A method and device for monitoring and managing carbon emissions from a substation

Abstract

This application relates to a method and apparatus for carbon emission monitoring and management in substations. This application pertains to the field of environmental governance technology and includes the following steps: acquiring a dataset from the substation; constructing a multi-scale carbon emission virtual model based on the dataset; monitoring carbon emissions based on the multi-scale carbon emission virtual model and the substation dataset to generate a carbon emission equipment-level traceability report; formulating a carbon emission equipment optimization strategy based on the carbon emission equipment-level traceability report and carbon emission gas leakage prediction data; and performing closed-loop feedback of the carbon emission equipment optimization strategy on the multi-scale carbon emission virtual model to execute the substation's carbon emission monitoring and management operations. Through the multi-scale carbon emission virtual model, intelligent traceability and prediction, and closed-loop feedback of the optimization strategy, the overall intelligence and systematization level of substation carbon emission management is improved.

Description

Technical Field

[0001] This application relates to the field of environmental governance technology, and in particular to a method and apparatus for monitoring and managing carbon emissions from a substation. Background Technology

[0002] The methods for monitoring carbon emissions in substations mainly rely on manual detection and periodic inspections. These methods are inefficient, inaccurate, and cannot keep track of carbon emissions in real time. They are also difficult to cope with complex power operating environments.

[0003] With the rapid increase in labor costs, there is an urgent need for new carbon emission monitoring models to meet the needs of substations. Summary of the Invention

[0004] This application provides a carbon emission monitoring and management method and apparatus for substations, to provide a new carbon emission monitoring mode to meet the needs of substations.

[0005] In a first aspect, this application provides a method for monitoring and managing carbon emissions from a substation, comprising the following steps:

[0006] The process involves acquiring a dataset from a substation and constructing a multi-scale virtual carbon emission model based on that dataset. Specifically, this can involve acquiring substation deployment information data, collecting substation data from the deployment information data to obtain a substation dataset, and constructing a multi-scale virtual 3D carbon emission model from the substation dataset to generate the multi-scale virtual carbon emission model.

[0007] Carbon emission monitoring is conducted based on a multi-scale carbon emission virtual model and data collected from substations to generate a carbon emission equipment-level source tracing report. Specifically, this can be achieved by monitoring carbon emissions from substation data collected using a multi-scale carbon emission virtual model to generate carbon emission monitoring data; calculating multi-level carbon emission contribution based on the carbon emission monitoring data using the multi-scale carbon emission virtual model to obtain a multi-dimensional emission contribution matrix; and using the multi-dimensional emission contribution matrix to trace abnormal emissions from the carbon emission monitoring data to generate a carbon emission equipment-level source tracing report.

[0008] Based on carbon emission equipment-level traceability reports and carbon emission gas leakage prediction data, carbon emission equipment optimization strategies are formulated. Specifically, this can involve using carbon emission equipment-level traceability reports to identify high-emission power plant areas in a multi-scale carbon emission virtual model, generating high-emission power plant area data; performing intelligent leakage prediction on the high-emission power plant area data, thereby generating carbon emission gas leakage prediction data; and generating local and overall carbon emission management strategies based on the carbon emission gas leakage prediction data, high-emission equipment location data, and carbon emission equipment-level traceability reports, resulting in high-carbon emission equipment optimization strategies and overall carbon emission equipment optimization strategies.

[0009] Based on carbon emission equipment optimization strategies, a multi-scale carbon emission virtual model is used to perform carbon emission closed-loop feedback to execute carbon emission monitoring and management operations for substations. Specifically, this can be achieved by performing carbon emission closed-loop feedback on a multi-scale carbon emission virtual model based on high-carbon emission equipment optimization strategies and overall carbon emission equipment optimization strategies, generating carbon emission optimization closed-loop feedback data; and then performing a comprehensive carbon emission evaluation on the carbon emission optimization closed-loop feedback data according to preset carbon emission evaluation indicators to execute carbon emission monitoring and management operations for substations.

[0010] This application, through the collection of deployment information data from substations and the construction of a multi-scale virtual 3D model of carbon emissions, can accurately reflect the operating status and environment of substations, providing a high-precision data foundation to support subsequent carbon emission monitoring and analysis. This virtual model helps to assess carbon emissions from multiple angles and levels, enhancing the comprehensiveness and precision of monitoring. By monitoring carbon emissions through the multi-scale virtual model and tracing sources based on emission contribution calculations, the source of abnormal emissions can be accurately located, and specific emitting equipment and processes can be identified. This provides an effective means for timely detection of emission problems, promotes precise carbon emission management, and helps to take targeted measures to reduce emissions. By identifying high-emission areas and performing intelligent leak prediction, potential carbon emission risks can be detected in advance, preventing problems such as gas leaks. Furthermore, carbon emission equipment optimization strategies and overall optimization strategies generated based on leak prediction data can improve equipment operating efficiency, reduce unnecessary carbon emissions, and optimize overall carbon emission management. Through closed-loop feedback of carbon emissions, the effectiveness of carbon emission optimization measures can be monitored in real time, and a comprehensive evaluation can be conducted based on preset evaluation indicators. This feedback mechanism ensures dynamic adjustment and continuous improvement of carbon emission management, enabling substations to continuously optimize their carbon emission control strategies and achieve long-term, stable low-carbon operation goals. Therefore, this invention improves the overall intelligence and systematization level of substation carbon emission management through multi-scale carbon emission virtual models, intelligent source tracing and prediction, and closed-loop feedback of optimization strategies.

[0011] In some embodiments, acquiring the data collection dataset from the substation and constructing a multi-scale carbon emission virtual model based on the data collection dataset from the substation includes:

[0012] Obtain deployment information data for substations;

[0013] The deployment information data of substations is analyzed to generate substation geographic topology data; based on the substation geographic topology data, multi-source sensors are used to collect substation data to obtain substation data collection dataset;

[0014] The data acquisition dataset from the substation is preprocessed to generate a standard substation acquisition dataset.

[0015] By using digital twin technology, a multi-scale virtual 3D model of carbon emissions is constructed from data collected from standard substations, thereby generating a multi-scale virtual carbon emissions model.

[0016] Substation data acquisition includes energy consumption, gas, and environmental data collection. Data preprocessing includes data cleaning, noise reduction, missing value imputation, and standardization. Data acquisition from multiple sensors (including energy consumption, gas, and environmental data) allows for a more comprehensive capture of various substation information. This not only enhances data richness but also ensures accuracy across multiple dimensions, providing a more reliable foundation for subsequent analysis. Preprocessing substation data (such as data cleaning, noise reduction, missing value imputation, and standardization) effectively removes invalid information and reduces data noise, thereby improving data quality and consistency and facilitating subsequent in-depth analysis and modeling. Through digital twin technology and the construction of multi-scale virtual 3D carbon emission models, dynamic visualization and refined management of substation carbon emissions can be achieved. Multi-scale models can perform multi-dimensional analysis of substation carbon emissions from different spatial and temporal scales, supporting more precise carbon emission control and optimization, and further promoting the achievement of green electricity and energy conservation and emission reduction goals. This process, through the construction of virtual models, not only enables real-time monitoring of substation carbon emissions but also provides data-driven decision support, helping managers adopt more rational energy-saving measures and optimize substation operating efficiency. The combination of digital twins and virtual 3D models allows for intelligent management of substations. Utilizing big data and AI analysis models, real-time predictions of substation energy consumption and environmental changes can be made, automatically adjusting and optimizing operations to achieve refined management of energy consumption and precise control of carbon emissions.

[0017] In some embodiments, the step of constructing a multi-scale virtual three-dimensional model of carbon emissions from data collected at a standard substation using digital twin technology, thereby generating a multi-scale virtual carbon emission model, includes:

[0018] Based on data collected from standard substations, a model-level deployment strategy is designed to generate model-level deployment design data, which includes equipment-level deployment data, system-level deployment data, and regional-level deployment data.

[0019] Based on equipment-level deployment data, a mathematical model is constructed between the operating parameters of a single device and its carbon emission characteristics using data collected from standard substations, thus obtaining a single-device carbon emission characteristic model; and a mathematical model of coordinated emission characteristics is constructed using system-level deployment data to obtain a system-level carbon emission model.

[0020] Based on regional deployment data, a mathematical model is constructed to represent the contribution of carbon emissions from the environment surrounding the substation to the system-level carbon emission model, thus obtaining a regional carbon emission model. Based on digital twin technology, the carbon emission characteristic model of a single device, the system-level carbon emission model, and the regional carbon emission model are integrated and virtualized at the model level to generate a multi-scale carbon emission virtual model.

[0021] By designing a hierarchical deployment strategy, data acquisition and carbon emission monitoring can be effectively planned at different levels (equipment, system, region). This multi-level data deployment not only improves substation management efficiency but also enables precise control and optimization of carbon emissions at each level, making substation operation and maintenance more scientific and efficient. Mathematical modeling of the carbon emission characteristics of individual devices allows for a precise understanding of the contribution of each device's operation to carbon emissions, further optimizing equipment energy efficiency and emission levels. Simultaneously, the construction of a system-level carbon emission model enables an overall assessment of the synergistic emission effects among various devices within the substation, supporting system-level optimization and enhancing the overall carbon emission management capabilities of the substation. Constructing a mathematical model of the carbon emission contribution of the substation's surrounding environment through regionally deployed data allows for the incorporation of environmental factors (such as climate and geographical location) into carbon emission prediction and optimization. This environmentally conscious carbon emission model more comprehensively reflects actual emissions, providing strong support for developing more precise carbon reduction measures. Digital twin technology integrates and virtualizes carbon emission models at the individual device, system, and regional levels, enabling substations to dynamically monitor carbon emissions across different spatial and temporal scales. This multi-scale virtual carbon emission model not only allows for real-time analysis but also helps managers optimize and adjust emissions at different levels, enhancing the substation's green operation capabilities. Based on this multi-scale virtual carbon emission model, substations can achieve more precise carbon emission control. Through real-time data feedback and model optimization, carbon emissions can be minimized while meeting electricity demand, which is crucial for achieving green electricity and complying with carbon emission policies.

[0022] In some embodiments, the step of monitoring carbon emissions based on a multi-scale carbon emission virtual model and a dataset collected from a substation to generate a carbon emission equipment-level traceability report includes:

[0023] Carbon emission monitoring is performed on the substation data set using a multi-scale carbon emission virtual model, generating carbon emission monitoring data. The carbon emission monitoring data is then used to define the causal structure of carbon emissions, generating data defining the causal structure of carbon emissions.

[0024] Based on the multi-scale carbon emission virtual model, the carbon emission causal structure is defined to calculate the carbon emission contribution of carbon emission monitoring data at multiple levels, thereby obtaining a multi-dimensional emission contribution matrix.

[0025] The multidimensional emission contribution matrix is ​​used to detect dynamic anomalies in carbon emission equipment based on carbon emission monitoring data, generating dynamic anomalies in carbon emission. Based on these dynamic anomalies, the abnormal emission equipment is located in the carbon emission monitoring data, generating high-emission equipment location data.

[0026] By integrating the location data of high-emission equipment and the source-tracing data of dynamic anomalies in carbon emissions, a carbon emission equipment-level source-tracing report is generated.

[0027] By applying multi-scale carbon emission virtual models, substations can accurately monitor carbon emissions from collected data. This not only allows for real-time tracking of carbon emissions from various devices and systems but also generates comprehensive carbon emission monitoring data, providing accurate foundational data for subsequent analysis and decision-making. Defining the causal structure of carbon emission monitoring data reveals the interrelationships between various factors within the substation, helping to identify key drivers and potential impact paths of carbon emissions, thus providing a scientific basis for optimizing emission management. Calculating multi-level carbon emission contributions based on the defined causal structure data enables analysis of carbon emission contributions from various stages, devices, and systems from multiple dimensions. This not only helps identify major emission sources but also allows for comparison and optimization of carbon emissions at different levels (such as equipment, systems, and regions), further improving management efficiency. Dynamic anomaly detection using a multi-dimensional emission contribution matrix in carbon emission monitoring data allows for timely detection of devices or stages with abnormal carbon emissions, enabling rapid measures to prevent excessive emissions. This real-time anomaly detection mechanism effectively improves the response speed of emission management and reduces the risk of exceeding carbon emission limits. Based on dynamic anomaly detection and high-emission equipment location data, specific equipment can be accurately traced to determine the source of abnormal emissions. This provides valuable information for substation operation and maintenance management, helping to quickly locate problematic equipment for repair or optimization, improving equipment operating efficiency and environmental protection levels. The generated carbon emission equipment-level traceability report can provide managers with detailed emission source analysis and abnormal emission situations. This report can support decision-makers in energy conservation and emission reduction, equipment optimization, and policy implementation, ensuring that substations improve energy efficiency while complying with environmental regulations.

