A carbon management cockpit system based on industrial energy data

Abstract

The present application relates to a kind of carbon management cockpit systems based on industrial energy data, which includes: multi-source heterogeneous data acquisition module, dynamic carbon flow topology reconstruction module, adaptive emission factor correction module, digital twin visualization module, predictive emission reduction decision module and carbon asset intelligent management module.The present application realizes the whole-chain carbon flow tracking from energy input to product output by constructing carbon emission flow reconstruction algorithm based on space-time graph neural network;Adaptive learning mechanism is used to dynamically correct emission factor, improve carbon accounting accuracy;Based on digital twin technology, multi-dimensional visualization cockpit is constructed, real-time monitoring, early warning and prediction of carbon emissions are realized;Through the predictive emission reduction decision engine integrated with reinforcement learning algorithm, the optimal emission reduction strategy is automatically generated.The present application solves the technical problems of carbon emission data lag, low accounting accuracy and insufficient decision support in the prior art.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring and management technology for industrial carbon emissions, and in particular to a carbon management cockpit system based on industrial energy data. Background Technology

[0002] As global climate change becomes increasingly severe, and given that the industrial sector is a major source of carbon emissions, existing industrial carbon emission management systems suffer from the following technical deficiencies:

[0003] First, the data collection dimensions are limited, lacking full-chain traceability capabilities. Existing systems primarily focus on the statistics of energy consumption data, failing to establish the correlation between energy input, material conversion, and product output, thus unable to achieve product carbon footprint traceability from "cradle to door." Carbon emission data is often aggregated on a monthly or annual basis, which cannot meet the needs of real-time and refined management.

[0004] Second, carbon accounting methods are rigid, and emission factors are outdated. Existing systems generally use nationally or industry-standard emission factors for carbon accounting, failing to consider the impact of factors such as fluctuations in raw material quality, changes in equipment operating conditions, and adjustments to process parameters on actual emission factors. This leads to significant deviations between the calculated results and measured values. In particular, in process industries such as chemicals and steel, even minor changes in raw material composition can cause large fluctuations in carbon emission factors, making it difficult to guarantee the accuracy of calculations using fixed emission factors.

[0005] Third, the visualization is static and lacks spatiotemporal correlation analysis. Existing cockpit systems mostly display historical carbon emission data using static charts, lacking three-dimensional visualization capabilities based on digital twin technology. This makes it impossible to intuitively show the distribution characteristics and temporal evolution patterns of carbon emissions in physical space. Managers find it difficult to quickly locate high-emission areas and cannot conduct "hypothesis-verification" scenario analysis.

[0006] Fourth, decision support is reactive and lacks predictive optimization capabilities. Existing systems mainly provide post-event statistics and early warning functions, failing to proactively generate forward-looking emission reduction strategies based on multi-dimensional information such as production plans, energy markets, and carbon markets. Emission reduction decisions rely heavily on human experience and lack the support of quantitative optimization models, making it difficult to achieve synergistic optimization of production costs and carbon emissions.

[0007] Fifth, carbon asset management is rudimentary and lacks dynamic optimization mechanisms. Existing systems primarily manage carbon allowances using an annual accounting model, failing to monitor allowance usage progress and surplus / deficit status in real time. Carbon trading strategies lack scientific forecasting of carbon price trends, easily leading to allowance shortages or idle assets, increasing compliance costs for enterprises.

[0008] To address the shortcomings of existing technologies, this invention provides a carbon management cockpit system based on industrial energy data. By constructing a dynamic carbon flow topology network, adaptive emission factor correction, digital twin visualization, predictive emission reduction decision-making, and intelligent carbon asset management, it achieves accurate monitoring, intelligent analysis, and optimized decision-making for industrial carbon emissions. Summary of the Invention

[0009] The main objective of this invention is to overcome the technical shortcomings of existing industrial carbon management systems, such as limited data collection dimensions, insufficient calculation accuracy, and passive decision support. Instead, it provides a new carbon management dashboard system based on industrial energy data. The technical problem to be solved is to achieve full-chain carbon flow tracking and accurate calculation from energy input to product output, making it more practical and having industrial application value.

