous data from multiple protocols in order to provide high-quality data support for carbon flow metering;

[0015] A coupled calculation model of industrial control communication, energy consumption, and equipment carbon emissions is constructed, and a four-level dynamic carbon factor real-time correction method is used to achieve accurate measurement of carbon flow time-series data of cross-network segment equipment.

[0016] Establish a classification and judgment mechanism for abnormal equipment carbon emission values ​​based on carbon flow time series data, realize the automatic identification and accurate marking of abnormal carbon flow time series data, and provide a basis for low-carbon regulation;

[0017] Based on the aforementioned anomaly classification and judgment mechanism, corresponding low-carbon collaborative control instructions are output, and a closed-loop management and control mechanism of "metering and accounting - anomaly judgment - low-carbon control - data archiving" is constructed to realize low-carbon optimized operation of cross-network segment equipment.

[0018] Carbon flow time-series data archiving and measurement result review are used to build a long-term closed-loop management and control mechanism to meet compliance requirements.

[0019] As a preferred embodiment, a method for cross-network segment device carbon flow metering and control based on industrial control communication data parsing, wherein the construction of a cross-network segment industrial control data universal parsing and aggregation system to achieve seamless aggregation and normalization processing of multi-protocol heterogeneous data includes:

[0020] In cross-network segment industrial control scenarios, industrial-grade network resolution terminals and data aggregation nodes are deployed, adopting a hybrid architecture of "distributed resolution + centralized aggregation";

[0021] Construct a multi-protocol fusion parsing and field mapping architecture;

[0022] Establish a hierarchical data preprocessing mechanism;

[0023] Lightweight hierarchical storage and real-time processing mechanisms are adopted to enable data storage and computation.

[0024] The solution requires detailed explanation of a method for cross-network segment device carbon flow metering and control based on industrial control communication data parsing. The establishment of a hierarchical data preprocessing mechanism includes:

[0025] A dual-dimensional threshold start / stop determination algorithm is adopted to remove invalid operating condition data. The algorithm combines communication activity and load value for joint determination to avoid misjudgment based on a single parameter.

[0026] For continuous numerical data such as load rate, power, and communication duration, outliers are removed using the complete algorithm based on the Laida 3σ criterion.

[0027] A standardized process combining linear normalization and unified unit conversion is used to normalize and calibrate the data;

[0028] Data is deduplicated using a hash deduplication algorithm.

[0029] As a preferred embodiment, a method for cross-network segment equipment carbon flow metering and control based on industrial control communication data parsing, wherein the construction of an industrial control communication-energy consumption-equipment carbon emission coupled calculation model, combined with a four-level dynamic carbon factor real-time correction method, to achieve accurate metering of cross-network segment equipment carbon flow time-series data includes:

[0030] The computational model employs a two-level lightweight computational logic.

[0031] The dynamic carbon factor is adjusted in real time through a four-level correction process and algorithm solidification linkage. The correction process is embedded with the energy consumption-carbon emission calculation algorithm throughout the process and is synchronized with the data processing cycle.

[0032] A dual verification mechanism is employed to ensure the accuracy of the carbon flow time-series data measurement.

[0033] The solution requires detailed explanation of a method for carbon flow metering and control of cross-network segment devices based on industrial control communication data parsing. The calculation model employs a two-stage lightweight calculation logic, including:

[0034] Level 1: Energy consumption conversion logic. For different types of industrial control equipment across network segments, based on the operating load rate, communication duration, and rated power parsed from communication data, a formula for calculating energy consumption per unit operating time is established:

[0035]

[0036] in, Energy consumption per equipment cycle (kWh); Rated power of the equipment (kW); Real-time load rate of the equipment (%) The effective operating time of the device (h); This is the working condition correction factor, with a value ranging from 0.8 to 1.0;

[0037] Level 2: The logic for measuring the current carbon emissions of the equipment introduces a dynamic carbon factor and combines it with real-time correction of multi-dimensional parameters to establish a formula for calculating the current carbon emissions of the equipment.

[0038]

[0039] in, The carbon emission value of the equipment in a single cycle (for the current period) ); For dynamic carbon factor ( / kWh).

[0040] As a preferred embodiment, a method for cross-network segment device carbon flow metering and control based on industrial control communication data parsing, wherein establishing a cross-network segment carbon flow time-series data anomaly classification and judgment mechanism to achieve automatic identification and accurate marking of carbon flow time-series data anomalies includes:

[0041] Based on the low-carbon emission reduction standards of the industrial sector, the corresponding rated carbon emission benchmark values ​​of equipment and the operating condition loss coefficient, establish a standard for determining the first level of normal operation, the second level of early warning, and the third level of exceeding the standard for cross-network segment equipment.

[0042] Establish a first-level normal judgment standard: δ≤90%, the algorithm marks it as a normal operating state, no control operation, continuously monitor the equipment operation data and equipment carbon emission values ​​in real time, record the equipment carbon emission values ​​once an hour, and store them in the data ledger;

[0043] Establish a two-level early warning judgment standard: 90% < δ ≤ 110%, the algorithm marks it as a slightly high emission state, automatically generates a yellow early warning label, synchronously triggers the mild control module, records the early warning time, equipment information, carbon emission exceedance range, and pushes early warning prompts to operation and maintenance personnel;

[0044] A three-level standard for judging excessive emissions is established: when δ>110%, the algorithm marks it as a state of severe high emissions, automatically generates a red excessive emission label, simultaneously triggers the severe control module, immediately pushes an emergency warning, and records the excessive data and equipment operating conditions in detail for easy subsequent traceability and analysis.

[0045] in, , To automatically calculate the ratio, This is the baseline value for the nameplate of a single device. This represents the current carbon emission value of the equipment.

[0046] The solution requires detailed explanation of a method for cross-network segment device carbon flow metering and control based on industrial control communication data parsing. Following the establishment of a cross-network segment carbon flow time-series data anomaly classification and judgment mechanism, the method further includes:

[0047] To address data loss caused by cross-network segment fluctuations and communication interruptions, an anomaly detection and calibration mechanism should be established.