[0028] In some embodiments, C device，i C contributes to the carbon emissions of device i system，k C is the carbon emission contribution of system k. region The contribution of carbon emissions is categorized by region, and Ctotal represents the total carbon emissions. The calculation of carbon emission contributions at multiple levels satisfies the following:

[0029]

[0030] Where: w i Let i be the operating weight of device i;

[0031] φ i Let be the emission factor of device i;

[0032] P i The operating power of device i;

[0033] N is the total number of devices;

[0034] S k Let k be the set of devices in system k;

[0035] M represents the total number of systems;

[0036] S l Let be the set of devices in system l.

[0037] By analyzing and integrating a multi-level carbon emission contribution calculation formula, the equipment-level formula can accurately calculate the carbon emission contribution of individual devices, clearly identifying which devices are the main emission sources. This provides direct data support for optimizing equipment operating parameters (such as power and load distribution), helping to reduce emissions. A transformer's weight, emission factor, and power value result in the highest equipment-level contribution, allowing for optimization of its operating strategy (such as reducing high-load periods). The system-level formula, by integrating equipment-level data, reveals the combined impact of synergistic effects or operating modes between devices on carbon emissions. It helps identify which system operating strategies or configurations contribute significantly to carbon emissions, thereby optimizing overall system operation. The high emission contribution generated by multiple devices operating collaboratively within a power system indicates that system scheduling strategies need adjustment (such as time-sharing operation or zoned scheduling). The regional-level formula separates the carbon emission contributions of devices and systems from the total contribution, quantifying the impact of external environmental factors (such as climate conditions and transmission losses) on emissions. This helps optimize the external layout and regional strategies of substations (such as improving insulation conditions or optimizing transmission efficiency). High humidity environments lead to increased leakage in insulation equipment, resulting in high regional-level emissions, indicating a need for enhanced regional meteorological monitoring or insulation performance improvements. The equipment-level, system-level, and regional-level emission data provided by the formula can be used as input, combined with real-time monitoring data, to achieve dynamic source tracing of high-emission equipment and abnormal processes. Source tracing results help to pinpoint the source of problems in real time and generate optimization suggestions, improving the timeliness of emission control. When regional emission anomalies are detected during a certain period, problematic equipment can be quickly located and emergency countermeasures formulated through equipment-level and system-level matrices. The hierarchical calculation formula simplifies the processing flow of complex emission data, achieving step-by-step analysis and optimization from single equipment to system to region. Data at different levels can flexibly support specific equipment optimization, system strategy improvement, or regional planning. Equipment-level analysis reveals excessively high emissions from older equipment, prioritizing upgrades; system-level analysis adjusts the operational priority of high-efficiency equipment. The weight parameters in the formula (such as operating weights and emission factors) and the emission contribution matrix can be dynamically adjusted to adapt to different operating conditions and external environments. Combining causal inference algorithms and deep learning models can further uncover complex relationships, supporting intelligent emission optimization and prediction. By training emission causal models using historical data and adjusting the weighting parameters in the formulas, dynamic predictions of future emission trends can be achieved. Regional-level calculation results can reflect the specific impact of invisible carbon emissions (such as SF6 leaks), providing a quantitative basis for promoting environmentally friendly alternative technologies (such as C4-FN gas). System-level data supports energy conservation and emission reduction by optimizing overall operating strategies (such as activating low-emission equipment during low-load periods). For high-emission regional data, priority can be given to the application of environmentally friendly insulation materials or the development of highly sensitive monitoring equipment.When using conventional formulas for calculating multi-level carbon emission contributions, equipment-level, system-level, and regional-level carbon emission contribution values ​​can be obtained. By applying the multi-level carbon emission contribution calculation formula provided by this invention, the equipment-level, system-level, and regional-level carbon emission contribution values ​​can be calculated more accurately, and the three contribution values ​​can be represented in a matrix to obtain the final multi-dimensional emission contribution matrix. This formula, through hierarchical analysis at the equipment, system, and regional levels, can accurately locate carbon emission sources, quantitatively assess synergistic effects, support intelligent source tracing management, and global optimization. This approach provides strong data support for carbon emission monitoring and management in substations, while promoting the application of intelligent and green technologies, ultimately achieving the goal of carbon reduction and efficiency improvement.

[0038] In some embodiments, the step of using a multidimensional emission contribution matrix to detect dynamic anomalies in carbon emission monitoring data, generating dynamic anomalies in carbon emission monitoring data, and locating abnormal emission equipment based on these anomalies to generate high-emission equipment location data includes:

[0039] A multidimensional emission contribution matrix is ​​used to detect dynamic anomalies in carbon emission equipment based on carbon emission monitoring data, and dynamic anomalies in carbon emission are generated.

[0040] Based on the dynamic anomalies in carbon emissions, carbon emission monitoring data is correlated with carbon emission equipment to generate a carbon emission equipment correlation dataset; the equipment operation status of the carbon emission equipment correlation dataset is analyzed to generate carbon emission equipment operation status data.

[0041] Carbon emission volatility characteristics are extracted from the operating status data of carbon emission equipment to obtain carbon emission volatility characteristic data; based on the carbon emission volatility characteristic data, suspicious equipment is screened from the operating data of carbon emission equipment to obtain suspicious carbon emission equipment screening data.

[0042] Using the carbon emission suspicious equipment screening data, emission surge detection is performed on the carbon emission equipment operation status data to generate emission surge detection data; and the emission surge detection data is used to screen abnormal equipment in the carbon emission suspicious equipment screening data to generate carbon emission abnormal equipment data.

[0043] Equipment identification codes are used to identify and locate equipment with abnormal carbon emissions, thereby obtaining location data for high-emission equipment.

[0044] By linking dynamic anomalies in carbon emissions to specific equipment, emission information from various devices can be systematically integrated to form a carbon emission equipment-related dataset. Detailed analysis of equipment operating status reveals the carbon emission performance of each device under different operating conditions, providing foundational data for subsequent anomaly detection and enhancing the accuracy and depth of the analysis. Extracting the carbon emission volatility characteristics of equipment operating status identifies devices exhibiting abnormal fluctuations; these devices are the sources of emission anomalies. Quantitative extraction of volatility characteristics enables rapid screening of equipment status, allowing for the timely detection of potentially high-emission devices. Based on the carbon emission volatility characteristic data, after screening out suspicious devices, further emission surge detection is performed. Surge detection accurately captures abnormal increases in equipment emissions, identifying sudden, high-emission events. Combining surge detection results allows for precise location of specific abnormal devices, avoiding misjudgments of normal equipment and enhancing management accuracy. Analysis of emission surge detection data effectively filters out high-emission devices. Then, by combining equipment identification codes, abnormal equipment can be accurately located to its specific equipment number, ensuring timely handling. This process ensures the efficiency and targeting of emissions management, preventing the spread of emissions problems. Through precise equipment location and anomaly detection, substations can detect emissions anomalies at an early stage, reducing the impact of high-emission equipment on the environment and systems. Timely handling of these devices not only avoids exceeding carbon emission limits but also reduces safety risks caused by equipment failure or performance degradation. The generated high-emission equipment location data allows management to make informed decisions based on specific data. For example, maintenance, optimization, or replacement measures can be taken for high-emission equipment to further reduce carbon emission levels. Simultaneously, this data provides valuable historical data and analytical basis for future equipment management and emissions optimization.

[0045] In some embodiments, the step of formulating a carbon emission equipment optimization strategy based on carbon emission equipment-level traceability reports and carbon emission gas leakage prediction data includes:

[0046] The carbon emission equipment-level traceability report is used to identify high-emission power plant areas using a multi-scale carbon emission virtual model, generate high-emission power plant area data, deploy high-sensitivity gas sensors on the high-emission power plant area data and collect corresponding data, thereby obtaining invisible carbon emission gas data.

[0047] Using deep learning algorithms to intelligently predict leaks of invisible carbon emission gas data, thereby generating carbon emission gas leak prediction data;

[0048] Based on carbon emission gas leakage prediction data, local environmental protection insulation gas substitution is performed on the location data of high-emission equipment to generate optimization strategies for high-carbon emission equipment; based on the optimization strategies for high-carbon emission equipment, overall carbon emission load scheduling is performed on the carbon emission equipment-level traceability report to generate overall carbon emission equipment optimization strategies.

[0049] By using equipment-level carbon emission traceability reports to identify high-emission power plant areas through multi-scale carbon emission virtual models, high-emission areas can be accurately located. This precise identification helps to concentrate resources and technical efforts on emission management, ensuring that key areas are monitored and optimized, thereby significantly reducing the environmental impact of carbon emissions. Deploying highly sensitive gas sensors in high-emission power plant areas can accurately monitor and capture invisible carbon emission gases (such as methane and nitrogen oxides) that are usually difficult to detect. This process not only improves the comprehensiveness of emission detection but also better captures leaks and sudden emission events, enhancing the real-time nature and effectiveness of carbon emission management. Using deep learning algorithms to intelligently predict leaks from invisible carbon emission gas data allows for real-time prediction of the likelihood of gas leaks and timely warnings. This predictive capability enables substations and power plants to identify potential leaks in advance and take necessary control measures to avoid carbon emission exceedances or environmental pollution incidents. Based on carbon emission gas leak prediction data, environmentally friendly insulating gases can be replaced with those used in high-emission equipment to optimize its carbon emissions. This strategy helps reduce the energy consumption of carbon-emitting equipment and lower its emission levels. By optimizing equipment through intelligent systems, substations can improve energy efficiency, reduce their carbon footprint, and promote green operations.

[0050] In some embodiments, the step of using deep learning algorithms to intelligently predict leaks of invisible carbon emission gas data, thereby generating carbon emission gas leak prediction data, includes:

[0051] Concentration change trends are extracted from invisible carbon emission gas data to obtain concentration change trend data; spatial feature analysis is then performed on the invisible carbon emission gas data using the concentration change trend data to generate gas spatial feature data.

[0052] The concentration change trend data and gas spatial characteristic data are divided into datasets to generate a model training set and a model test set; the long short-term memory neural network algorithm is used to train the model on the model training set to generate a pre-model for predicting emission gas leakage.

[0053] The emission gas leakage prediction model is generated by iterating and optimizing the pre-model through the model test set; invisible carbon emission gas data is then imported into the emission gas leakage prediction model for intelligent leakage prediction, thereby generating carbon emission gas leakage prediction data.

[0054] By extracting concentration change trends from invisible carbon emission data, the time-series changes in gas emissions can be analyzed, revealing the periodicity, suddenness, or abnormal fluctuations of gas leaks. This provides dynamic time-series information for intelligent leak prediction, ensuring that the prediction model can make accurate judgments based on real-time data. Analyzing the spatial characteristics of the gas allows for the identification of the location of gas leak sources and their spatial distribution characteristics. Spatial feature data provides a basis for understanding the diffusion model and spatial relationships of emitted gases, further enhancing the comprehensiveness and accuracy of the prediction model. Dividing the concentration change trend data and gas spatial feature data into separate datasets ensures the sufficiency and diversity of model training. The resulting training and test sets guarantee the independence of training and testing, improving the model's generalization ability and enabling it to provide accurate predictions for gas leaks under different conditions. The advantage of LSTM neural networks lies in their ability to handle data with long-term dependencies, making them suitable for processing time-series data. LSTM models can effectively capture the long-term dependencies of gas concentration changes, accurately identifying potential trends and patterns of gas leaks. Through validation and feedback on the model test set, the LSTM model undergoes iterative optimization, improving prediction accuracy and reducing errors. This allows for continuous improvement of the emission gas leak prediction model, adapting to more complex real-world situations and continuously enhancing its predictive performance. Using the trained and optimized LSTM model for intelligent leak prediction of invisible carbon emission gas data, high-risk areas and time periods for gas leaks can be identified in advance. This provides early warnings to relevant departments, enabling them to take effective measures to prevent potential environmental pollution or accidents. Through continuous prediction, substations and power plants can achieve real-time monitoring of carbon emission gases, promptly detect gas leak events, and enhance the flexibility and accuracy of emergency response.