[0010] Another objective of this invention is to provide a carbon management cockpit system based on industrial energy data. The technical problem to be solved is to establish an adaptive emission factor correction mechanism that dynamically adjusts the emission factor according to real-time monitoring data, thereby improving the accuracy and reliability of carbon accounting and making it more suitable for practical use.

[0011] Another objective of this invention is to provide a carbon management cockpit system based on industrial energy data. The technical problem to be solved is to construct a multi-dimensional visualization cockpit based on digital twin technology to realize real-time monitoring, spatial positioning and trend prediction of carbon emissions, improve the intuitiveness and convenience of management, and thus be more suitable for practical use.

[0012] Another objective of this invention is to provide a carbon management cockpit system based on industrial energy data. The technical problem to be solved is to establish a predictive emission reduction decision-making mechanism, generate multi-objective optimized emission reduction strategies based on deep reinforcement learning algorithms, and realize the transformation from passive response to active optimization, thereby making it more suitable for practical use.

[0013] The objective of this invention and the technical problem it solves are achieved through the following technical solution. A carbon management cockpit system based on industrial energy data, according to this invention, includes:

[0014] The multi-source heterogeneous data acquisition module is used to collect energy consumption data, production process data, material flow data and environmental monitoring data in real time through IoT sensors, enterprise resource planning systems, manufacturing execution systems and distributed control systems.

[0015] The dynamic carbon flow topology reconstruction module is used to construct an energy-material-carbon emission correlation topology network based on the collected data using a spatiotemporal graph neural network algorithm, so as to realize the full-chain carbon flow tracking and traceability from primary energy input to end product output;

[0016] The adaptive emission factor correction module is used to dynamically correct the emission factor based on the deviation between real-time monitoring data and the standard emission factor using a sliding window adaptive learning algorithm, and to establish an emission factor uncertainty quantification model.

[0017] The digital twin visualization module is used to build a virtual carbon management cockpit that maps to the actual physical space. It uses 3D visualization technology to display the spatiotemporal distribution of carbon emissions, carbon flow Sankey diagram, equipment-level carbon emission heat map, and carbon emission trend prediction curve.

[0018] The predictive emission reduction decision module is used to generate multi-objective optimized predictive emission reduction strategies based on deep reinforcement learning algorithms, combined with production plans, energy prices, carbon market prices and environmental constraints.

[0019] The intelligent carbon asset management module is used to calculate carbon quota surpluses and shortages in real time, predict carbon price trends, optimize carbon trading strategies, and automatically generate carbon emission reports and verification documents.

[0020] The objectives of this invention and the technical problems it addresses can be further achieved by the following technical measures.

[0021] The aforementioned carbon management cockpit system based on industrial energy data, wherein the dynamic carbon flow topology reconfiguration module includes:

[0022] The energy-material relationship graph construction unit is used to establish a directed graph relationship between energy consumption nodes and material conversion nodes based on the process flow.

[0023] The spatiotemporal feature extraction unit is used to extract the spatial dependence features of carbon emissions using a graph convolutional network and to extract the time series features of carbon emissions using a long short-term memory network.

[0024] The carbon flow allocation calculation unit is used to calculate the carbon emission allocation ratio of each process step based on the network flow algorithm, so as to realize product-level carbon footprint traceability.

[0025] The aforementioned carbon management cockpit system based on industrial energy data, wherein the adaptive emission factor correction module includes:

[0026] The real-time deviation monitoring unit is used to compare the measured carbon emission concentration with the calculated value based on the standard emission factor to identify systematic deviations.

[0027] The factor dynamic correction unit is used to dynamically adjust the emission factor based on raw material quality fluctuations, equipment aging and operating condition changes using an exponentially weighted moving average algorithm.