[0048] The solution requires further detailed explanation of a cross-network segment device carbon flow metering and control method based on industrial control communication data parsing. The step of outputting corresponding low-carbon collaborative control instructions according to the anomaly classification and judgment mechanism includes:

[0049] For normally operating equipment: The dynamic carbon factor is automatically calibrated every cycle, a daily report of equipment carbon emission values ​​is generated, data fluctuations are continuously monitored, and an early warning is automatically triggered if the fluctuation exceeds ±5%, so as to avoid high emission risks in advance and maintain low-carbon and stable operation.

[0050] For secondary, low-emission equipment: optimize heartbeat interval and load rate, and reduce invalid communication;

[0051] For Level 3 heavy-emission equipment: optimize communication paths and regulate peak-shifting timing and load reduction under different operating conditions.

[0052] The solution requires detailed explanation of a cross-network segment device carbon flow metering and control method based on industrial control communication data parsing. After outputting the corresponding low-carbon collaborative control command according to the anomaly classification and judgment mechanism, the method further includes:

[0053] The control effects on normally operating equipment, secondary light-emission equipment, and tertiary heavy-emission equipment are verified, and the emission reduction effect is calculated and a control effect report is generated.

[0054] Determine whether the emission reduction effect has achieved the expected results;

[0055] If yes, then the control effect is determined to meet the requirements; if no, then the control parameters are automatically optimized, and the process returns to the step of outputting the corresponding low-carbon coordinated control command according to the anomaly classification judgment mechanism, until the expected effect is achieved.

[0056] The solution requires detailed explanation of a method for carbon flow metering and control of cross-network segment devices based on industrial control communication data parsing. The carbon flow time-series data archiving and metering result review include:

[0057] The raw data, anomaly identification and judgment records, low-carbon control schemes, and on-site implementation effects of carbon flow time-series data measurement for all equipment across network segments are uniformly archived according to the aforementioned lightweight hierarchical storage method, forming a standardized carbon measurement ledger that is traceable, verifiable, and exportable. The standardized carbon measurement ledger includes equipment information, network segment information, equipment carbon emission values, anomaly records, control records, and effect verification data, meeting the compliance requirements for industrial carbon verification, emission reduction assessment, and environmental data traceability.

[0058] Monthly carbon flow time-series data archiving and measurement result review are completed regularly, and quarterly review reports are generated to summarize the carbon emission statistics and control effectiveness of equipment across the entire network segment, providing data support for subsequent low-carbon management decisions.

[0059] The present invention provides a method for carbon flow metering and control of cross-network segment devices based on industrial control communication data parsing, which has at least the following beneficial effects:

[0060] 1. It proposes a universal parsing rule for cross-network segment multi-protocol, a keyword hash field mapping template, and a quantitative data cleaning standard, along with a fixed algorithm. This effectively solves the problems of inconsistent aggregation and poor adaptability of heterogeneous data across network segments, and realizes seamless aggregation and normalization of industrial control data from multiple vendors and protocols. It can be adapted to various existing industrial control scenarios.

[0061] 2. A coupled calculation model of industrial control communication, energy consumption, and equipment carbon emissions was constructed. Combined with a four-level dynamic carbon factor real-time correction method, the pain points of traditional carbon metering, such as reliance on dedicated hardware, fixed carbon factors, and low metering accuracy, were solved. This enabled accurate carbon metering in scenarios without dedicated sensors, effectively improving the accuracy of carbon emission accounting and significantly reducing accounting errors.

[0062] 3. A lightweight hierarchical storage algorithm and a timed automatic cleanup mechanism are designed to solve the problems of redundancy and low processing efficiency of massive cross-network segment carbon stream time-series data storage. It can operate stably with existing industrial control equipment without the need for additional dedicated hardware, effectively reducing implementation costs.

[0063] 4. Establish a graded judgment mechanism for abnormal equipment carbon emission values ​​based on carbon flow time series data and a quantitative low-carbon collaborative control system. This deeply integrates carbon measurement with low-carbon management and control, solving the problem of traditional carbon measurement and control being disconnected and unable to actively reduce emissions. It can accurately locate high-emission equipment and achieve graded control, effectively reducing carbon emissions from industrial cross-network equipment.

[0064] 5. By solidifying the algorithm throughout the entire process and combining it with data processing to form a complete closed loop of "data processing - metering and accounting - anomaly judgment - low-carbon control - data archiving", the system solves the industry pain points of poor implementation and weak universality of cross-network segment carbon flow metering, improves the efficiency and accuracy of industrial low-carbon management and control, and can meet the compliance requirements of carbon verification and emission reduction assessment. Attached Figure Description

[0065] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0066] Figure 1 A schematic diagram of a cross-network segment device carbon flow metering and control method based on industrial control communication data parsing provided in an embodiment of this application;

[0067] Figure 2 This is a flowchart of a dynamic carbon factor correction method provided in an embodiment of this application;

[0068] Figure 3 This is a schematic diagram of anomaly classification and low-carbon synergistic regulation provided in an embodiment of this application. Detailed Implementation

[0069] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0070] The core of this application is to provide a method for carbon flow measurement and control of cross-network segment devices based on industrial control communication data parsing. It is applicable to various industrial automation production lines, industrial park industrial control networks, and general industrial control equipment clusters with cross-network segment communication. It is especially suitable for carbon accounting and low-carbon optimization scenarios in existing industrial sites. It solves the technical pain points of traditional carbon measurement, such as reliance on dedicated sensors, inability to uniformly aggregate heterogeneous data across network segments, inability to accurately trace carbon flow, and poor dynamic adaptability. It also solves the practical problems of cross-network segment data aggregation, general cleaning, and dynamic factor correction.

[0071] Figure 1 A schematic diagram of a cross-network segment device carbon flow metering and control method based on industrial control communication data parsing provided in an embodiment of this application; Figure 2This is a flowchart of a dynamic carbon factor correction method provided in an embodiment of this application; Figure 3 This is a schematic diagram of anomaly classification and low-carbon synergistic regulation provided in an embodiment of this application, as shown below. Figures 1 to 3 As shown.

[0072] A method for carbon flow metering and control of cross-network segment devices based on industrial control communication data parsing includes the following steps:

[0073] Step S101: Construct a cross-network segment industrial control data general parsing and aggregation system to achieve seamless aggregation and normalization of heterogeneous data from multiple protocols in order to provide high-quality data support for carbon flow metering.