[0055] In some embodiments, the step of performing carbon emission closed-loop feedback on a multi-scale carbon emission virtual model based on a carbon emission equipment optimization strategy to execute carbon emission monitoring and management operations at the substation includes:

[0056] Based on the optimization strategies for high-carbon emission equipment and the overall carbon emission equipment optimization strategy, a comprehensive optimization strategy for carbon emission equipment is generated. The comprehensive optimization strategy for carbon emission equipment is then used to perform closed-loop feedback on a multi-scale carbon emission virtual model to generate closed-loop feedback data for carbon emission optimization.

[0057] Based on the preset carbon emission evaluation indicators, a comprehensive carbon emission evaluation is conducted on the closed-loop feedback data of carbon emission optimization to carry out carbon emission monitoring and management operations at the substation.

[0058] By integrating optimization strategies for high-carbon-emission equipment with overall carbon-emission equipment optimization strategies, substations can be provided with more comprehensive and systematic optimization solutions. This integration combines optimization strategies at different levels (such as equipment, system, and regional levels) to ensure carbon emissions are optimized from multiple perspectives, maximizing emissions reduction. Through closed-loop feedback using comprehensive carbon emission equipment optimization strategies, the effectiveness of optimization measures can be monitored and fed back in real time. If carbon emissions fail to meet expected targets, the system can automatically adjust strategies based on feedback data, continuously optimizing carbon emission control measures. This feedback mechanism ensures the flexibility and efficiency of carbon emission management, achieving dynamic carbon emission adjustment and management. Based on closed-loop feedback data from carbon emission optimization, substations can adjust operating parameters in real time, optimizing energy consumption and equipment operating status. The closed-loop control system automatically monitors carbon emission status and adjusts optimization strategies based on real-time data, achieving continuous optimization during carbon emission control and ensuring that emission levels are always at their optimal state. Based on the closed-loop feedback data for carbon emission optimization, the system can adjust multiple aspects (such as equipment load, energy consumption, and gas emissions) to maintain carbon emission levels within set standards. This dynamic adjustment prevents carbon emission levels from exceeding or falling short of standards during actual operation, ensuring optimal performance for the enterprise in terms of environmental protection and energy efficiency. The system comprehensively evaluates the closed-loop feedback data for carbon emission optimization based on preset carbon emission evaluation indicators, providing substations with scientific and objective carbon emission assessment results. These indicators can include emissions, emission intensity, and energy utilization efficiency, ensuring a comprehensive assessment of the carbon emission optimization effect.

[0059] Secondly, this application provides a carbon emission monitoring and management device for a substation, comprising:

[0060] The carbon emission modeling unit is used to acquire the data collection dataset from the substation and construct a multi-scale carbon emission virtual model based on the data collection dataset from the substation.

[0061] The carbon emission traceability unit is used to monitor carbon emissions based on a multi-scale carbon emission virtual model and data collected from substations, and to generate a carbon emission equipment-level traceability report.

[0062] The carbon emission optimization unit is used to formulate carbon emission equipment optimization strategies based on carbon emission equipment-level traceability reports and carbon emission gas leakage prediction data; and

[0063] The carbon emission management unit is used to perform carbon emission closed-loop feedback on the multi-scale carbon emission virtual model based on the carbon emission equipment optimization strategy, and to carry out carbon emission monitoring and management operations of the substation. Attached Figure Description

[0064] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0065] Figure 1 This is a flowchart of a carbon emission monitoring and management method for a substation according to an embodiment of this application.

[0066] Figure 2 This is a flowchart of a carbon emission monitoring and management method for a substation according to an embodiment of this application.

[0067] Figure 3 This is a flowchart of a carbon emission monitoring and management method for a substation according to an embodiment of this application.

[0068] Figure 4 This is a flowchart of a carbon emission monitoring and management method for a substation according to an embodiment of this application.

[0069] Figure 5 This is a flowchart of a carbon emission monitoring and management method for a substation according to an embodiment of this application.

[0070] Figure 6 This is a flowchart of a carbon emission monitoring and management method for a substation according to an embodiment of this application.

[0071] Figure 7 This is a flowchart of a carbon emission monitoring and management method for a substation according to an embodiment of this application.

[0072] Figure 8 This is a flowchart of a carbon emission monitoring and management method for a substation according to an embodiment of this application.

[0073] Figure 9 This is a schematic diagram of a carbon emission monitoring and management device for a substation according to an embodiment of this application. Detailed Implementation

[0074] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with the embodiments of this application. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0075] The methods for monitoring carbon emissions in substations mainly rely on manual detection and periodic inspections. These methods are inefficient, inaccurate, and cannot keep track of carbon emissions in real time. They are also difficult to cope with complex power operating environments.

[0076] With the sharp rise in labor costs and the tightening of carbon emission policies, there is an urgent need for new carbon emission monitoring models to meet the needs of substations.

[0077] In view of this, this application provides a carbon emission monitoring and management method and apparatus for substations, so as to provide a new carbon emission monitoring mode to meet the needs of substations.

[0078] First, as Figure 1 As shown, in a first aspect, this application provides a method for monitoring and managing carbon emissions from a substation, comprising the following steps:

[0079] S10. Obtain the data collection dataset from the substation and construct a multi-scale carbon emission virtual model based on the data collection dataset from the substation.

[0080] Specifically, this can involve acquiring substation deployment information data; collecting substation data from this deployment information data to obtain a substation dataset; constructing a multi-scale virtual 3D model of carbon emissions from the substation dataset to generate a multi-scale virtual carbon emission model; and collecting basic deployment information of the substation, covering its location, construction year, equipment configuration, energy consumption, and other relevant data. This data can come from the substation management system, engineering construction archives, or energy monitoring platform. Real-time data collection of substation energy consumption and equipment operating status is performed using sensors, monitoring systems, or existing data platforms. This includes data on power load, equipment operating time, and environmental conditions (such as temperature and humidity). Internet of Things (IoT) devices and smart sensor networks can be used for dynamic data collection. The collected raw data is then organized, cleaned, and integrated to form a complete substation dataset. This dataset should contain key information such as energy usage, equipment operating status, and carbon emission estimates. Based on the substation's equipment configuration, energy consumption, and operating mode, a multi-scale virtual 3D model is constructed using the information in the dataset. This model not only needs to display the equipment structure, layout, and operating status of the substation, but also needs to estimate carbon emissions based on data such as equipment usage and power consumption. The multi-scale construction approach means generating carbon emission data at different levels based on different times, spaces, and operational stages (such as equipment, region, and the substation as a whole). It integrates 3D modeling software (such as Blender and SketchUp) and carbon emission estimation models (such as emission models based on equipment power consumption and load factors). Using virtual 3D modeling and carbon emission estimation techniques, a multi-scale virtual carbon emission model is ultimately generated. This model can visualize the carbon emissions of the substation at different time and spatial scales. The model should be interactive, supporting the viewing of carbon emission contributions from different equipment or regions, and changes in carbon emissions over different time periods.

[0081] S20. Conduct carbon emission monitoring based on the multi-scale carbon emission virtual model and the data collected from the substation, and generate a carbon emission equipment-level traceability report.

[0082] Specifically, carbon emission monitoring can be performed on substation datasets collected using a multi-scale carbon emission virtual model to generate carbon emission monitoring data. Based on the multi-scale carbon emission virtual model, multi-level carbon emission contribution calculations are performed on the carbon emission monitoring data to obtain a multi-dimensional emission contribution matrix. The multi-dimensional emission contribution matrix is ​​then used to trace abnormal emissions from the carbon emission monitoring data, generating a carbon emission equipment-level traceability report. Using the multi-scale carbon emission virtual model constructed in step S10, the collected substation dataset (such as power load, equipment status, and operating time) is used to calculate and monitor the substation's carbon emissions in real time. The carbon emission estimation methods in the model are used to quantify the carbon emissions of each device or area. A carbon emission monitoring dataset is generated by calculating the carbon emissions of multiple devices and areas. This ensures that the model can dynamically monitor carbon emissions in both time and space dimensions, providing detailed data support for subsequent analysis. A multi-level contribution analysis method is used to calculate the carbon emission contribution of the monitoring data at multiple levels. Specifically, this can be divided into the following levels: calculating the carbon emission contribution of each device based on its operating load and power consumption. Based on the equipment layout within the substation, the equipment is divided into different areas, and the carbon emission contribution of each area is calculated. The total carbon emission contribution of the entire substation is then calculated based on the regional carbon emission data. The calculated carbon emission contributions at each level form a multidimensional emission contribution matrix, containing carbon emission data for equipment, areas, and the overall substation. Anomaly detection is performed on the generated multidimensional emission contribution matrix. Potential abnormal emissions are identified by comparing normal emission data with actual monitoring data. First, carbon emission monitoring data is compared with historical normal emission data to identify emissions deviating from the normal range. Based on the multidimensional emission contribution matrix, the source of abnormal emission data is analyzed to determine whether the abnormal emissions originate from specific equipment, areas, or the entire substation. A time-series analysis is performed on the abnormal data, combining the time dimension, to determine whether it is a short-term or long-term anomaly and whether it is related to equipment failure or external factors. After source tracing is completed, a detailed carbon emission equipment-level source tracing report is generated based on the abnormal emission source tracing results. The report should include: a list of equipment with detected abnormal emissions and their specific locations (e.g., specific areas or equipment within the substation); a list of the specific emissions from each abnormal emission equipment and the degree of deviation from the normal emission range. Provide the time period in which abnormal emissions occurred and analyze their causes (e.g., equipment failure, load fluctuations). Analyze the contribution of the equipment, region, and the entire substation to carbon emissions using a multidimensional emission contribution matrix. Based on the causes of the emission anomalies, propose improvement measures or equipment optimization suggestions to help achieve optimized carbon emission management.

[0083] S30. Based on the carbon emission equipment-level traceability report and carbon emission gas leakage prediction data, formulate carbon emission equipment optimization strategies;

[0084] Specifically, this can involve identifying high-emission power plant areas using a multi-scale carbon emission virtual model based on carbon emission equipment-level traceability reports, generating high-emission power plant area data; performing intelligent leakage prediction on the high-emission power plant area data to generate carbon emission gas leakage prediction data; generating local and overall carbon emission management strategies based on the carbon emission gas leakage prediction data, high-emission equipment location data, and carbon emission equipment-level traceability reports, resulting in high-emission equipment optimization strategies and overall carbon emission equipment optimization strategies; and identifying any equipment or area exceeding an emission threshold as a high-emission area. Based on carbon emission data and an emission contribution matrix, equipment or areas with concentrated emissions are identified, forming high-emission power plant areas. Detailed information about these high-emission areas is recorded, including equipment location, emission levels, equipment type, and operational status, generating high-emission power plant area data, including the identifier, emission amount, equipment, and operational status of each high-emission area. Machine learning algorithms (such as decision trees, support vector machines, neural networks, etc.) or physics-based leakage models are used to train the prediction model. Input data includes high-emission area data, historical emission data, equipment failure records, and environmental conditions. The model predicts carbon emission gas leakage trends based on existing data and identifies potential leakage risk areas. Based on the prediction model, it outputs carbon emission gas leakage prediction data, including leakage probability, leakage amount, and leakage time window information, generating carbon emission gas leakage prediction data to provide a basis for subsequent management decisions. Combining leakage prediction data, equipment-level traceability reports, and high-emission equipment location data, it generates optimization strategies for high-emission equipment and overall carbon emission management strategies. Specific implementation methods include: conducting local carbon emission management for high-emission equipment based on leakage prediction data, including adjusting equipment load, optimizing operating time, and increasing equipment maintenance frequency. For equipment with high leakage risk, it adopts stricter operating procedures or reduces emissions through equipment upgrades and replacements. Based on local management strategies, it formulates overall carbon emission management strategies for substations or power plants, including optimizing overall equipment operating modes, increasing the proportion of green energy, and introducing advanced emission control technologies. For individual high-emission equipment, it proposes specific optimization schemes, such as reducing energy consumption, optimizing equipment operating efficiency, equipment upgrades, and maintenance plans. Based on leakage prediction data, it proposes enhanced monitoring and management measures for equipment leakage points to reduce carbon emissions and gas leaks. Based on overall emissions and carbon emission forecasts, a global optimization scheme is designed, encompassing all equipment and operational aspects, with priority given to optimizing high-emission equipment. This involves introducing intelligent scheduling, optimizing load allocation, and adjusting the power plant's operation and maintenance strategies to minimize overall carbon emissions. It also includes integration with green energy sources such as wind and solar power to further reduce overall carbon emissions.