[0028] The uncertainty quantification unit is used to establish a probability distribution model of emission factors based on a Bayesian neural network and output the uncertainty range of carbon accounting results.

[0029] The aforementioned carbon management cockpit system based on industrial energy data, wherein the digital twin visualization module includes:

[0030] The 3D scene construction unit is used to build a detailed 3D model of the factory, production line, and energy-consuming equipment based on factory CAD drawings and GIS data.

[0031] The real-time data mapping unit is used to map real-time carbon emission data from the physical space to the virtual space via the OPC UA protocol, achieving millisecond-level synchronization.

[0032] The multi-dimensional display unit provides multi-level drill-down analysis views at the group, factory, workshop, production line, and equipment levels, as well as correlation analysis views between carbon emissions and output, energy consumption, and quality.

[0033] The aforementioned carbon management cockpit system based on industrial energy data, wherein the predictive emission reduction decision module includes:

[0034] The multi-objective optimization modeling unit is used to establish a constrained optimization model with the objective functions of minimizing production costs, minimizing carbon emissions, and minimizing carbon quota gaps.

[0035] A deep reinforcement learning decision unit is used to train an agent to learn optimal production scheduling and energy allocation strategies in a simulated environment using a proximal policy optimization algorithm.

[0036] The strategy simulation verification unit is used to virtually verify the generated emission reduction strategy through a digital twin model, and to evaluate the effectiveness and risks of the strategy implementation.

[0037] The aforementioned carbon management cockpit system based on industrial energy data, wherein the intelligent carbon asset management module includes:

[0038] The carbon quota accounting unit is used to calculate the quota surplus or deficit status based on a real-time comparison between actual carbon emissions and allocated quotas.

[0039] The carbon price forecasting unit is used to predict carbon market price trends by combining attention-based time series forecasting models with policy signals, market supply and demand, and macroeconomic indicators.

[0040] The trading optimization unit is used to optimize the timing and quantity of carbon allowance trading based on a stochastic programming model, thereby maximizing the value of carbon assets.

[0041] The aforementioned carbon management cockpit system based on industrial energy data, wherein the system further includes:

[0042] The blockchain evidence storage module is used to perform hash calculations on raw carbon emission data, accounting process data, and report documents and then store them on the blockchain to ensure that the data is tamper-proof and traceable.

[0043] Edge computing nodes are used to deploy lightweight computing units on-site in factories to enable local data preprocessing and real-time response, reducing cloud transmission latency.

[0044] The aforementioned carbon management dashboard system based on industrial energy data, wherein the blockchain evidence storage module adopts a consortium blockchain architecture, includes:

[0045] The data fingerprint generation unit is used to perform SHA-256 hash operations on carbon emission data to generate data fingerprints.

[0046] Smart contract units are used to automatically execute the on-chain, query, and cross-institutional sharing of carbon emission data access control.

[0047] Cross-chain interoperability unit is used to enable data interoperability with government regulatory chains and carbon trading platform chains.

[0048] The objective of this invention and the technical problem it solves are further achieved by the following technical solution. A carbon management method based on industrial energy data, according to this invention, using the aforementioned system, includes the following steps:

[0049] S1: Real-time acquisition of energy consumption, production operation and environmental monitoring data through multi-source heterogeneous data acquisition modules;

[0050] S2: Construct an energy-material-carbon emission correlation network using the dynamic carbon flow topology reconstruction module, and calculate the carbon emissions at each stage;

[0051] S3: Based on the adaptive emission factor correction module, the emission factor is dynamically corrected according to the measured data to improve the accuracy of the calculation;

[0052] S4: Through the digital twin visualization module, carbon emission status and early warning information are displayed in real time in the virtual cockpit;

[0053] S5: Utilize the predictive emission reduction decision module to generate the optimal emission reduction strategy for the future production cycle and push it to the execution system;

[0054] S6: Through the carbon asset intelligent management module, monitor the carbon quota status in real time and optimize carbon trading strategies.