[0074] Preferably, this step includes the following process:

[0075] First, in cross-network segment industrial control scenarios, industrial-grade network resolution terminals and data aggregation nodes are deployed, adopting a hybrid architecture of "distributed resolution + centralized aggregation".

[0076] The industrial-grade network parsing terminal model adopts IEC-61850-3. This terminal supports the parsing of five mainstream industrial control protocols: TCP / IP, Modbus-RTU, Modbus-TCP, Profinet, and S7-200 / 300. The sampling rate can reach 100Hz, the parsing latency is ≤10ms, and the accuracy can reach 0.1%. It also has a built-in ARM Cortex-A9 processor (800MHz main frequency), which can complete the protocol frame identification and core field extraction every 5ms. When an abnormal packet is detected, an abnormal identifier (Code 0x6B) is generated within 20ms.

[0077] At the core routing nodes of each network segment, a data aggregation node is deployed. The model is GW-8000. This node supports simultaneous access of dual network segments and has two built-in gigabit Ethernet interfaces, corresponding to the upper-layer monitoring network segment and the field device control network segment respectively. It has read-only access permissions, does not modify the routing configuration, and does not penetrate the firewall.

[0078] At each industrial control device, a small data acquisition module, model DAQ-200, is deployed. This module supports passive reading of device communication messages without modifying the original device configuration. The sampling period can be adaptively adjusted (5ms-100ms). It has a built-in cache module (capacity 16MB) that can cache data when the network is interrupted (up to 2 hours), and automatically retransmit it after the network is restored.

[0079] Through the aforementioned terminals, nodes, and modules, communication data from cross-network segment industrial control equipment is collected in real time, providing accurate and continuous data support for subsequent carbon flow measurement and low-carbon regulation.

[0080] Second, a multi-protocol fusion parsing and field mapping architecture is constructed. The parsing terminal and the aggregation node communicate via an Ethernet interface, using the Modbus-TCP protocol (transmission rate 100Mbps, common-mode interference resistance ≥2500V), supporting Time-Sensitive Networking (TSN) extensions, ensuring end-to-end latency of less than 5ms when the parsing terminal sends an anomaly flag; the data acquisition module communicates with the parsing terminal via an RS-485 interface, with a transmission rate of 9600bps, using CRC-16 checksum, and an automatic retransmission interval of less than 50ms for error frames; the aggregation node and the algorithm processing center transmit normalized data via WiFi 6 (5GHz band), enabling OFDMA technology, achieving a single-node throughput of 1.2Gbps and latency jitter of less than 100μs.

[0081] In addition, based on the IEEE 802.1AS Generalized Precise Time Protocol (gPTP), the clock synchronization of the entire network, including the parsing terminal, aggregation node, acquisition module and algorithm processing center, is achieved with a synchronization error of <1ms, ensuring the time consistency of data across network segments.

[0082] To address the challenges of multi-segment network isolation and the coexistence of multiple protocols in industrial environments, a unified cross-protocol parsing logic has been developed. This logic covers five mainstream industrial protocols: TCP / IP, Modbus-RTU, Modbus-TCP, Profinet, and S7-200 / 300. The entire parsing process employs a read-only mode, avoiding any device-specific proprietary protocols or encrypted messages, thus ensuring non-intrusiveness and minimal disruption to production. Specific parsing methods are as follows:

[0083] Frame header and trailer identification: Preset hexadecimal identifiers for various protocol standard frame headers and trailers. Modbus protocol uses a fixed frame header of 0x3A, TCP / IP protocol uses a standard IP header version number of 4 / 6, Profinet uses Ethernet frame type of 0x8892, and S7 protocol uses a fixed TPKT header of 0x03. Valid data frames are located by matching characters byte by byte, and protocol handshake frames, debugging instruction frames, and empty data frames are automatically removed.

[0084] Core field extraction: For different protocols, publicly accessible core data fields are uniformly extracted. Five basic fields are fixed to be extracted: device address, running status, load value, communication duration, and heartbeat timestamp. Vendor-private debugging fields and encrypted configuration fields are discarded to ensure parsing compliance.

[0085] Cross-segment virtual aggregation: Based on the read-only access permissions of existing network segment routes, without modifying the routing configuration, penetrating the firewall, or occupying the industrial control network bandwidth, the valid data frames parsed from different network segments are aggregated into a unified algorithm processing queue in chronological order through the UTC unified timestamp synchronization mechanism, realizing logical-level data aggregation rather than physical-level data migration, and without affecting the operational stability of the original industrial control network.

[0086] This parsing rule has the advantages of read-only access, non-intrusiveness, cross-protocol compatibility, no need for additional hardware deployment, and no impact on normal device operation. It only parses publicly available fields of the protocol, is compatible with mainstream compliant industrial control equipment on the market, breaks down manufacturer barriers, and has high parsing accuracy and strong stability.

[0087] Furthermore, to address the issue of inconsistent fields across heterogeneous devices, this invention incorporates a universal field mapping template, coupled with fully automatic matching logic, to achieve standardized conversion of non-standard data. This template is a pre-built standardized field mapping table with an embedded algorithm core module. It requires no additional hardware or independent software deployment and is directly invoked synchronously with the algorithm logic. To balance universal adaptability and non-standard compatibility, the template adopts a two-level hierarchical structure, covering various conventional and special non-standard devices, achieving universal applicability across all devices.

[0088] The basic mapping layer is responsible for mapping the non-standard fields of different devices to the eight core fields of carbon metering, namely, unique device identifier, network segment information, communication timestamp, operating load rate, start / stop status, heartbeat interval, communication duration, and rated power. The order and naming of the fields are fixed to prevent field confusion from the source.

[0089] The adaptation and correction layer is designed for special non-standard equipment. A fault-tolerant matching mechanism is set up. By using fuzzy matching of keywords such as "load", "power", "running time" and "start-stop", non-standard fields are normalized. The matching threshold can be adaptively adjusted, with a minimum matching degree of no less than 70%, ensuring the compatibility effect of non-standard equipment.