[0085] S40. Based on the carbon emission equipment optimization strategy, perform carbon emission closed-loop feedback on the multi-scale carbon emission virtual model and execute carbon emission monitoring and management operations of the substation.

[0086] Specifically, this can involve using a multi-scale carbon emission virtual model to generate carbon emission optimization closed-loop feedback data based on optimization strategies for high-carbon emission equipment and overall carbon emission equipment. The carbon emission optimization closed-loop feedback data is then used to conduct a comprehensive carbon emission evaluation based on preset carbon emission evaluation indicators to perform carbon emission monitoring and management operations at the substation. Optimization measures for high-carbon emission equipment (such as reducing equipment load, improving equipment operating efficiency, and reducing maintenance intervals) are input into the multi-scale carbon emission virtual model. Overall carbon emission management strategies for the power station (such as intelligent scheduling, optimized load allocation, and integration of green energy) are input into the virtual model for global optimization. The multi-scale carbon emission virtual model simulates the carbon emission situation after applying the optimization strategies. The model dynamically simulates equipment and areas based on these strategies, providing real-time feedback on the optimization effects and generating carbon emission optimization closed-loop feedback data. During this process, the virtual model should be able to dynamically adjust the operating status of equipment and energy efficiency optimization measures based on actual operating data and prediction results, generating continuous optimization feedback and forming a closed-loop mechanism. The model generates real-time feedback data based on actual equipment emissions, predicted carbon emissions, and the effectiveness of optimization measures. This feedback data includes carbon emission reduction, optimization effect assessment, and equipment responsiveness. As the optimization strategy is implemented, the system periodically updates model parameters, and the optimization strategy is gradually adjusted based on feedback results, ultimately achieving global optimization. A series of quantitative carbon emission evaluation indicators are set to comprehensively evaluate the effectiveness of the optimization strategy. Common evaluation indicators include: total carbon emissions, carbon emission intensity, carbon emission reduction ratio, green energy ratio, and operational efficiency. The total carbon emissions before and after optimization are calculated using mathematical formulas to obtain the emission reduction. For example: ΔC = Cbefore - Cafter; where ΔC is the carbon emission reduction, Cbefore is the total carbon emissions before optimization, and Cafter is the total carbon emissions after optimization. A comprehensive score is obtained by considering multiple evaluation indicators, such as total carbon emissions, emission reduction ratio, and green energy utilization rate. For example, a weighted average method can be used: S = w1 × Reduction + w2 × Efficiency + w3 × Green; where S is the comprehensive evaluation score, and w1, w2, and w3 are the weights of each indicator, representing the relative importance of emission reduction, energy efficiency, and green energy, respectively. The evaluation result will output a comprehensive score to measure the overall effectiveness of the current optimization strategy. The comprehensive evaluation can be divided into several levels, such as excellent, good, average, and poor, providing a basis for subsequent optimization directions. Based on the carbon emission optimization closed-loop feedback data and the comprehensive evaluation results, the carbon emission monitoring and management operations of the substation are adjusted. Specific tasks include: dynamically scheduling equipment and adjusting operating loads based on the optimized carbon emission situation to ensure that carbon emissions are minimized; and arranging equipment maintenance and inspections based on the analysis results of equipment status in the feedback data to reduce carbon emission increases caused by failures and decreased operating efficiency.Regularly generate carbon emission reports, monitor the emission dynamics of substations, and provide data support for environmental supervision.

[0087] This application, through the collection of deployment information data from substations and the construction of a multi-scale virtual 3D model of carbon emissions, can accurately reflect the operating status and environment of substations, providing a high-precision data foundation to support subsequent carbon emission monitoring and analysis. This virtual model helps to assess carbon emissions from multiple angles and levels, enhancing the comprehensiveness and precision of monitoring. By monitoring carbon emissions through the multi-scale virtual model and tracing sources based on emission contribution calculations, the source of abnormal emissions can be accurately located, and specific emitting equipment and processes can be identified. This provides an effective means for timely detection of emission problems, promotes precise carbon emission management, and helps to take targeted measures to reduce emissions. By identifying high-emission areas and performing intelligent leak prediction, potential carbon emission risks can be detected in advance, preventing problems such as gas leaks. Furthermore, carbon emission equipment optimization strategies and overall optimization strategies generated based on leak prediction data can improve equipment operating efficiency, reduce unnecessary carbon emissions, and optimize overall carbon emission management. Through closed-loop feedback of carbon emissions, the effectiveness of carbon emission optimization measures can be monitored in real time, and a comprehensive evaluation can be conducted based on preset evaluation indicators. This feedback mechanism ensures dynamic adjustment and continuous improvement of carbon emission management, enabling substations to continuously optimize their carbon emission control strategies and achieve long-term, stable low-carbon operation goals. Therefore, this invention improves the overall intelligence and systematization level of substation carbon emission management through multi-scale carbon emission virtual models, intelligent source tracing and prediction, and closed-loop feedback of optimization strategies.

[0088] In conjunction with the first aspect, such as Figure 2 As shown, in some embodiments provided in this application, obtaining the data collection dataset from the substation and constructing a multi-scale carbon emission virtual model based on the data collection dataset from the substation includes:

[0089] S101. Obtain substation deployment information data;

[0090] S102. Perform substation geographic topology analysis on the deployment information data of the substation to generate substation geographic topology data. Based on the substation geographic topology data, use multi-source sensors to collect substation data to obtain the substation data collection dataset.

[0091] S103. Perform data preprocessing on the substation data acquisition dataset to generate a standard substation data acquisition dataset;

[0092] S104. Using digital twin technology, a multi-scale virtual three-dimensional model of carbon emissions is constructed from data collected from standard substations, thereby generating a multi-scale virtual carbon emissions model.

[0093] Substation data acquisition includes energy consumption, gas, and environmental data collection. Data preprocessing includes data cleaning, noise reduction, missing value imputation, and standardization. Data acquisition from multiple sensors (including energy consumption, gas, and environmental data) allows for a more comprehensive capture of various substation information. This not only enhances data richness but also ensures accuracy across multiple dimensions, providing a more reliable foundation for subsequent analysis. Preprocessing substation data (such as data cleaning, noise reduction, missing value imputation, and standardization) effectively removes invalid information and reduces data noise, thereby improving data quality and consistency and facilitating subsequent in-depth analysis and modeling. Through digital twin technology and the construction of multi-scale virtual 3D carbon emission models, dynamic visualization and refined management of substation carbon emissions can be achieved. Multi-scale models can perform multi-dimensional analysis of substation carbon emissions from different spatial and temporal scales, supporting more precise carbon emission control and optimization, and further promoting the achievement of green electricity and energy conservation and emission reduction goals. This process, through the construction of virtual models, not only enables real-time monitoring of substation carbon emissions but also provides data-driven decision support, helping managers adopt more rational energy-saving measures and optimize substation operating efficiency. The combination of digital twins and virtual 3D models allows for intelligent management of substations. Utilizing big data and AI analysis models, real-time predictions of substation energy consumption and environmental changes can be made, automatically adjusting and optimizing operations to achieve refined management of energy consumption and precise control of carbon emissions.

[0094] Specifically, this can be achieved by starting with the substation's design and planning documents. These documents detail key parameters such as the site selection basis, construction scale, and layout of major equipment within the substation during the initial planning phase. For example, they specify the installation location and quantity of main transformers, as well as the distribution of outgoing line bays for different voltage levels. Simultaneously, the asset management system of the power grid company is accessed to extract detailed information on the models, specifications, and commissioning times of various electrical equipment within the substation. Furthermore, Geographic Information System (GIS) data is used to accurately obtain the substation's latitude, longitude, altitude, and surrounding topography. This deployment information is then imported into professional geographic information analysis software. Utilizing its spatial analysis capabilities, the geographical adjacency relationships between different functional areas within the substation, such as the substation area, distribution area, and control room, are identified. The routing and layout of power transmission lines and communication lines within the substation are also analyzed, generating substation geographic topology data. This data presents the spatial distribution hierarchy of the substation's components, resembling a precise "substation map." Based on this generated geographic topology data, multi-source sensors are strategically deployed. In terms of energy consumption data collection, smart meters are installed on major energy-consuming equipment such as main transformers and switchgear to monitor the power consumption and energy usage of these devices in real time. For substation gas data collection, high-precision gas sensors are deployed in areas prone to generating harmful gases, such as sealed compartments of critical electrical equipment and cable trenches, to monitor the concentration of gases like sulfur hexafluoride (SF6). Regarding substation environmental data collection, temperature and humidity sensors and light sensors are installed in various corners of the substation to comprehensively perceive the environmental conditions. In this way, multiple sensors work together to continuously collect data, which is then aggregated into a substation data collection dataset. Each data record in the substation data collection dataset is checked one by one, and obviously erroneous data, such as negative power values ​​that violate physical principles, is removed. Simultaneously, garbled data caused by sensor malfunctions and communication interference is investigated to ensure the basic accuracy of the data. Filtering algorithms, such as commonly used median filtering and mean filtering, are employed to filter out high-frequency noise mixed in during the data collection process. Taking ambient temperature data as an example, if there are sudden spikes in the data due to brief electromagnetic interference, the filtering algorithm can smooth the data and restore the true temperature change trend. For data loss caused by temporary sensor outages or malfunctions, interpolation is used to fill in the missing values. For time-series data, linear interpolation is more suitable, calculating reasonable intermediate values ​​based on adjacent valid data. For non-linear data, more flexible methods such as spline interpolation can be used to ensure data continuity. Z-score standardization is employed to unify data of different physical dimensions and magnitudes into a similar numerical range. For example, energy consumption data may be large, while environmental temperature and humidity data may be relatively small. After standardization, these data features can participate more fairly in subsequent model construction and analysis, avoiding model bias caused by differences in magnitude.After completing the above processing, a standard substation data set is generated. Leveraging the powerful modeling capabilities of digital twin technology, the 3D model is constructed based on the standard substation data. First, at the basic geometric modeling level, a precise 3D framework of the substation's buildings and equipment is constructed based on the substation's geographical topology data, macroscopically presenting the overall appearance and layout of the substation. Then, physical attributes are assigned to the model, linking energy consumption data, gas emission data, etc., with corresponding equipment and areas to simulate the dynamic changes in energy consumption during equipment operation and potential carbon emission processes. At different scales, the model can macroscopically display the carbon emission contribution of the entire substation to the regional power grid, while also focusing on the microscopic details of carbon emission from individual devices, ultimately successfully generating a multi-scale carbon emission virtual model.

[0095] In conjunction with the first aspect, such as Figure 3 As shown in some embodiments provided in this application, the step of constructing a multi-scale virtual three-dimensional model of carbon emissions from data collected at a standard substation using digital twin technology, thereby generating a multi-scale virtual carbon emissions model, includes:

[0096] S1041. Design a model-level deployment strategy based on data collected from standard substations, and generate model-level deployment design data, which includes equipment-level deployment data, system-level deployment data, and regional-level deployment data.

[0097] S1042. Based on the equipment-level deployment data, a mathematical model is constructed between the single-equipment operating parameters and carbon emission characteristics of the standard substation collected data, thereby obtaining a single-equipment carbon emission characteristic model; a mathematical model of the coordinated emission characteristics is constructed based on the single-equipment carbon emission characteristic model using system-level deployment data, thereby obtaining a system-level carbon emission model.

[0098] S1043. Based on regional deployment data, construct a mathematical model of the contribution of carbon emissions from the environment surrounding the substation to the system-level carbon emission model, thereby obtaining a regional carbon emission model; based on digital twin technology, integrate and virtualize the single-equipment carbon emission characteristic model, the system-level carbon emission model, and the regional carbon emission model at the model level to generate a multi-scale carbon emission virtual model.