[0055] The objectives of this invention and the technical problems it addresses can be further achieved by the following technical measures.

[0056] The aforementioned carbon management method based on industrial energy data, wherein the dynamic carbon flow topology reconstruction in step S2 includes:

[0057] Construct a directed graph G=(V,E) containing energy input nodes, process conversion nodes, and product output nodes;

[0058] Based on the principles of material balance and energy conservation, a carbon flow transfer equation between nodes is established;

[0059] By using graph neural networks to learn the carbon emission transfer weights between nodes, the precise allocation of carbon flows in complex process networks can be achieved.

[0060] Compared with the prior art, the present invention has significant advantages and beneficial effects. As can be seen from the above technical solution, in order to achieve the aforementioned objectives, the main technical contents of the present invention are as follows:

[0061] This invention proposes a carbon management cockpit system based on industrial energy data. Through the collaborative work of modules such as multi-source heterogeneous data acquisition, dynamic carbon flow topology reconstruction, adaptive emission factor correction, digital twin visualization, predictive emission reduction decision-making, and intelligent carbon asset management, it achieves accurate monitoring, intelligent analysis, and optimized decision-making of industrial carbon emissions.

[0062] As described above, this invention solves the technical problems of isolated carbon emission data and lack of correlation analysis in existing technologies by constructing a carbon emission flow reconstruction algorithm based on spatiotemporal graph neural networks; it solves the technical problem of the disconnect between standard emission factors and actual operating conditions by using an adaptive emission factor correction mechanism; it solves the technical problem of unintuitive perception of carbon emission status by using digital twin visualization technology; it solves the technical problem of emission reduction strategy formulation relying on human experience by using a predictive emission reduction decision engine; and it solves the technical problems of extensive carbon quota management and lack of optimized trading strategies by using intelligent carbon asset management.

[0063] By employing the above technical solution, the carbon management cockpit system based on industrial energy data of the present invention has at least the following advantages:

[0064] 1. It enables full-chain carbon flow tracking from primary energy input to end-product output, supporting accurate calculation and traceability of product-level carbon footprint;

[0065] 2. By dynamically correcting emission factors through an adaptive learning mechanism, the accuracy of carbon accounting is improved by more than 20%, meeting the accuracy requirements of carbon verification;

[0066] 3. Construct a multi-dimensional visualization dashboard based on digital twin technology to support multi-level drill-down analysis from the group level to the equipment level, thereby improving management efficiency;

[0067] 4. Through the predictive emission reduction decision-making module, the synergistic optimization of production costs and carbon emissions can be achieved, which is expected to reduce overall operating costs by 5%-10%;

[0068] 5. Establish an intelligent carbon asset management system, optimize carbon trading strategies, reduce compliance costs, and enhance the value of carbon assets.

[0069] In summary, the unique carbon management cockpit system based on industrial energy data of this invention achieves digitalization, intelligence, and precision in industrial carbon management through technological innovation. It possesses numerous advantages and practical value, and is truly innovative as no similar designs have been publicly disclosed or used in the same technology. It represents a significant improvement in both system architecture and functional implementation, demonstrating substantial technological progress and producing user-friendly and practical results. Furthermore, it offers superior performance compared to existing industrial carbon management systems, making it more suitable for practical application and possessing broad industrial applicability. It is indeed a novel, progressive, and practical new design.

[0070] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0071] The specific embodiments and system architecture of the present invention are given in detail in the following examples and accompanying drawings. Attached Figure Description

[0072] Figure 1 This is a flowchart of the dynamic carbon flow topology reconstruction module of the present invention. Detailed Implementation

[0073] To further illustrate the technical means and effects adopted by the present invention to achieve the intended purpose, the following detailed description, in conjunction with the accompanying drawings and preferred embodiments, describes the specific implementation, system architecture, functional modules, and effects of a carbon management cockpit system based on industrial energy data proposed according to the present invention.