[0090] The entire mapping logic is deeply integrated with the cross-network segment aggregation logic, employing a keyword hash matching algorithm for fully automated matching. The algorithm logic is fixed throughout, with no manual intervention required, and the execution steps are rigorous and standardized: First, the non-standard field names of the devices after protocol parsing are uniformly formatted, removing special characters, spaces, and differences in capitalization, and converting them into standard string format; second, the hash values ​​of the non-standard fields and the template standard fields are calculated using a preset hash function to establish a hash comparison relationship; then, a quantitative matching threshold is set, and field mapping is automatically completed when the hash matching degree is ≥85%, fault-tolerant fuzzy matching is triggered when the matching degree is between 70% and 85%, and fields with a matching degree <70% are marked as fields to be verified, with the algorithm automatically supplementing the device's default parameters; the entire algorithm execution order is fixed as "first cross-network segment protocol parsing, then field normalization mapping, and finally data aggregation processing," achieving integrated flow of parsing, aggregation, and mapping without the need for separate module deployment, thus eliminating data gaps.

[0091] Third, a hierarchical data preprocessing mechanism is established. Referring to GB / T 32151-2015 "Requirements for Greenhouse Gas Emission Accounting and Reporting" and industrial control data processing specifications, and relying on the general data cleaning standard for industrial carbon measurement proposed in this invention, precise data filtering and standardization are achieved. The specific content and implementation methods are as follows:

[0092] Level 1 Processing (Invalid Data Removal): A dual-dimensional threshold start / stop determination algorithm is used to remove invalid operating condition data. This algorithm combines communication activity and load values ​​for joint determination, avoiding misjudgment based on a single parameter.

[0093] 1. Communication activity determination: Extract the heartbeat packet frequency and communication message transmission and reception volume of the device per unit time (1 minute), and set a fixed threshold: the heartbeat interval of the normally operating device is ≤30s and the number of messages transmitted and received per unit time is ≥5; the heartbeat interval is >60s and there are no messages transmitted or received, which is determined to be shutdown; the heartbeat interval is 30s-60s and is determined to be standby.

[0094] 2. Load value determination: Real-time load rate of the equipment. A load rate of 0% or less, lasting for ≥5 minutes, is considered a shutdown; a load rate of 0% or less is considered a shutdown. <5%, duration ≥2min, is judged as standby;

[0095] 3. Joint Judgment Rule: If any one dimension of the shutdown / standby condition is met, the algorithm automatically marks all data in the corresponding time period as invalid operating condition data, removes them in batches, and retains only the operating condition data that meets the normal operating condition in both dimensions. It is compatible with the start and stop status recognition of all industrial control equipment, with a false judgment rate of ≤1%.

[0096] Secondary processing (outlier data filtering): After removing invalid data, for continuous numerical data such as load rate, power, and communication duration, outliers are removed using the complete Raida 3σ criterion algorithm, as detailed below:

[0097] 1. Select 100 consecutive valid data sets as the sample set and calculate the arithmetic mean of the dataset. Where n=100, This is a single set of sample data;

[0098] 2. Calculate the standard deviation of the dataset. ;

[0099] 3. Set the anomaly detection range Values ​​exceeding this range are identified as abnormal data and automatically removed. Simultaneously, combined with dual verification of equipment rated parameter thresholds, power data exceeding 1.2 times the rated power and load rate exceeding the 0-100% range are directly identified as abnormal. This dual filtering ensures data accuracy, with an abnormality filtering accuracy rate of ≥96%.

[0100] Level 3 processing (data normalization calibration): A standardized process combining linear normalization and unified unit conversion is used to normalize and calibrate the data. The specific process is as follows:

[0101] 1. Unit Conversion: Power units are converted according to the formula. Convert watts to kilowatts; time units are calculated using the formula. Seconds will be uniformly converted to hours;

[0102] 2. Consistent numerical precision:

[0103] All values ​​are rounded to two decimal places to ensure consistent calculation methods.

[0104] 3. Time base calibration:

[0105] Extract the device's local timestamp and compensate using the UTC standard time difference formula. ,in, For the calibrated standard timestamp, For device local timestamps The algorithm automatically acquires and calibrates the clock deviation value for devices across network segments, eliminating clock differences across network segments.

[0106] 4. Consistent format:

[0107] All calibrated data is stored in a unified key-value pair format with a fixed field order to facilitate subsequent model calculations.

[0108] Level 4 processing (duplicate data deduplication): Data is deduplicated using a hash deduplication algorithm, specifically:

[0109] Based on the primary key composed of a unique device identifier and a communication timestamp, a hash deduplication algorithm is used to calculate the hash value of the combined primary key. Data with duplicate hash values ​​are identified as duplicate data, and redundant entries are automatically deleted, retaining only a single valid data entry.

[0110] Fourth, a lightweight, tiered storage and real-time processing mechanism is adopted to achieve efficient data storage and rapid computation. Data is stored using existing industrial control hosts or servers on-site. The specific creation and operation logic for each storage level is as follows:

[0111] The real-time processing memory queue adopts a first-in-first-out (FIFO) circular memory queue, which is automatically initialized when the algorithm starts. A fixed memory space is pre-allocated (default 256M, adaptively adjustable according to the computing power of the on-site equipment, minimum 128M, maximum 512M). The queue length is fixed at 1000 data entries, requiring no manual creation. Each cleaned and valid data entry enters the tail of the queue. The algorithm reads data sequentially from the head of the queue for model calculation. After each data entry is calculated, it is immediately removed from the queue, releasing the corresponding memory space. The queue operates in a continuous loop, preventing data accumulation and maintaining a memory utilization rate consistently below 60%, avoiding memory overflow. The enqueue-dequeue latency for a single data entry is ≤50ms.

[0112] Short-term archived data: Stored in compressed CSV format, retaining only eight core fields (device unique identifier, network segment information, communication timestamp, operating load rate, start / stop status, heartbeat interval, communication duration, and rated power). It uses the LZ77 compression algorithm to compress the data volume to 1 / 10 of the original data. The storage path defaults to the existing server's non-system disk, and folders are automatically created for archiving by date.