[0099] By designing a hierarchical deployment strategy, data acquisition and carbon emission monitoring can be effectively planned at different levels (equipment, system, region). This multi-level data deployment not only improves substation management efficiency but also enables precise control and optimization of carbon emissions at each level, making substation operation and maintenance more scientific and efficient. Mathematical modeling of the carbon emission characteristics of individual devices allows for a precise understanding of the contribution of each device's operation to carbon emissions, further optimizing equipment energy efficiency and emission levels. Simultaneously, the construction of a system-level carbon emission model enables an overall assessment of the synergistic emission effects among various devices within the substation, supporting system-level optimization and enhancing the overall carbon emission management capabilities of the substation. Constructing a mathematical model of the carbon emission contribution of the substation's surrounding environment through regionally deployed data allows for the incorporation of environmental factors (such as climate and geographical location) into carbon emission prediction and optimization. This environmentally conscious carbon emission model more comprehensively reflects actual emissions, providing strong support for developing more precise carbon reduction measures. Digital twin technology integrates and virtualizes carbon emission models at the individual device, system, and regional levels, enabling substations to dynamically monitor carbon emissions across different spatial and temporal scales. This multi-scale virtual carbon emission model not only allows for real-time analysis but also helps managers optimize and adjust emissions at different levels, enhancing the substation's green operation capabilities. Based on this multi-scale virtual carbon emission model, substations can achieve more precise carbon emission control. Through real-time data feedback and model optimization, carbon emissions can be minimized while meeting electricity demand, which is crucial for achieving green electricity and complying with carbon emission policies.

[0100] Specifically, this can be achieved by deeply analyzing data collected from standard substations, focusing on each key piece of equipment, such as main transformers, circuit breakers, and disconnectors. Basic parameters of the equipment can be identified, including model, rated power, operating time, and corresponding energy consumption and emissions data. For different types of equipment, data subsets can be divided based on their functional characteristics and operating patterns to generate equipment-level deployment data. For example, for main transformers, data on oil temperature, load rate, and power loss should be collected to prepare for accurate characterization of their carbon emission characteristics. The electrical connections and energy transmission links between various devices within the substation should be considered to integrate equipment data from a system perspective. The collaborative operation modes of different equipment combinations under different operating conditions should be analyzed to identify power interaction and start-stop coordination information between devices, which should be summarized into system-level deployment data. For example, in the incoming and outgoing lines of a substation system, multiple devices operate collaboratively; the overall energy consumption fluctuations and joint emissions should be statistically analyzed to establish systemic relationships between devices. Finally, the geographical location of the substation should be considered, taking into account external factors such as surrounding industrial layout, residential distribution, and climate conditions. This data is integrated with system-level data within the substation, such as the impact of peak electricity consumption from surrounding factories on the substation load, and the relationship between climatic factors like wind direction and temperature and the substation's gas diffusion and heat dissipation, to generate regional-level deployment data. Using equipment-level deployment data as input, appropriate mathematical tools, such as linear regression and nonlinear fitting, are selected. Equipment operating parameters, such as the speed of motors and the temperature setpoints of heating equipment, are used as independent variables, with corresponding carbon emission indicators as dependent variables. For a single high-voltage switchgear, the functional relationship between its contact opening and closing frequency, current magnitude, and carbon emissions from trace sulfur hexafluoride leakage is analyzed. Through extensive data training and optimization, a single-equipment carbon emission characteristic model is constructed, accurately reflecting the inherent laws governing the operation and carbon emissions of a single piece of equipment. Based on system-level deployment data, each single-equipment carbon emission characteristic model is used as a basic module. Considering the synergistic effects between equipment, multivariate simultaneous equations and dynamic system modeling are used to describe the joint carbon emissions of the equipment cluster under complex scenarios such as switching between different operating conditions and emergency fault response. For example, when some equipment within a substation switches operating modes for maintenance, the dynamic changes in carbon emissions across the entire substation system are simulated to construct a system-level carbon emission model. Based on regional deployment data, the system-level carbon emission model is further expanded. Diffusion models and energy balance models from environmental science are introduced to establish mathematical relationships between the heat and greenhouse gases emitted by the substation into the surrounding environment and regional environmental factors such as atmospheric flow and surface absorption. For instance, during periods of high summer temperatures, the contribution of substation heat emissions to local warming and the diffusion range and concentration changes of harmful gases under different wind conditions are simulated to obtain a regional-level carbon emission model.Leveraging the virtual-real mapping capabilities of digital twin technology, single-device carbon emission characteristic models, system-level carbon emission models, and regional carbon emission models are integrated hierarchically. Presented in a three-dimensional visualization within a virtual space, this ranges from microscopic details of carbon emissions within equipment to a macroscopic overview of regional environmental impacts. Data interaction links are established between models at different levels, enabling real-time synchronization and linkage of carbon emission data at different scales. Ultimately, a multi-scale virtual carbon emission model is generated, providing a one-stop digital platform for comprehensive monitoring, analysis, and control of substation carbon emissions.

[0101] In conjunction with the first aspect, such as Figure 4 As shown, in some embodiments provided in this application, the step of monitoring carbon emissions based on a multi-scale carbon emission virtual model and a dataset collected from a substation to generate a carbon emission equipment-level traceability report includes:

[0102] S201. Use a multi-scale carbon emission virtual model to monitor carbon emissions from the substation dataset, generate carbon emission monitoring data, define the causal structure of carbon emissions from the carbon emission monitoring data, and generate carbon emission causal structure definition data.

[0103] S202. Based on the multi-scale carbon emission virtual model, the carbon emission causal structure is defined to calculate the carbon emission monitoring data at multiple levels, thereby obtaining a multi-dimensional emission contribution matrix.

[0104] S203. Utilize the multidimensional emission contribution matrix to detect dynamic anomalies in carbon emission equipment based on carbon emission monitoring data, generate dynamic anomalies in carbon emission, locate abnormal emission equipment based on the dynamic anomalies in carbon emission monitoring data, and generate high emission equipment location data.

[0105] S204. Integrate the location data of high-emission equipment and the source traceability data of dynamic anomalies in carbon emissions to generate a carbon emission equipment-level source traceability report.

[0106] By applying multi-scale carbon emission virtual models, substations can accurately monitor carbon emissions from collected data. This not only allows for real-time tracking of carbon emissions from various devices and systems but also generates comprehensive carbon emission monitoring data, providing accurate foundational data for subsequent analysis and decision-making. Defining the causal structure of carbon emission monitoring data reveals the interrelationships between various factors within the substation, helping to identify key drivers and potential impact paths of carbon emissions, thus providing a scientific basis for optimizing emission management. Calculating multi-level carbon emission contributions based on the defined causal structure data enables analysis of carbon emission contributions from various stages, devices, and systems from multiple dimensions. This not only helps identify major emission sources but also allows for comparison and optimization of carbon emissions at different levels (such as equipment, systems, and regions), further improving management efficiency. Dynamic anomaly detection using a multi-dimensional emission contribution matrix in carbon emission monitoring data allows for timely detection of devices or stages with abnormal carbon emissions, enabling rapid measures to prevent excessive emissions. This real-time anomaly detection mechanism effectively improves the response speed of emission management and reduces the risk of exceeding carbon emission limits. Based on dynamic anomaly detection and high-emission equipment location data, specific equipment can be accurately traced to determine the source of abnormal emissions. This provides valuable information for substation operation and maintenance management, helping to quickly locate problematic equipment for repair or optimization, improving equipment operating efficiency and environmental protection levels. The generated carbon emission equipment-level traceability report can provide managers with detailed emission source analysis and abnormal emission situations. This report can support decision-makers in energy conservation and emission reduction, equipment optimization, and policy implementation, ensuring that substations improve energy efficiency while complying with environmental regulations.

[0107] Specifically, real-time updated substation datasets can be input into a multi-scale carbon emission virtual model. This model, leveraging its built-in physical laws and data correlation logic, accurately tracks carbon emissions at the substation equipment, system, and even regional levels. For example, at the equipment level, the model precisely calculates the current carbon emissions of the equipment based on real-time energy consumption data and corresponding carbon emission conversion coefficients. At the system level, it comprehensively considers the total energy consumption and total gas emissions under coordinated equipment operation to calculate the system's carbon emissions, thus generating comprehensive carbon emission monitoring data. In-depth analysis of the carbon emission monitoring data clarifies the causal relationships between different variables. Starting with changes in equipment operating parameters, it identifies which parameter changes directly or indirectly cause increases or decreases in carbon emissions. For example, an increase in the load rate of a certain substation equipment leads to increased energy consumption, which in turn increases carbon emissions. Clarifying these relationships, marking the direction of causal arrows, influence weights, and other key information, ultimately generates carbon emission causal structure definition data. Based on the powerful data processing and analysis capabilities of the multi-scale carbon emission virtual model, the pre-constructed carbon emission causal structure definition data is used to perform refined calculations on the carbon emission monitoring data. At the equipment level, the proportion of carbon emissions from a single device in the total carbon emissions of the entire substation is measured. At the system level, the carbon emission contribution of each subsystem combination is analyzed, clarifying the differences in the contribution of different systems, such as substation systems and distribution systems, to the overall carbon emissions. At the regional level, the environmental factors surrounding the substation are also considered to calculate the substation's share of carbon emission influence in its region. These contribution values ​​at different levels are organized and arranged to form a multidimensional emission contribution matrix, laying the foundation for subsequent precise location of abnormal emission sources. Carbon emission monitoring data is compared with the multidimensional emission contribution matrix. Once the carbon emission contribution at a certain moment or at a certain level deviates from the normal fluctuation range, an anomaly is identified. For example, if the daily carbon emission contribution of a certain device is stable at 5% and suddenly jumps to 15%, it indicates an abnormal carbon emission situation in the process involving that device. By connecting these anomalies, it is possible to accurately identify whether the anomaly is caused by a sudden change in equipment operating parameters, a system coordination failure, or regional environmental interference, thus generating dynamic anomaly links in carbon emissions. By investigating the dynamic anomaly links in carbon emissions, the source equipment is traced back based on the connection relationships between devices and the data flow. Leveraging the visualization advantages of multi-scale carbon emission virtual models, abnormal equipment is highlighted in a virtual 3D scene, recording detailed equipment information such as equipment number, location, and operating conditions. This generates high-emission equipment location data, facilitating rapid target identification by maintenance personnel. All data related to the dynamic anomaly process of carbon emissions is collected, including historical equipment operating data, recent maintenance records, and environmental change logs. This data is then organized and integrated according to time sequence and causal logic to detail how the abnormal emissions occurred, from which equipment they originated, and what environmental or system factors influenced them, ultimately forming a clear and comprehensive carbon emission equipment-level traceability report.

[0108] In conjunction with the first aspect, in some embodiments provided in this application, C device，i C contributes to the carbon emissions of device i system，k C is the carbon emission contribution of system k. region The contribution of carbon emissions is categorized by region, and Ctotal represents the total carbon emissions. The calculation of carbon emission contributions at multiple levels satisfies the following:

[0109]

[0110] Where: w i Let i be the operating weight of device i;

[0111] φ i Let be the emission factor of device i;

[0112] P i The operating power of device i;

[0113] N is the total number of devices;

[0114] S k Let k be the set of devices in system k;

[0115] M represents the total number of systems;

[0116] S l Let be the set of devices in system l.