[0074] Example 1:

[0075] The carbon management cockpit system based on industrial energy data in a preferred embodiment of the present invention adopts a layered design with cloud-edge-device collaboration, mainly including the following layers:

[0076] Sensing Layer (End): This layer comprises various sensors and smart instruments deployed on-site in the factory, including electricity meters, gas meters, steam flow meters, temperature sensors, pressure sensors, and flue gas analyzers, used to collect real-time energy consumption data, process parameters, and environmental monitoring data. Furthermore, it interfaces with existing enterprise systems such as ERP, MES, and DCS via industrial communication protocols like OPC UA, Modbus, and MQTT to obtain management data such as production plans, bills of materials, and equipment status.

[0077] Edge Layer (Edge): Deploy edge computing gateways on-site at the factory, incorporating lightweight data preprocessing algorithms to clean, compress, aggregate, and store data locally. Edge nodes run carbon flow computing microservices, supporting millisecond-level real-time carbon emission calculation and early warning response, reducing cloud transmission bandwidth pressure and ensuring local autonomy in the event of network outages.

[0078] Platform Layer (Cloud): Data centers deployed in the cloud, including data lakes, algorithm engines, business middleware, and application services. The data lake stores massive amounts of historical and real-time data streams; the algorithm engine integrates advanced algorithms such as spatiotemporal graph neural networks and deep reinforcement learning; the business middleware encapsulates general service capabilities such as carbon accounting, carbon analysis, and carbon prediction; and application services provide dashboard access interfaces for web and mobile devices.

[0079] Application Layer: Provides customized applications for different user roles, including: Group Management Dashboard (for senior executives, providing a global perspective), Factory Operations Dashboard (for factory managers, providing production optimization suggestions), Workshop Monitoring Dashboard (for workshop supervisors, providing real-time alerts), and Carbon Asset Management Platform (for carbon management personnel, providing trading decision support).

[0080] Example 2:

[0081] Please see Figure 1 As shown, the dynamic carbon flow topology reconstruction module is one of the core innovations of this invention, and its workflow is as follows:

[0082] Step S201: Construct an energy-material relationship graph. Based on the plant's process flow diagram (PFD) and piping and instrumentation diagram (P&ID), identify energy input nodes (such as grid connection points, natural gas pressure regulating stations), process conversion nodes (such as blast furnaces, converters, and heating furnaces), material storage nodes (such as raw material warehouses and finished product warehouses), and product output nodes. Establish directed connections between these nodes to form an initial topology graph G=(V,E), where V is the set of nodes and E is the set of edges.

[0083] Step S202: Collect multi-dimensional data. Collect energy consumption data (electricity, coal, gas, oil, steam), material flow data (input, output, inventory), process parameter data (temperature, pressure, speed), and quality inspection data (composition, calorific value, carbon content) of each node in real time.

[0084] Step S203: Calculate direct carbon emissions. For nodes with flue gas emission monitoring, calculate direct carbon emissions based on measured concentrations and flow rates; for nodes without monitoring, calculate indirect carbon emissions based on energy consumption and standard emission factors.

[0085] Step S204: Carbon Flow Allocation Calculation. For process nodes with multiple inputs and outputs (such as the blast furnace process in an integrated iron and steel enterprise), a carbon flow allocation algorithm based on mass balance and energy balance is adopted. Let the carbon input of node i be C_in, and the output product set be P, then the carbon allocation C_p of product p is:

[0086] C_p = C_in × (m_p × α_p) / Σ(m_j × α_j)

[0087] Where m_p is the mass flow rate of product p, and α_p is the carbon distribution coefficient of product p, which is determined by process characteristics and thermodynamic calculations.

[0088] Step S205: Graph Neural Network Learning. A spatiotemporal graph neural network (ST-GNN) model is trained using historical data to learn the carbon emission transfer patterns between nodes. The model input consists of the graph's topology and node features, and the output is the predicted carbon emission value for each node. A graph attention mechanism (GAT) is used to identify the critical paths and nodes with the greatest impact on carbon emissions.