[0113] To ensure efficient data processing, this invention employs a streaming data processing algorithm, with a latency of ≤100ms for the entire process from data entry to completion of calculation for a single data item, supporting concurrent data processing for up to 100 devices across network segments.

[0114] Long-term historical data cleanup mechanism: An automatic cleanup algorithm using scheduled tasks is employed. The specific process is as follows:

[0115] 1. Time-triggered: The algorithm has a built-in real-time calendar clock, which automatically starts the cleanup task at 2:00 AM every day (during periods of low load and no production in industrial sites), without affecting normal production;

[0116] 2. Data Filtering: The algorithm traverses the historical storage folder, filters out the original detailed data and compressed CSV files with a storage duration of >365 days, and extracts the daily average carbon emission value, total energy consumption, and total carbon emission of the corresponding date.

[0117] 3. Summary Retention: The three summary values ​​are appended to the historical summary data table and permanently retained to ensure data traceability;

[0118] 4. Automatic Deletion: The algorithm batch deletes the corresponding original detailed data and compressed CSV files. After the cleanup is completed, a cleanup log is automatically generated for future reference, including the cleanup time, the amount of data cleaned, and the summary value of the remaining data.

[0119] This step constructs a parsing and aggregation system for deep integration of heterogeneous data across network segments and protocols. By deploying basic industrial-grade parsing and acquisition equipment and cooperating with standardized data preprocessing processes, it achieves full-link, 24 / 7 real-time reliable processing from data parsing, aggregation, cleaning to storage, providing high-quality data support for carbon flow metering.

[0120] Step S102: Construct a coupled calculation model of industrial control communication-energy consumption-equipment carbon emission, and combine it with a four-level dynamic carbon factor real-time correction method to achieve accurate measurement of cross-network segment equipment carbon flow time-series data.

[0121] In this step, an industrial server with an i5-12400 CPU is used as the algorithm processing center. The coupled computational model and dynamic carbon factor correction algorithm proposed in this invention are deployed there. The server has 8GB of built-in memory and a 512GB solid-state drive, supports multi-threaded parallel computing, and has a carbon emission calculation latency of ≤10ms per device. It can support concurrent calculation for 100 devices across network segments. The specific process is as follows:

[0122] First, the coupled computation model employs a two-level lightweight computational logic, eliminating the need for complex high-order operations. All formula parameters are derived from communication parsing data or device nameplates. The specific computational process is as follows:

[0123] Level 1: Energy consumption conversion logic. For different types of industrial control equipment across network segments, based on the operating load rate, communication duration, and rated power parsed from communication data, a formula for calculating energy consumption per unit operating time is established:

[0124]

[0125] In the formula: Energy consumption per equipment cycle (kWh); The rated power of the equipment (kW) is taken from the parameters on the equipment nameplate. The algorithm supports manual input or automatic reading. The device's real-time load rate (%) is derived from communication data parsing. The effective operating time of the device (h) is calculated by excluding invalid communication and standby time. This is the working condition correction factor, with a value of 0.8-1.0. When the equipment load is stable, it is taken as 1.0, and when the load fluctuates greatly, it is taken as 0.8-0.9. The algorithm automatically assigns a value according to the load fluctuation range.

[0126] Level Two: Current carbon emission measurement of equipment, introducing dynamic carbon factors, and combining real-time correction with multi-dimensional parameters to establish a formula for calculating current carbon emission of equipment:

[0127]

[0128] In the formula: The carbon emission value of the equipment in a single cycle (for the current period) ); For dynamic carbon factor ( / kWh).

[0129] Second, the dynamic carbon factor is adjusted in real time through a four-level correction process and algorithm-based solidification. The correction process is fully embedded with energy consumption-carbon emission calculation algorithms and synchronized with the data processing cycle. The specific correction process is as follows:

[0130] 1. Algorithm trigger node: A correction calculation is triggered every 10 minutes, which is fully synchronized with the equipment data acquisition cycle. After the correction is completed, the global variable of carbon factor in the algorithm is automatically updated, and all calculations in the current period use the latest factor;

[0131] 2. Basic Factor Retrieval: The algorithm incorporates the national standard GB / T 32151-2015 basic carbon factor library, and by default retrieves the power grid supply reference factor. / kWh can be directly assigned to the initial factor variable without manual configuration.

[0132] 3. Regional Green Energy Ratio Correction: The algorithm periodically reads publicly available green energy ratio data from the regional power grid. , corrected formula: The green electricity ratio data is automatically updated daily; when real-time data is unavailable, the average value of the previous month is used.

[0133] 4. Equipment load correction: The algorithm obtains the equipment load rate in real time. Automatically determine the load range: In the efficient range, ; In the inefficient range, ; , The range is not adjusted;

[0134] 5. Grid Peak-Valley Time Correction: The algorithm reads the system's local time and automatically determines the grid peak-valley time period. Peak segment No corrections are made for level sections;

[0135] 6. Factor locking and application: Final dynamic factor after correction The algorithm automatically locks this factor and substitutes it into the formula for calculating the current carbon emission value of the equipment. Once the current carbon emission calculation is completed, the full correction process will be triggered again in the next 10-minute cycle.

[0136] Third, a dual verification mechanism is adopted to ensure the accuracy of carbon flow measurement. After each carbon emission calculation, the algorithm automatically compares the carbon emission values ​​of the same type of equipment with the historical data of the same period. If the deviation exceeds ±10%, the data cleaning and model calculation are automatically re-executed. If the deviation still exists, the data is marked as data to be verified, a verification prompt is generated and pushed to the operation and maintenance personnel. At the same time, the dynamic carbon factor correction parameters are automatically calibrated every hour. Combined with real-time data of the regional power grid, the correction formula coefficients are optimized to improve the measurement accuracy.

[0137] Based on the above coupled calculation model, three-level carbon flow classification and metering can be achieved for a single cross-network segment device, a single communication link, and the entire plant network segment, clearly tracing the entire path of carbon flow generation, transmission, and loss, and accurately locating high-emission equipment and inefficient communication links.