[0117] By analyzing and integrating a multi-level carbon emission contribution calculation formula, the equipment-level formula can accurately calculate the carbon emission contribution of individual devices, clearly identifying which devices are the main emission sources. This provides direct data support for optimizing equipment operating parameters (such as power and load distribution), helping to reduce emissions. A transformer's weight, emission factor, and power value result in the highest equipment-level contribution, allowing for optimization of its operating strategy (such as reducing high-load periods). The system-level formula, by integrating equipment-level data, reveals the combined impact of synergistic effects or operating modes between devices on carbon emissions. It helps identify which system operating strategies or configurations contribute significantly to carbon emissions, thereby optimizing overall system operation. The high emission contribution generated by multiple devices operating collaboratively within a power system indicates that system scheduling strategies need adjustment (such as time-sharing operation or zoned scheduling). The regional-level formula separates the carbon emission contributions of devices and systems from the total contribution, quantifying the impact of external environmental factors (such as climate conditions and transmission losses) on emissions. This helps optimize the external layout and regional strategies of substations (such as improving insulation conditions or optimizing transmission efficiency). High humidity environments lead to increased leakage in insulation equipment, resulting in high regional-level emissions, indicating a need for enhanced regional meteorological monitoring or insulation performance improvements. The equipment-level, system-level, and regional-level emission data provided by the formula can be used as input, combined with real-time monitoring data, to achieve dynamic source tracing of high-emission equipment and abnormal processes. Source tracing results help to pinpoint the source of problems in real time and generate optimization suggestions, improving the timeliness of emission control. When regional emission anomalies are detected during a certain period, problematic equipment can be quickly located and emergency countermeasures formulated through equipment-level and system-level matrices. The hierarchical calculation formula simplifies the processing flow of complex emission data, achieving step-by-step analysis and optimization from single equipment to system to region. Data at different levels can flexibly support specific equipment optimization, system strategy improvement, or regional planning. Equipment-level analysis reveals excessively high emissions from older equipment, prioritizing upgrades; system-level analysis adjusts the operational priority of high-efficiency equipment. The weight parameters in the formula (such as operating weights and emission factors) and the emission contribution matrix can be dynamically adjusted to adapt to different operating conditions and external environments. Combining causal inference algorithms and deep learning models can further uncover complex relationships, supporting intelligent emission optimization and prediction. By training emission causal models using historical data and adjusting the weighting parameters in the formulas, dynamic predictions of future emission trends can be achieved. Regional-level calculation results can reflect the specific impact of invisible carbon emissions (such as SF6 leaks), providing a quantitative basis for promoting environmentally friendly alternative technologies (such as C4-FN gas). System-level data supports energy conservation and emission reduction by optimizing overall operating strategies (such as activating low-emission equipment during low-load periods). For high-emission regional data, priority can be given to the application of environmentally friendly insulation materials or the development of highly sensitive monitoring equipment.When using conventional formulas for calculating multi-level carbon emission contributions, equipment-level, system-level, and regional-level carbon emission contribution values ​​can be obtained. By applying the multi-level carbon emission contribution calculation formula provided by this invention, the equipment-level, system-level, and regional-level carbon emission contribution values ​​can be calculated more accurately, and the three contribution values ​​can be represented in a matrix to obtain the final multi-dimensional emission contribution matrix. This formula, through hierarchical analysis at the equipment, system, and regional levels, can accurately locate carbon emission sources, quantitatively assess synergistic effects, support intelligent source tracing management, and global optimization. This approach provides strong data support for carbon emission monitoring and management in substations, while promoting the application of intelligent and green technologies, ultimately achieving the goal of carbon reduction and efficiency improvement.

[0118] In conjunction with the first aspect, such as Figure 5 As shown, in some embodiments provided in this application, the step of using a multidimensional emission contribution matrix to detect dynamic anomalies in carbon emission monitoring data, generating dynamic anomalies in carbon emission monitoring data, and locating abnormal emission equipment based on these dynamic anomalies to generate high-emission equipment location data includes:

[0119] S2031. Use a multidimensional emission contribution matrix to detect dynamic anomalies in carbon emission equipment based on carbon emission monitoring data, and generate dynamic anomalies in carbon emission.

[0120] S2032. Based on the dynamic anomalies in carbon emissions, correlate carbon emission monitoring data with carbon emission equipment to generate a carbon emission equipment correlation dataset; analyze the equipment operation status of the carbon emission equipment correlation dataset to generate carbon emission equipment operation status data.

[0121] S2033. Extract carbon emission fluctuation characteristics from the carbon emission equipment operation status data to obtain carbon emission fluctuation characteristic data; based on the carbon emission fluctuation characteristic data, screen suspicious equipment from the carbon emission equipment operation data to obtain suspicious carbon emission equipment screening data.

[0122] S2034. Use the carbon emission suspected equipment screening data to detect emission surges in the operating status data of carbon emission equipment, and generate emission surge detection data; use the emission surge detection data to screen abnormal equipment in the carbon emission suspected equipment screening data, and generate carbon emission abnormal equipment data.

[0123] S2035. Identify and locate equipment with abnormal carbon emission data using equipment identification codes to obtain high-emission equipment location data.

[0124] By linking dynamic anomalies in carbon emissions to specific equipment, emission information from various devices can be systematically integrated to form a carbon emission equipment-related dataset. Detailed analysis of equipment operating status reveals the carbon emission performance of each device under different operating conditions, providing foundational data for subsequent anomaly detection and enhancing the accuracy and depth of the analysis. Extracting the carbon emission volatility characteristics of equipment operating status identifies devices exhibiting abnormal fluctuations; these devices are the sources of emission anomalies. Quantitative extraction of volatility characteristics enables rapid screening of equipment status, allowing for the timely detection of potentially high-emission devices. Based on the carbon emission volatility characteristic data, after screening out suspicious devices, further emission surge detection is performed. Surge detection accurately captures abnormal increases in equipment emissions, identifying sudden, high-emission events. Combining surge detection results allows for precise location of specific abnormal devices, avoiding misjudgments of normal equipment and enhancing management accuracy. Analysis of emission surge detection data effectively filters out high-emission devices. Then, by combining equipment identification codes, abnormal equipment can be accurately located to its specific equipment number, ensuring timely handling. This process ensures the efficiency and targeting of emissions management, preventing the spread of emissions problems. Through precise equipment location and anomaly detection, substations can detect emissions anomalies at an early stage, reducing the impact of high-emission equipment on the environment and systems. Timely handling of these devices not only avoids exceeding carbon emission limits but also reduces safety risks caused by equipment failure or performance degradation. The generated high-emission equipment location data allows management to make informed decisions based on specific data. For example, maintenance, optimization, or replacement measures can be taken for high-emission equipment to further reduce carbon emission levels. Simultaneously, this data provides valuable historical data and analytical basis for future equipment management and emissions optimization.

[0125] Specifically, this can be achieved by obtaining data related to dynamic anomalies in carbon emissions and cross-referencing it with complete carbon emission monitoring data. Based on the electrical connections, data transmission links, and physical adjacency of equipment, data from all devices associated with the anomaly is extracted and integrated to form a carbon emission equipment association dataset. For example, if abnormal fluctuations in carbon emissions are detected on a certain outgoing line, data from circuit breakers, disconnectors, current transformers, and other equipment on that line are correlated to ensure no potentially related equipment information is overlooked. For the carbon emission equipment association dataset, the various operating parameters of each device are analyzed in depth, covering indicators such as start / stop status, load rate, operating time, temperature, and pressure. Using a pre-defined normal operating range model, the current operating state of the equipment is determined, such as stable operation, overload operation, or standby, thereby generating carbon emission equipment operating status data. Taking a transformer as an example, by analyzing oil temperature and load current, it is determined whether it is in a normal heat dissipation and power transmission state. Statistical analysis and feature extraction are then performed on the carbon emission equipment operating status data. The system calculates the mean, variance, and standard deviation of carbon emission data for each device at different time scales to capture the fluctuation characteristics of carbon emission data over time. Signal processing techniques such as wavelet transform and Fourier transform are used to decompose the carbon emission data sequence, highlighting high-frequency and low-frequency fluctuation components to obtain accurate carbon emission fluctuation characteristic data. For example, for some intermittently operating equipment, these methods can be used to accurately capture the sudden changes in carbon emissions caused by each start-up and shutdown. Using the carbon emission fluctuation characteristic data as a reference, reasonable fluctuation thresholds and screening rules are set. When the carbon emission fluctuation amplitude and frequency of equipment exceed the normal range, it is marked and screened out, forming a list of suspicious carbon emission equipment. For example, the daily carbon emission fluctuation amplitude of normal equipment is within ±5%. If a device's fluctuation exceeds 10% for several consecutive days, it is included in the list of suspicious equipment. Using the suspicious carbon emission equipment screening data, the system focuses on the short-term carbon emission changes of equipment. A sliding window algorithm is used to monitor the carbon emission slope of equipment in real time over the past few minutes and hours. Once a sharp increase in the slope is detected, far exceeding the historical level for the same period, it is judged as an emission surge, generating emission surge detection data. For example, if a piece of equipment previously had a stable hourly carbon emission increase of 1 kg, but suddenly experiences an hourly increase of 10 kg, it triggers the detection mechanism. Based on the emission surge detection data, a second screening is performed on the data of equipment suspected of high carbon emissions. Equipment with abnormal fluctuations but not a sudden increase is eliminated, retaining only equipment that has truly experienced a sharp increase in carbon emissions within a short period, generating data on equipment with abnormal carbon emissions, and accurately identifying the most problematic equipment causing high emissions. Finally, for each piece of equipment in the data on equipment with abnormal carbon emissions, its unique equipment identification code is extracted, such as equipment number, QR code, RFID tag information, etc. Combined with the geographical topology model and equipment layout map of the substation, the specific physical location of the abnormal equipment within the station is determined, thus obtaining the location data of high-emission equipment.

[0126] In conjunction with the first aspect, such as Figure 6 As shown, in some embodiments provided in this application, the step of forming a carbon emission equipment optimization strategy based on carbon emission equipment-level traceability reports and carbon emission gas leakage prediction data includes:

[0127] S301. Identify high-emission power plant areas using multi-scale carbon emission virtual models through carbon emission equipment-level traceability reports, generate high-emission power plant area data, deploy high-sensitivity gas sensors on the high-emission power plant area data and collect corresponding data, thereby obtaining invisible carbon emission gas data.

[0128] S302. Utilize deep learning algorithms to perform intelligent leakage prediction on invisible carbon emission gas data, thereby generating carbon emission gas leakage prediction data.

[0129] S303. Based on the carbon emission gas leakage prediction data, the location data of high emission equipment is used to replace local equipment with environmental protection insulating gas, thereby generating an optimization strategy for high carbon emission equipment. Based on the optimization strategy for high carbon emission equipment, the carbon emission equipment-level traceability report is used to perform overall carbon emission load scheduling, thereby generating an overall carbon emission equipment optimization strategy.

[0130] By using equipment-level carbon emission traceability reports to identify high-emission power plant areas through multi-scale carbon emission virtual models, high-emission areas can be accurately located. This precise identification helps to concentrate resources and technical efforts on emission management, ensuring that key areas are monitored and optimized, thereby significantly reducing the environmental impact of carbon emissions. Deploying highly sensitive gas sensors in high-emission power plant areas can accurately monitor and capture invisible carbon emission gases (such as methane and nitrogen oxides) that are usually difficult to detect. This process not only improves the comprehensiveness of emission detection but also better captures leaks and sudden emission events, enhancing the real-time nature and effectiveness of carbon emission management. Using deep learning algorithms to intelligently predict leaks from invisible carbon emission gas data allows for real-time prediction of the likelihood of gas leaks and timely warnings. This predictive capability enables substations and power plants to identify potential leaks in advance and take necessary control measures to avoid carbon emission exceedances or environmental pollution incidents. Based on carbon emission gas leak prediction data, environmentally friendly insulating gases can be replaced with those used in high-emission equipment to optimize its carbon emissions. This strategy helps reduce the energy consumption of carbon-emitting equipment and lower its emission levels. By optimizing equipment through intelligent systems, substations can improve energy efficiency, reduce their carbon footprint, and promote green operations.