[0089] Step S206: Product Carbon Footprint Traceability. Based on the trained model, the carbon emission composition of the entire life cycle is traced back from the final product, identifying carbon emission hotspots and providing a basis for the formulation of emission reduction strategies.

[0090] Example 3:

[0091] The workflow of the adaptive emission factor correction module is as follows:

[0092] Step S301: Establish a baseline emission factor database. Collect national industry emission factor standards, IPCC guideline recommendations, and historical measured data from enterprises to establish a baseline database that includes fuel combustion emission factors, process emission factors, and power generation emission factors.

[0093] Step S302: Real-time Deviation Monitoring. For key emission sources equipped with CEMS (Continuous Emission Monitoring System), compare the measured carbon emission concentration with the calculated value based on the baseline emission factor, and calculate the deviation rate:

[0094] δ = (C_measured - C_calculated) / C_calculated × 100%

[0095] Step S303: Deviation Cause Analysis. If the deviation rate exceeds a preset threshold (e.g., ±10%), cause analysis is triggered. Analysis dimensions include: changes in raw material quality (e.g., fluctuations in coal calorific value and carbon content), equipment operating conditions (e.g., changes in boiler load rate and combustion efficiency), and adjustments to process parameters (e.g., changes in oxygen content and flue gas temperature).

[0096] Step S304: Dynamic Factor Adjustment. The emission factor is dynamically adjusted based on recent deviations using the Exponentially Weighted Moving Average (EWMA) algorithm.

[0097] EF_new = λ × EF_measured + (1-λ) × EF_old

[0098] Wherein, EF_measured is the emission factor calculated based on measured data, EF_old is the historical emission factor, and λ is the smoothing coefficient (usually taken as 0.3-0.5).

[0099] Step S305: Uncertainty Quantification. A probability distribution model of emission factors is established based on a Bayesian neural network (BNN), outputting the confidence interval of the carbon accounting results to provide an explanation of the uncertainty in carbon verification.

[0100] Step S306: Factor Update and Synchronization. The revised emission factors are updated to the database and synchronized to relevant nodes via the blockchain network to ensure the consistency of factors across the entire system.

[0101] Example 4:

[0102] The interface layout of the digital twin visualization cockpit includes:

[0103] Global Overview Area: Using a 3D factory model as a background, this area displays the real-time carbon emission intensity of each workshop and piece of equipment via a color heatmap. Red indicates high emissions, yellow indicates medium emissions, and green indicates low emissions. Rotation, zooming, and panning are supported; clicking on specific equipment allows you to drill down for detailed information.

[0104] Carbon flow Sankey diagram: A Sankey diagram that shows the distribution of carbon flow from energy input to product output. The line width represents the amount of carbon flow, and it visually shows the main pathways and conversion efficiency of carbon emissions.

[0105] Key Indicators Area: Displays key carbon emission indicators at the group / factory level, including: total carbon emissions (tCO2e), carbon emission intensity (tCO2e / 10,000 yuan of output value), clean energy share (%), remaining carbon allowance (%), and annual emission reduction progress (%). The indicator cards use a traffic light warning mechanism, automatically flashing to alert when the standard is exceeded.

[0106] Trend Analysis Area: Displays historical and projected carbon emission trend curves, supporting switching between daily, weekly, monthly, and yearly time dimensions. The projected curves are generated based on an LSTM model and display 95% confidence intervals.

[0107] Warning Information Area: Displays real-time warning information in a scrolling manner, including: emission exceeding standards warnings, equipment malfunction warnings, quota shortage warnings, policy update reminders, etc. Supports one-click location of the anomaly.