[0138] Step S103: Establish a graded judgment mechanism for abnormal equipment carbon emission values ​​based on carbon flow time series data, realize the automatic identification and accurate marking of abnormal carbon flow time series data, and provide a basis for low-carbon regulation. The specific process is as follows:

[0139] Based on the low-carbon emission reduction standards of the industrial sector, the corresponding rated carbon emission benchmark values ​​of equipment and the operating condition loss coefficient, establish a standard for determining the first level of normal operation, the second level of early warning, and the third level of exceeding the standard for cross-network segment equipment.

[0140] Preset benchmark values: The algorithm processing center has a built-in library of rated carbon emission benchmark values ​​for commonly used industrial control equipment, and also supports manual input of benchmark values ​​from the nameplate of individual equipment. It is automatically saved after entry, eliminating the need for repeated configuration;

[0141] Real-time calculation and comparison: The algorithm calculates the current carbon emission value of the equipment. After calculation, the carbon emission ratio will be automatically calculated: The calculation result should be rounded to one decimal place.

[0142] Automatic grading determination: Automatically grade and mark based on ratio, synchronously triggering the corresponding subsequent control modules:

[0143] Establish a first-level normal judgment standard: δ≤90%, the algorithm marks it as a normal operating state, no control operation, continuously monitor the equipment operation data and equipment carbon emission values ​​in real time, record the equipment carbon emission values ​​once an hour, and store them in the data ledger;

[0144] Establish a two-level early warning judgment standard: 90% < δ ≤ 110%, the algorithm marks it as a slightly high emission state, automatically generates a yellow early warning label, synchronously triggers the mild control module, records the early warning time, equipment information, carbon emission exceedance range, and pushes early warning prompts to operation and maintenance personnel;

[0145] A three-level standard for judging excessive emissions is established: when δ>110%, the algorithm marks it as a state of severe high emissions, automatically generates a red excessive emission label, simultaneously triggers the severe control module, immediately pushes an emergency warning, and records the excessive data and equipment operating conditions in detail for easy subsequent traceability and analysis.

[0146] in, To automatically calculate the ratio, This is the baseline value for the nameplate of a single device. This represents the current carbon emission value of the equipment.

[0147] In a preferred embodiment, after step S103, step S104 is further included:

[0148] Step S104: Establish an anomaly detection and calibration mechanism to address data loss caused by cross-network segment fluctuations and communication interruptions.

[0149] In this step, the algorithm determines the network anomaly by checking the continuity of heartbeat packets. If no heartbeat packets are received from the device for three consecutive cycles, the current anomaly determination is suspended, and the data caching mechanism is activated. After the network is restored, the device carbon emission values ​​and anomaly determination results for the missing period are recalculated, calibrated, and updated to the data ledger to avoid misjudgments caused by network fluctuations. The determination results are automatically generated and stored in the data ledger.

[0150] Step S105: Output the corresponding low-carbon collaborative control instructions according to the anomaly classification and judgment mechanism, and construct a closed-loop management and control mechanism of "metering and accounting - anomaly judgment - low-carbon control - data archiving" to realize low-carbon optimized operation of cross-network segment equipment.

[0151] In this step, the entire control scheme is solidified into executable algorithm instructions, supporting automatic output of quantitative parameter adjustment suggestions. Some parameters support direct transmission of the algorithm to the device configuration port. The corresponding low-carbon collaborative control instructions output according to the anomaly classification and judgment mechanism include:

[0152] For normally operating equipment (maintaining optimized processes): the dynamic carbon factor is automatically calibrated every hour, a daily report of equipment carbon emission values ​​is generated, data fluctuations are continuously monitored, and an early warning is automatically triggered if the fluctuation exceeds ±5% to avoid high emission risks in advance. The equipment operating parameters are optimized monthly based on historical data to maintain low-carbon and stable operation.

[0153] For secondary, low-emission equipment (low-level control process): optimize heartbeat intervals and load rates, and reduce invalid communication. Specifically:

[0154] 1. Heartbeat Interval Optimization: Automatically calculates the optimal heartbeat interval, formula: ,in, The calculated optimal heartbeat interval, The original heartbeat interval of the device is compressed by 20% by default, with a minimum of 5 seconds. The optimal interval parameter is output and transmitted directly to the device heartbeat configuration item through the data aggregation node, without the need for manual calculation.

[0155] 2. Load rate optimization: Calculate the target load range of 60%-80%, output the load adjustment gradient, increase it by 5% each time, and gradually adjust it to the high-efficiency range. During the adjustment process, monitor the equipment operation status in real time to avoid sudden load changes from affecting the stability of equipment operation.

[0156] 3. Invalid Communication Reduction: Filter out communication messages with no operation during idle periods, automatically mark redundant communication, output shutdown commands, reduce unnecessary energy consumption, and generate a configuration list of all adjustment parameters that can be directly executed. On-site personnel can import it with one click, or it can be remotely pushed to the device through the algorithm.

[0157] For Level 3 heavy-emission equipment (heavy-duty control process): optimize communication paths and regulate peak-shaving timing and load reduction under different operating conditions. Specifically:

[0158] 1. Communication path optimization: Traverse cross-network segment routing nodes, detect the communication latency of each path in real time (detection period 1s), filter the communication path with the lowest latency, output path switching instructions, and replace the original high-latency path through the aggregation node to reduce data retransmission loss;

[0159] 2. Peak-shifting timing control: Read the running timing of other devices in the same network segment, avoid high-load periods (times with load rate > 80%), generate a device peak-shifting operation schedule accurate to the minute, adjust the device startup and running times, and reduce the overall load peak;

[0160] 3. Operating condition load reduction control: For equipment with severe overload, a step-by-step load reduction plan is output, with each load reduction being 10% to gradually adjust to a reasonable range. Dynamic carbon factors are matched simultaneously, and priority is given to switching to low factor periods during off-peak hours. After each adjustment, the carbon emission value of the equipment is recalculated until the carbon emission value of the equipment returns to the normal range.

[0161] In a preferred embodiment, after step S105, step S106 is further included:

[0162] Step S106: Verify the control effects on normally operating equipment, secondary lightly high-emission equipment, and tertiary heavily high-emission equipment, calculate the emission reduction effect, and generate a control effect report. Specifically, after each control is completed, continuous monitoring is performed for 24 hours, and the carbon emission values ​​and operating parameters of the equipment before and after the control are compared to calculate the emission reduction effect and generate a control effect report.