[0131] Specifically, by carefully studying the carbon emission equipment-level traceability report, which details the source equipment, abnormal processes, and corresponding propagation paths of carbon emissions, high-emission power plant areas can be accurately located based on the equipment location and system correlation information in the report, combined with the regional carbon emission heat map presented by the multi-scale carbon emission virtual model. For example, if the traceability report indicates that a certain group of substation equipment frequently exhibits high carbon emission anomalies, the model can locate the plant area where the equipment is located, mark it as a high-emission area, and generate high-emission power plant area data. Within the identified high-emission power plant area, the layout of high-sensitivity gas sensors is scientifically planned. These sensors are primarily placed at equipment sealing connections and around gas storage tanks—areas prone to gas leaks. These sensors can detect invisible high-greenhouse-effect gases such as sulfur hexafluoride (SF6). After continuous operation for a period of time, a large amount of invisible carbon emission gas data, including gas concentration and changing trends, is collected. The collected invisible carbon emission gas data is then input into a deep learning algorithm model. First, the data is preprocessed, including normalization, to bring data of different magnitudes to the same scale for easier model computation. The dataset is then divided into training, validation, and test sets. A suitable deep learning architecture, such as Long Short-Term Memory (LSTM), is selected, as it excels at processing time-series data and is well-suited for capturing the changing patterns of gas concentration over time. The model is trained using the training set to learn the relationship between gas concentration changes and potential leaks. The model parameters are then adjusted using the validation set, and finally, the model performance is evaluated using the test set. The trained model can then predict carbon emission gas leaks over a future period, generating carbon emission gas leak prediction data. Based on this carbon emission gas leak prediction data, the composition, quantity, and leakage risk of the insulating gas used in the equipment identified by the high-emission equipment location data are analyzed. Existing environmentally friendly insulating gases on the market, such as novel fluorocyanide-nitrile mixtures, are investigated. While ensuring the equipment's insulation and electrical performance, replacement plans are developed to gradually replace traditional high-emission, high-leakage gases with environmentally friendly gases, generating optimization strategies for high-carbon emission equipment. Based on existing optimization strategies for high-carbon emission equipment, and taking into account the operating status and load demand of other equipment within the substation, combined with carbon emission link information from the carbon emission equipment-level traceability report, linear programming and integer programming methods from operations research are used to re-plan the start-up and shutdown sequence of equipment and load allocation. This achieves optimized scheduling of the overall carbon emission load within the substation, minimizing the total carbon emissions of the entire substation. Ultimately, an overall carbon emission equipment optimization strategy is generated, providing clear guidelines for subsequent energy-saving and emission-reduction retrofits and daily operation and maintenance of the substation.

[0132] In conjunction with the first aspect, such as Figure 7 As shown, in some embodiments provided in this application, the step of using deep learning algorithms to intelligently predict leaks of invisible carbon emission gas data, thereby generating carbon emission gas leak prediction data, includes:

[0133] S3021. Extract the concentration change trend from the invisible carbon emission gas data to obtain concentration change trend data; perform spatial feature analysis on the invisible carbon emission gas data using the concentration change trend data to generate gas spatial feature data.

[0134] S3022. Divide the concentration change trend data and gas spatial characteristic data into datasets to generate a model training set and a model test set; use the long short-term memory neural network algorithm to train the model on the model training set to generate a pre-model for predicting gas leakage.

[0135] S3023. The emission gas leakage prediction pre-model is optimized and iterated through the model test set to generate the emission gas leakage prediction model; the invisible carbon emission gas data is imported into the emission gas leakage prediction model for intelligent leakage prediction to generate carbon emission gas leakage prediction data.

[0136] By extracting concentration change trends from invisible carbon emission data, the time-series changes in gas emissions can be analyzed, revealing the periodicity, suddenness, or abnormal fluctuations of gas leaks. This provides dynamic time-series information for intelligent leak prediction, ensuring that the prediction model can make accurate judgments based on real-time data. Analyzing the spatial characteristics of the gas allows for the identification of the location of gas leak sources and their spatial distribution characteristics. Spatial feature data provides a basis for understanding the diffusion model and spatial relationships of emitted gases, further enhancing the comprehensiveness and accuracy of the prediction model. Dividing the concentration change trend data and gas spatial feature data into separate datasets ensures the sufficiency and diversity of model training. The resulting training and test sets guarantee the independence of training and testing, improving the model's generalization ability and enabling it to provide accurate predictions for gas leaks under different conditions. The advantage of LSTM neural networks lies in their ability to handle data with long-term dependencies, making them suitable for processing time-series data. LSTM models can effectively capture the long-term dependencies of gas concentration changes, accurately identifying potential trends and patterns of gas leaks. Through validation and feedback on the model test set, the LSTM model undergoes iterative optimization, improving prediction accuracy and reducing errors. This allows for continuous improvement of the emission gas leak prediction model, adapting to more complex real-world situations and continuously enhancing its predictive performance. Using the trained and optimized LSTM model for intelligent leak prediction of invisible carbon emission gas data, high-risk areas and time periods for gas leaks can be identified in advance. This provides early warnings to relevant departments, enabling them to take effective measures to prevent potential environmental pollution or accidents. Through continuous prediction, substations and power plants can achieve real-time monitoring of carbon emission gases, promptly detect gas leak events, and enhance the flexibility and accuracy of emergency response.

[0137] Specifically, by using statistical analysis methods and time series processing techniques to collect invisible carbon emission gas data, the rate of change of gas concentration can be calculated at fixed time intervals (e.g., hourly, minutely) to form concentration change trend data. For example, the difference method can be used to calculate the concentration difference between adjacent time points to observe whether the concentration is rising, falling, or remaining stable. Moving average algorithms can be used to further smooth the data, highlighting long-term concentration change trends and filtering out short-term noise interference. Using the acquired concentration change trend data, combined with the deployment location information of gas sensors within the high-emission power plant area, spatial feature analysis can be conducted. Analyzing the differences in gas concentration changes at different locations identifies areas with large concentration gradients, which often indicate gas leakage sources or diffusion paths. Using Geographic Information System (GIS) technology, the concentration change data can be mapped onto a regional map, visually presenting the spatial distribution of gas concentrations and generating spatial feature data of the gas. For example, if the gas concentration in a corner of a substation continues to rise rapidly and the surrounding concentration gradient is steep, a leak is very likely to exist there. Concentration change trend data and gas spatial characteristic data are integrated and divided into model training and test sets according to a certain ratio (commonly 70%-30%, 80%-20%, etc.). The training set is ensured to be sufficiently rich to cover various normal and abnormal gas concentration changes and spatial distributions, while the test set is used to independently evaluate model performance. For example, with 1000 sets of data, 800 sets are allocated to the training set and 200 sets to the test set. The model training set is input into a Long Short-Term Memory (LSTM) neural network algorithm framework. The special gating structure of LSTM makes it adept at capturing long-term dependencies in time-series data, which perfectly matches the characteristics of gas concentration changes over time. During training, an appropriate loss function is set, such as the mean squared error loss function, and the network weights are continuously adjusted through backpropagation, allowing the model to learn the mapping relationship between the input data and potential gas leaks, gradually generating a pre-model for predicting gas leaks. The model test set is input into the pre-model for predicting gas leaks, and the error between the prediction result and the true label on the test set is calculated. Based on the magnitude of the error, the model parameters are fine-tuned using optimization algorithms (such as stochastic gradient descent, Adam optimizer, etc.). This process is iterated repeatedly until the model's performance metrics (such as accuracy, recall, F1 score, etc.) on the test set reach a satisfactory level, at which point the emission gas leakage prediction model is obtained. The latest collected invisible carbon emission gas data is then imported into the optimized emission gas leakage prediction model. Based on previously learned concentration change patterns, spatial characteristics, and leakage relationships, the model outputs prediction information such as the probability of carbon emission gas leakage, leakage location, and leakage scale estimates for a future period. Ultimately, carbon emission gas leakage prediction data is generated, providing crucial decision-making basis for subsequent operation, maintenance, and prevention measures.

[0138] In conjunction with the first aspect, such as Figure 8As shown in some embodiments provided in this application, the step of performing carbon emission closed-loop feedback on a multi-scale carbon emission virtual model based on a carbon emission equipment optimization strategy to execute carbon emission monitoring and management operations for substations includes:

[0139] S401. Based on the optimization strategies for high-carbon emission equipment and the overall carbon emission equipment, strategy integration is performed to generate a comprehensive optimization strategy for carbon emission equipment. The comprehensive optimization strategy for carbon emission equipment is then used to perform carbon emission closed-loop feedback on a multi-scale carbon emission virtual model to generate carbon emission optimization closed-loop feedback data.

[0140] S402. Conduct a comprehensive carbon emission evaluation on the carbon emission optimization closed-loop feedback data based on the preset carbon emission evaluation indicators, so as to carry out the carbon emission monitoring and management operations of the substation.

[0141] By integrating optimization strategies for high-carbon-emission equipment with overall carbon-emission equipment optimization strategies, substations can be provided with more comprehensive and systematic optimization solutions. This integration combines optimization strategies at different levels (such as equipment, system, and regional levels) to ensure carbon emissions are optimized from multiple perspectives, maximizing emissions reduction. Through closed-loop feedback using comprehensive carbon emission equipment optimization strategies, the effectiveness of optimization measures can be monitored and fed back in real time. If carbon emissions fail to meet expected targets, the system can automatically adjust strategies based on feedback data, continuously optimizing carbon emission control measures. This feedback mechanism ensures the flexibility and efficiency of carbon emission management, achieving dynamic carbon emission adjustment and management. Based on closed-loop feedback data from carbon emission optimization, substations can adjust operating parameters in real time, optimizing energy consumption and equipment operating status. The closed-loop control system automatically monitors carbon emission status and adjusts optimization strategies based on real-time data, achieving continuous optimization during carbon emission control and ensuring that emission levels are always at their optimal state. Based on the closed-loop feedback data for carbon emission optimization, the system can adjust multiple aspects (such as equipment load, energy consumption, and gas emissions) to maintain carbon emission levels within set standards. This dynamic adjustment prevents carbon emission levels from exceeding or falling short of standards during actual operation, ensuring optimal performance for the enterprise in terms of environmental protection and energy efficiency. The system comprehensively evaluates the closed-loop feedback data for carbon emission optimization based on preset carbon emission evaluation indicators, providing substations with scientific and objective carbon emission assessment results. These indicators can include emissions, emission intensity, and energy utilization efficiency, ensuring a comprehensive assessment of the carbon emission optimization effect.

[0142] High-carbon emission equipment optimization strategies focus on the retrofitting of individual or localized high-emission equipment and gas substitution, while overall carbon emission equipment optimization strategies consider the carbon emission load scheduling of all equipment in the station. By combining these two strategies and considering factors such as equipment operation synergy and resource allocation rationality, complementary and unified solutions are developed to generate a comprehensive carbon emission equipment optimization strategy. For example, after replacing the insulating gas in some equipment with environmentally friendly gas, the start-up and shutdown times and power allocation of the equipment must be replanned during overall load scheduling to ensure synergy and maximize emission reduction. The generated comprehensive carbon emission equipment optimization strategy is then applied to a multi-scale carbon emission virtual model. This model adjusts equipment parameters and operating modes based on the strategy to simulate the carbon emission situation after implementing the optimization strategy. Various carbon emission data, such as changes in equipment energy consumption and greenhouse gas emission rates, are tracked in real time during the simulation, forming a closed-loop feedback data for carbon emission optimization. This is analogous to inputting a set of "optimization instructions" into the model and then collecting the "response data" after the model executes the instructions, thus constructing a dynamic feedback mechanism. The pre-set carbon emission assessment indicators cover multiple dimensions, including carbon emissions per unit of electricity generation, the reduction rate of total carbon emissions for the entire station, and changes in carbon emission intensity. These indicators are compared one by one with the closed-loop feedback data from carbon emission optimization, and the assessment is conducted through a combination of quantitative calculations and qualitative analysis. For example, the percentage reduction in carbon emissions per unit of electricity generation after implementing the optimization strategy is calculated, and the optimization effect is judged based on the magnitude of the reduction. If carbon emission intensity continues to decline, it indicates that the optimization strategy is on the right track; otherwise, further analysis and adjustments are needed. Based on the above comprehensive evaluation results, if the evaluation meets the standards, it means that the current optimization strategies for high-carbon emission equipment and the overall carbon emission equipment optimization strategies are feasible, and their implementation should be maintained and strengthened, with continuous collection of feedback data to ensure a continuous reduction in carbon emissions. If the evaluation does not meet the standards, the entire process must be reviewed, from strategy formulation and model simulation to data collection, to find the root cause of the problem and adjust the optimization strategy accordingly, thereby truly achieving accurate monitoring and effective management of substation carbon emissions.

[0143] Secondly, such as Figure 9 As shown, this application provides a carbon emission monitoring and management device for a substation, comprising:

[0144] The carbon emission modeling unit is used to acquire the data collection dataset from the substation and construct a multi-scale carbon emission virtual model based on the data collection dataset from the substation.

[0145] The carbon emission traceability unit is used to monitor carbon emissions based on a multi-scale carbon emission virtual model and data collected from substations, and to generate a carbon emission equipment-level traceability report.

[0146] The carbon emission optimization unit is used to formulate carbon emission equipment optimization strategies based on carbon emission equipment-level traceability reports and carbon emission gas leakage prediction data; and

[0147] The carbon emission management unit is used to perform carbon emission closed-loop feedback on the multi-scale carbon emission virtual model based on the carbon emission equipment optimization strategy, and to carry out carbon emission monitoring and management operations of the substation.