[0108] Example 5: Predictive Emission Reduction Decisions

[0109] The architecture of the predictive emissions reduction decision-making module includes:

[0110] Environmental Modeling Layer: Constructs a digital twin model of the factory production system, including equipment energy consumption models, process emission models, and material flow models. The model is built based on a combination of physical mechanisms and data-driven methods, supporting hypothesis-verification scenario simulation.

[0111] Status Awareness Layer: Acquires real-time basic data such as current production status, energy inventory, equipment availability, and order demand, as well as external data such as carbon market prices, electricity prices, and weather forecasts.

[0112] Decision Engine Layer: The agent is trained using the Proximal Policy Optimization (PPO) algorithm. The state space includes: the operating status of each device, energy inventory level, production task queue, and carbon allowance balance; the action space includes: device start / stop decisions, production load adjustments, energy switching selection, and maintenance plan arrangements; the reward function comprehensively considers production costs, carbon emission costs, and penalties for delayed delivery.

[0113] Strategy Output Layer: The intelligent agent outputs the optimal production scheduling plan for the next 24 hours to 7 days, including: equipment operation plans for each time period, energy procurement suggestions, and emission reduction measures recommendations. The plan is displayed through a visual interface and provides a "one-click deployment" function to send instructions to the MES system for execution.

[0114] Example 6:

[0115] The workflow of the carbon asset management module is as follows:

[0116] Step S601: Real-time Quota Calculation. Based on real-time carbon emission data, dynamically calculate the annual quota usage progress and remaining quota. Let the annual quota be Q, and the cumulative emissions be E, then the quota surplus / deficit ΔQ = Q - E.

[0117] Step S602: Carbon Price Forecast. Based on the attention mechanism, the Transformer model takes historical carbon price series, policy events, macroeconomic indicators (GDP, PMI), energy prices (coal price, electricity price), and weather factors (affecting renewable energy output) as input, and outputs the carbon price forecast range for the next 1-30 days.

[0118] Step S603: Optimize Trading Strategy. Establish a stochastic programming model with the goal of minimizing expected costs, taking into account carbon price uncertainty, to optimize the timing and quantity of quota buying and selling. Constraints include: compliance period, capital budget, inventory capacity, and risk tolerance.

[0119] Step S604: Transaction Execution Monitoring. Connect to the carbon trading platform API to automatically execute trading instructions or push trading suggestions to manual confirmation. Monitor position status, profit and loss, and market risk in real time.

[0120] Step S605: Automatic Report Generation. Automatically generate carbon emission reports and verification support documents that meet the required format, supporting PDF export and blockchain storage.

[0121] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the methods and techniques disclosed above without departing from the scope of the present invention to create equivalent embodiments. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A carbon management cockpit system based on industrial energy data, characterized in that, include: The multi-source heterogeneous data acquisition module is used to collect energy consumption data, production process data, material flow data and environmental monitoring data in real time through IoT sensors, enterprise resource planning systems, manufacturing execution systems and distributed control systems. The dynamic carbon flow topology reconstruction module is used to construct an energy-material-carbon emission correlation topology network based on the collected data using a spatiotemporal graph neural network algorithm, so as to realize the full-chain carbon flow tracking and traceability from primary energy input to end product output; The adaptive emission factor correction module is used to dynamically correct the emission factor based on the deviation between real-time monitoring data and the standard emission factor using a sliding window adaptive learning algorithm, and to establish an emission factor uncertainty quantification model. The digital twin visualization module is used to build a virtual carbon management cockpit that maps to the actual physical space. It uses 3D visualization technology to display the spatiotemporal distribution of carbon emissions, carbon flow Sankey diagram, equipment-level carbon emission heat map, and carbon emission trend prediction curve. The predictive emission reduction decision module is used to generate multi-objective optimized predictive emission reduction strategies based on deep reinforcement learning algorithms, combined with production plans, energy prices, carbon market prices and environmental constraints. The intelligent carbon asset management module is used to calculate carbon quota surpluses and shortages in real time, predict carbon price trends, optimize carbon trading strategies, and automatically generate carbon emission reports and verification documents.