[0163] Step S107: Determine whether the emission reduction effect has reached the expected effect (emission reduction rate <5%). If yes, proceed to S108; otherwise, proceed to S109.

[0164] Step S108: Then determine that the regulation effect has met the requirements;

[0165] Step S109: The control parameters are automatically optimized, and the process returns to step S105 until the desired effect is achieved.

[0166] Step S110: Archive carbon flow time-series data and review metrology results to establish a long-term closed-loop management mechanism and meet compliance requirements. The specific process is as follows:

[0167] The raw data, anomaly identification and judgment records, low-carbon control schemes and on-site implementation effects of carbon flow time-series data measurement for all equipment across network segments are uniformly archived according to the lightweight hierarchical storage method in step S101 to form a standardized carbon measurement ledger that is traceable, verifiable and exportable. The standardized carbon measurement ledger includes equipment information, network segment information, equipment carbon emission values, anomaly records, control records and effect verification data, which meet the compliance requirements of industrial carbon verification, emission reduction assessment and environmental data traceability.

[0168] Monthly carbon flow time-series data archiving and measurement result review are completed regularly. Specifically, the algorithm automatically iteratively optimizes the dynamic carbon factor correction coefficient and anomaly judgment threshold parameters based on historical equipment carbon emission values ​​and control records, while optimizing low-carbon control parameters to continuously improve the accuracy of carbon measurement and the adaptability of low-carbon management, forming a long-term closed-loop management mechanism.

[0169] A quarterly review report is generated, summarizing the carbon emission statistics and control effectiveness of all network equipment, providing data support for subsequent low-carbon management decisions.

[0170] This application provides a cross-network segment device carbon flow metering and control method based on industrial control communication data analysis. It constructs an integrated monitoring and early warning system for collaborative sensing and intelligent response of electricity and carbon. Through heterogeneous sensor network design (including sensor selection, communication protocol adaptation, and dynamic spectrum sensing), deep fusion of multi-source data, industrial-grade time alignment, and intelligent algorithms embedded with physical mechanisms, it solves the pain points of insufficient sensor network coverage, low communication reliability, data fragmentation, high false alarm rate, and large response delay in traditional technologies. Its breakthroughs are reflected in: achieving dynamic sensor response, reliable transmission of carbon flow time-series data across the entire chain, and reliable 24 / 7 monitoring through heterogeneous sensor network networking; a dynamic coupling mechanism between electricity and carbon flow time-series data in real time to accurately identify the causal relationship between abnormal energy consumption and leakage; hierarchical spatiotemporal alignment technology, combining time synchronization and process constraints to achieve high-precision time-series matching of multi-frequency data; a mechanism-data dual-drive model, integrating physical laws and deep learning to improve the interpretability and reliability of leakage prediction; and fully closed-loop intelligent control, linking sensing, decision-making, and execution units to form an autonomous response chain from monitoring to sealing.

[0171] To address the technical challenges of unified aggregation of heterogeneous data across industrial network segments and protocols, reliance on dedicated hardware for carbon metering, low metering accuracy, and disconnect between carbon metering and low-carbon management, this paper proposes a universal protocol parsing rule for cross-network segment industrial control data, a keyword hash field mapping template, a universal data cleaning standard for industrial carbon metering, and supporting algorithms. It also constructs a coupled calculation model of industrial control communication, energy consumption, and carbon emissions, designs a four-level dynamic carbon factor real-time correction method, and forms a complete technical solution encompassing "data processing, metering and accounting, anomaly detection, low-carbon regulation, and data archiving." The key breakthrough lies in achieving standardized processing of heterogeneous data across network segments and accurate carbon flow metering without relying on dedicated gateways, sensors, or other additional hardware. It utilizes existing industrial control networks and equipment, and through a fully integrated algorithm, it achieves standardized processing of heterogeneous data across network segments and accurate carbon flow metering. Simultaneously, it deeply integrates carbon metering with low-carbon management, resolving the industry pain points of poor implementation, weak universality, and passive management associated with traditional carbon metering.

Claims

1. A method for carbon flow metering and control of cross-network segment devices based on industrial control communication data parsing, characterized in that, include: A cross-network segment industrial control data parsing and aggregation system is constructed to achieve seamless aggregation and normalization of heterogeneous data from multiple protocols, thereby providing high-quality data support for carbon flow metering. Among them, the cross-network segment industrial control data parsing supports the parsing of five mainstream industrial control protocols: TCP / IP, Modbus-RTU, Modbus-TCP, Profinet, and S7-200 / 300. A coupled calculation model of industrial control communication, energy consumption, and equipment carbon emissions is constructed, and a four-level dynamic carbon factor real-time correction method is used to achieve accurate measurement of carbon flow time-series data of cross-network segment equipment. Establish a classification and judgment mechanism for abnormal equipment carbon emission values ​​based on carbon flow time series data, realize the automatic identification and accurate marking of abnormal carbon flow time series data, and provide a basis for low-carbon regulation; Based on the aforementioned anomaly classification and judgment mechanism, corresponding low-carbon collaborative control instructions are output, and a closed-loop management and control mechanism of "metering and accounting - anomaly judgment - low-carbon control - data archiving" is constructed to realize low-carbon optimized operation of cross-network segment equipment; Carbon flow time-series data archiving and measurement result review are used to build a long-term closed-loop management and control mechanism to meet compliance requirements. The computational model employs a two-level lightweight computational logic, including: Level 1: Energy consumption conversion logic. For different types of industrial control equipment across network segments, based on the operating load rate, communication duration, and rated power parsed from communication data, a formula for calculating energy consumption per unit operating time is established: in, Energy consumption per equipment cycle (kWh); Rated power of the equipment (kW); Real-time load rate of the equipment (%) The effective operating time of the device (h); This is the working condition correction factor, with a value ranging from 0.8 to 1.0; Level 2: The logic for measuring the current carbon emissions of the equipment introduces a dynamic carbon factor and combines it with real-time correction of multi-dimensional parameters to establish a formula for calculating the current carbon emissions of the equipment. in, The carbon emission value of the equipment in a single cycle (for the current period) ); For dynamic carbon factor ( / kWh).