[0148] By acquiring deployment information data from substations and conducting detailed data collection through a multi-scale modeling module, an accurate multi-scale virtual 3D model of carbon emissions can be generated. This virtual model not only reflects the emissions situation in different areas within the substation but can also be dynamically adjusted according to actual conditions, supporting future expansion and optimization. The construction of this multi-scale model provides a high-quality spatial data foundation for carbon emission analysis, optimization strategy formulation, and pollution source location, offering crucial support for subsequent carbon emission monitoring and optimization. Based on the multi-scale carbon emission virtual model, the carbon emission contribution analysis module enables real-time and accurate carbon emission monitoring. Through in-depth analysis of the collected data, the generated carbon emission monitoring data helps substations comprehensively understand their emission situation and promptly identify potential emission anomalies. By calculating the carbon emission contribution of the monitoring data at multiple levels, the system can identify the carbon emission contribution ratio of each device and area. This analysis provides detailed evidence for subsequent carbon emission source tracing and helps quickly locate high-emission sources. Through a multi-dimensional emission contribution matrix, the system can effectively trace abnormal emissions, providing managers with clear equipment-level source tracing reports, thus facilitating the optimization and adjustment of high-emission equipment. By combining equipment-level carbon emission traceability reports and virtual models, high-emission power plant areas within substations can be identified. This allows managers to quickly pinpoint key emission areas, concentrate resources on optimization, and reduce carbon emissions. Intelligent leak prediction technology enables the system to identify carbon emission leaks in advance, allowing for preventative measures. This predictive capability helps substations identify potential carbon emission risks early, reducing the occurrence of unforeseen events. Based on leak prediction data and equipment-level traceability reports, the system can develop optimization strategies for high-emission equipment, ensuring effective control of the substation's overall carbon emission level. The combination of local and overall optimization strategies maximizes carbon emission reduction while ensuring normal operation. Feedback from high-emission equipment optimization strategies and overall optimization strategies generates closed-loop feedback data for carbon emission optimization. This closed-loop feedback system reflects the effectiveness of optimization measures in real time, supports dynamic adjustments to management strategies, and ensures that carbon emission control remains at its optimal state. By comprehensively evaluating the closed-loop feedback data of carbon emission optimization, the system can provide substations with a quantitative and scientific carbon emission assessment system. This assessment system can not only determine the actual effect of carbon emissions but also provide data support for further optimization, ensuring the effectiveness of carbon emission control measures over a long period. Therefore, this invention improves the overall intelligence and systematization level of substation carbon emission management through multi-scale carbon emission virtual models, intelligent source tracing and prediction, and closed-loop feedback of optimization strategies.

[0149] In the description of this specification, references to the terms "one embodiment / mode," "some embodiments / modes," "example," "comprising," and "having," as well as any variations thereof, in the specification, claims, and accompanying drawings, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or devices. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not represent a sequence, nor do they limit "first," "second," and "third" to different types.

[0150] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0151] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0152] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0153] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0154] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for monitoring and managing carbon emissions from a substation, characterized in that, Includes the following steps: Obtain deployment information data for substations; Perform substation geographic topology analysis on the substation deployment information data to generate substation geographic topology data; Based on the geographical topology data of the substation, multi-source sensors are used to collect substation data to obtain the substation data collection dataset. The data acquisition dataset from the substation is preprocessed to generate a standard substation acquisition dataset. Based on data collected from standard substations, a model-level deployment strategy is designed to generate model-level deployment design data, which includes equipment-level deployment data, system-level deployment data, and regional-level deployment data. Based on equipment-level deployment data, a mathematical model is constructed between the operating parameters of a single device and its carbon emission characteristics using data collected from standard substations, thus obtaining a single-device carbon emission characteristic model; and a mathematical model of coordinated emission characteristics is constructed using system-level deployment data to obtain a system-level carbon emission model. Based on regional deployment data, a mathematical model is constructed to represent the contribution of carbon emissions from the environment surrounding the substation to the system-level carbon emission model, thereby obtaining the regional carbon emission model. Based on digital twin technology, the carbon emission characteristic model of a single device, the system-level carbon emission model, and the regional carbon emission model are integrated and virtualized at the model level to generate a multi-scale carbon emission virtual model. Carbon emissions are monitored based on multi-scale carbon emission virtual models and data collected from substations, and carbon emission equipment-level traceability reports are generated. Based on carbon emission equipment-level traceability reports and carbon emission gas leakage prediction data, a carbon emission equipment optimization strategy is formulated. Based on carbon emission equipment optimization strategies, a multi-scale carbon emission virtual model is used to perform closed-loop feedback of carbon emissions and execute carbon emission monitoring and management operations for substations.

2. The carbon emission monitoring and management method for substations as described in claim 1, characterized in that, The process of monitoring carbon emissions based on a multi-scale carbon emission virtual model and data collected from substations, and generating a carbon emission equipment-level traceability report, includes: Carbon emission monitoring is performed on the substation data set using a multi-scale carbon emission virtual model, generating carbon emission monitoring data. The carbon emission monitoring data is then used to define the causal structure of carbon emissions, generating data defining the causal structure of carbon emissions. Based on the multi-scale carbon emission virtual model, the carbon emission causal structure is defined to calculate the carbon emission contribution of carbon emission monitoring data at multiple levels, thereby obtaining a multi-dimensional emission contribution matrix. The multidimensional emission contribution matrix is ​​used to detect dynamic anomalies in carbon emission equipment based on carbon emission monitoring data, generating dynamic anomalies in carbon emission. Based on these dynamic anomalies, the abnormal emission equipment is located in the carbon emission monitoring data, generating high-emission equipment location data. By integrating the location data of high-emission equipment and the source-tracing data of dynamic anomalies in carbon emissions, a carbon emission equipment-level source-tracing report is generated.

3. The carbon emission monitoring and management method for substations as described in claim 2, characterized in that, C device，i C contributes to the carbon emissions of device i system，k C is the carbon emission contribution of system k. region For regional carbon emissions contribution, C total For total carbon emissions, the calculation of multi-level carbon emission contributions satisfies: Where: w i Let i be the operating weight of device i; φ i Let be the emission factor of device i; P i The operating power of device i; N is the total number of devices; S k Let k be the set of devices in system k; M represents the total number of systems; S l Let be the set of devices in system l.

4. The carbon emission monitoring and management method for substations as described in claim 2, characterized in that, The process of using a multidimensional emission contribution matrix to detect dynamic anomalies in carbon emission monitoring data, generating dynamic anomalies in carbon emission monitoring data, and locating high-emission equipment based on these anomalies to generate high-emission equipment location data includes: A multidimensional emission contribution matrix is ​​used to detect dynamic anomalies in carbon emission equipment based on carbon emission monitoring data, and dynamic anomalies in carbon emission are generated. Based on the dynamic anomalies in carbon emissions, carbon emission monitoring data is correlated with carbon emission equipment to generate a carbon emission equipment correlation dataset; the equipment operation status of the carbon emission equipment correlation dataset is analyzed to generate carbon emission equipment operation status data. Carbon emission volatility characteristics are extracted from the operating status data of carbon emission equipment to obtain carbon emission volatility characteristic data; based on the carbon emission volatility characteristic data, suspicious equipment is screened from the operating data of carbon emission equipment to obtain suspicious carbon emission equipment screening data. Using the carbon emission suspicious equipment screening data, emission surge detection is performed on the carbon emission equipment operation status data to generate emission surge detection data; and the emission surge detection data is used to screen abnormal equipment in the carbon emission suspicious equipment screening data to generate carbon emission abnormal equipment data. Equipment identification codes are used to identify and locate equipment with abnormal carbon emissions, thereby obtaining location data for high-emission equipment.

5. The carbon emission monitoring and management method for substations as described in claim 1, characterized in that, The carbon emission equipment optimization strategy, formulated based on carbon emission equipment-level traceability reports and carbon emission gas leakage prediction data, includes: The carbon emission equipment-level traceability report is used to identify high-emission power plant areas using a multi-scale carbon emission virtual model, generate high-emission power plant area data, deploy high-sensitivity gas sensors on the high-emission power plant area data and collect corresponding data, thereby obtaining invisible carbon emission gas data. Using deep learning algorithms to intelligently predict leaks of invisible carbon emission gas data, thereby generating carbon emission gas leak prediction data; Based on carbon emission gas leakage prediction data, local environmental protection insulation gas substitution is performed on the location data of high-emission equipment to generate optimization strategies for high-carbon emission equipment; based on the optimization strategies for high-carbon emission equipment, overall carbon emission load scheduling is performed on the carbon emission equipment-level traceability report to generate overall carbon emission equipment optimization strategies.

6. The carbon emission monitoring and management method for substations as described in claim 5, characterized in that, The method of using deep learning algorithms to intelligently predict leaks of invisible carbon emission gas data, thereby generating carbon emission gas leak prediction data, includes: Concentration change trends are extracted from invisible carbon emission gas data to obtain concentration change trend data; spatial feature analysis is then performed on the invisible carbon emission gas data using the concentration change trend data to generate gas spatial feature data. The concentration change trend data and gas spatial characteristic data are divided into datasets to generate a model training set and a model test set; the long short-term memory neural network algorithm is used to train the model on the model training set to generate a pre-model for predicting emission gas leakage. The emission gas leakage prediction model is generated by iterating and optimizing the pre-model through the model test set; invisible carbon emission gas data is then imported into the emission gas leakage prediction model for intelligent leakage prediction, thereby generating carbon emission gas leakage prediction data.

7. The carbon emission monitoring and management method for substations as described in claim 1, characterized in that, The carbon emission closed-loop feedback of the multi-scale carbon emission virtual model based on the carbon emission equipment optimization strategy, and the execution of carbon emission monitoring and management operations of the substation, include: Based on the optimization strategies for high-carbon emission equipment and the overall carbon emission equipment optimization strategy, a comprehensive optimization strategy for carbon emission equipment is generated. The comprehensive optimization strategy for carbon emission equipment is then used to perform closed-loop feedback on a multi-scale carbon emission virtual model to generate closed-loop feedback data for carbon emission optimization. Based on the preset carbon emission evaluation indicators, a comprehensive carbon emission evaluation is conducted on the closed-loop feedback data of carbon emission optimization to carry out carbon emission monitoring and management operations at the substation.

8. A carbon emission monitoring and management device for a substation, characterized in that, include: The carbon emission modeling unit acquires deployment information data for substations; Perform substation geographic topology analysis on the substation deployment information data to generate substation geographic topology data; Based on the geographical topology data of the substation, multi-source sensors are used to collect substation data to obtain the substation data collection dataset. The data acquisition dataset from the substation is preprocessed to generate a standard substation acquisition dataset. Based on data collected from standard substations, a model-level deployment strategy is designed to generate model-level deployment design data, which includes equipment-level deployment data, system-level deployment data, and regional-level deployment data. Based on equipment-level deployment data, a mathematical model is constructed between the operating parameters of individual equipment and carbon emission characteristics of standard substations to obtain a single-equipment carbon emission characteristic model. Then, a mathematical model of the coordinated emission characteristics of the single-equipment carbon emission characteristic model is constructed using system-level deployment data to obtain a system-level carbon emission model. Finally, a mathematical model is constructed based on the regional-level deployment data to assess the contribution of the system-level carbon emission model to carbon emissions from the substation's surrounding environment, resulting in a regional-level carbon emission model. Finally, using digital twin technology, the single-equipment carbon emission characteristic model, the system-level carbon emission model, and the regional-level carbon emission model are integrated and virtualized at a hierarchical level to generate a multi-scale carbon emission virtual model. The carbon emission traceability unit is used to monitor carbon emissions based on a multi-scale carbon emission virtual model and data collected from substations, and to generate a carbon emission equipment-level traceability report. The carbon emission optimization unit is used to formulate carbon emission equipment optimization strategies based on carbon emission equipment-level traceability reports and carbon emission gas leakage prediction data. as well as The carbon emission management unit is used to perform carbon emission closed-loop feedback on the multi-scale carbon emission virtual model based on the carbon emission equipment optimization strategy, and to carry out carbon emission monitoring and management operations of the substation.

Images