2. The method for cross-network segment device carbon flow metering and control based on industrial control communication data parsing according to claim 1, characterized in that, The construction of a cross-network segment industrial control data universal parsing and aggregation system to achieve seamless aggregation and normalization of multi-protocol heterogeneous data includes: In cross-network segment industrial control scenarios, industrial-grade network resolution terminals and data aggregation nodes are deployed, adopting a hybrid architecture of "distributed resolution + centralized aggregation"; Construct a multi-protocol fusion parsing and field mapping architecture; Establish a hierarchical data preprocessing mechanism; Lightweight hierarchical storage and real-time processing mechanisms are adopted to enable data storage and computation.

3. The method for carbon flow metering and control of cross-network segment devices based on industrial control communication data parsing according to claim 2, characterized in that, The establishment of a hierarchical data preprocessing mechanism includes: A dual-dimensional threshold start / stop determination algorithm is adopted to remove invalid operating condition data. The algorithm combines communication activity and load value for joint determination to avoid misjudgment based on a single parameter. For continuous numerical data such as load rate, power, and communication duration, outliers are removed using the complete algorithm based on the Laida 3σ criterion. A standardized process combining linear normalization and unified unit conversion is used to normalize and calibrate the data; Data is deduplicated using a hash deduplication algorithm.

4. The method for carbon flow metering and control of cross-network segment devices based on industrial control communication data parsing according to claim 1, characterized in that, The construction of the coupled calculation model of industrial control communication-energy consumption-equipment carbon emissions, combined with the four-level dynamic carbon factor real-time correction method, to achieve accurate measurement of cross-network segment equipment carbon flow time-series data includes: The computational model employs a two-level lightweight computational logic. The dynamic carbon factor is adjusted in real time through a four-level correction process and algorithm solidification linkage. The correction process is embedded with the energy consumption-carbon emission calculation algorithm throughout the process and is synchronized with the data processing cycle. A dual verification mechanism is employed to ensure the accuracy of the carbon flow time-series data measurement.

5. The method for carbon flow metering and control of cross-network segment devices based on industrial control communication data parsing according to claim 1, characterized in that, The establishment of a device carbon emission value anomaly classification and judgment mechanism based on carbon flow time series data, to achieve automatic identification and accurate marking of carbon flow time series data anomalies, includes: Based on the low-carbon emission reduction standards of the industrial sector, the corresponding rated carbon emission benchmark values ​​of equipment and the operating condition loss coefficient, establish a standard for determining the first level of normal operation, the second level of early warning, and the third level of exceeding the standard for cross-network segment equipment. Establish a first-level normal judgment standard: δ≤90%, the algorithm marks it as a normal operating state, no control operation, continuously monitor the equipment operation data and equipment carbon emission values ​​in real time, record the equipment carbon emission values ​​once an hour, and store them in the data ledger; Establish a two-level early warning judgment standard: 90% < δ ≤ 110%, the algorithm marks it as a slightly high emission state, automatically generates a yellow early warning label, synchronously triggers the mild control module, records the early warning time, equipment information, carbon emission exceedance range, and pushes early warning prompts to operation and maintenance personnel; A three-level standard for judging excessive emissions is established: when δ>110%, the algorithm marks it as a state of severe high emissions, automatically generates a red excessive emission label, simultaneously triggers the severe control module, immediately pushes an emergency warning, and records the excessive data and equipment operating conditions in detail for easy subsequent traceability and analysis. in, , To automatically calculate the ratio, This is the baseline value for the nameplate of a single device. This represents the current carbon emission value of the equipment.

6. The method for carbon flow metering and control of cross-network segment devices based on industrial control communication data parsing according to claim 5, characterized in that, After establishing the device carbon emission value anomaly classification and judgment mechanism based on carbon flow time series data, the following is also included: To address data loss caused by cross-network segment fluctuations and communication interruptions, an anomaly detection and calibration mechanism should be established.

7. The method for carbon flow metering and control of cross-network segment devices based on industrial control communication data parsing according to claim 5, characterized in that, The low-carbon collaborative regulation instruction output according to the anomaly classification and determination mechanism includes: For normally operating equipment: the dynamic carbon factor is automatically calibrated every cycle, a daily statistical report of equipment carbon emissions is generated, data fluctuations are continuously monitored, and an early warning is automatically triggered if the fluctuation exceeds ±5%, so as to avoid high emission risks in advance and maintain low-carbon and stable operation. For secondary, low-emission equipment: optimize heartbeat interval and load rate, and reduce invalid communication; For Level 3 heavy-emission equipment: optimize communication paths and regulate peak-shifting timing and load reduction under different operating conditions.

8. The method for carbon flow metering and control of cross-network segment devices based on industrial control communication data parsing according to claim 7, characterized in that, After outputting the corresponding low-carbon coordinated control command according to the anomaly classification and determination mechanism, the method further includes: The control effects on normally operating equipment, secondary light-emission equipment, and tertiary heavy-emission equipment are verified, and the emission reduction effect is calculated and a control effect report is generated. Determine whether the emission reduction effect has achieved the expected results; If yes, then the control effect is determined to meet the requirements; if no, then the control parameters are automatically optimized, and the process returns to the step of outputting the corresponding low-carbon coordinated control command according to the anomaly classification judgment mechanism, until the expected effect is achieved.

9. The method for carbon flow metering and control of cross-network segment devices based on industrial control communication data parsing according to claim 2, characterized in that, The carbon flow time-series data archiving and metrology result review includes: The raw carbon flow time series data of all equipment across network segments, the calculated equipment carbon emission values, anomaly identification and judgment records, low-carbon control schemes and on-site implementation effects are uniformly archived according to the aforementioned lightweight hierarchical storage method to form a standardized carbon metering ledger that is traceable, verifiable and exportable. The standardized carbon metering ledger includes equipment information, network segment information, equipment carbon emission values, anomaly records and effect verification data, which meets the compliance requirements of industrial carbon verification, emission reduction assessment and environmental data traceability. Monthly carbon flow time-series data archiving and measurement result review are completed regularly, and quarterly review reports are generated to summarize the carbon emission statistics and control effectiveness of equipment across the entire network segment, providing data support for subsequent low-carbon management decisions.

Images