2. The carbon management cockpit system based on industrial energy data according to claim 1, characterized in that, The dynamic carbon flow topology reconstruction module includes: The energy and material relationship graph construction unit is used to establish a directed graph relationship between energy consumption nodes and material conversion nodes according to the process flow. The spatiotemporal feature extraction unit is used to extract the spatial dependence features of carbon emissions using a graph convolutional network and to extract the time series features of carbon emissions using a long short-term memory network. The carbon flow allocation calculation unit is used to calculate the carbon emission allocation ratio of each process step based on the network flow algorithm, so as to realize product-level carbon footprint traceability.

3. The carbon management cockpit system based on industrial energy data according to claim 2, characterized in that, The adaptive emission factor correction module includes: The real-time deviation monitoring unit is used to compare the measured carbon emission concentration with the calculated value based on the standard emission factor to identify systematic deviations. The factor dynamic correction unit is used to dynamically adjust the emission factor based on fluctuations in raw material quality, equipment aging, and changes in operating conditions using an exponentially weighted moving average algorithm. The uncertainty quantification unit is used to establish a probability distribution model of emission factors based on a Bayesian neural network and output the uncertainty range of carbon accounting results.

4. The carbon management cockpit system based on industrial energy data according to claim 1, characterized in that, The digital twin visualization module includes: The 3D scene construction unit is used to build a detailed 3D model of the factory, production line, and energy-consuming equipment based on factory CAD drawings and GIS data. The real-time data mapping unit is used to map real-time carbon emission data from the physical space to the virtual space via the OPC UA protocol, achieving millisecond-level synchronization. The multi-dimensional display unit provides multi-level drill-down analysis views at the group, factory, workshop, production line, and equipment levels, as well as correlation analysis views between carbon emissions and output, energy consumption, and quality.

5. The carbon management cockpit system based on industrial energy data according to claim 1, characterized in that, The predictive emission reduction decision module includes: The multi-objective optimization modeling unit is used to establish a constrained optimization model with the objective functions of minimizing production costs, minimizing carbon emissions, and minimizing carbon quota gaps. A deep reinforcement learning decision unit is used to train an agent to learn optimal production scheduling and energy allocation strategies in a simulated environment using a proximal policy optimization algorithm. The strategy simulation verification unit is used to virtually verify the generated emission reduction strategy through a digital twin model, and to evaluate the effectiveness and risks of the strategy implementation.

6. The carbon management cockpit system based on industrial energy data according to claim 1, characterized in that, The intelligent carbon asset management module includes: The carbon quota accounting unit is used to calculate the quota surplus or deficit status based on a real-time comparison between actual carbon emissions and allocated quotas. The carbon price forecasting unit is used in attention-based time series forecasting models to predict carbon market price trends by combining policy signals, market supply and demand, and macroeconomic indicators. The trading optimization unit is used to optimize the timing and quantity of carbon allowance trading based on a stochastic programming model, thereby maximizing the value of carbon assets.

7. The carbon management cockpit system based on industrial energy data according to any one of claims 1 to 6, characterized in that, The system also includes: The blockchain evidence storage module is used to perform hash calculations on raw carbon emission data, accounting process data, and report documents and then store them on the blockchain to ensure that the data is tamper-proof and traceable. Edge computing nodes are used to deploy lightweight computing units on-site in factories to enable local data preprocessing and real-time response, reducing cloud transmission latency.

8. The carbon management cockpit system based on industrial energy data according to claim 7, characterized in that, The blockchain evidence storage module adopts a consortium blockchain architecture and includes: The data fingerprint generation unit is used to perform SHA-256 hash operations on carbon emission data to generate data fingerprints. Smart contract units are used to automatically execute the on-chain, query, and cross-institutional sharing of carbon emission data; Cross-chain interoperability unit is used to enable data interoperability with government regulatory chains and carbon trading platform chains.

9. A method for using a carbon management system based on industrial energy data as described in any one of claims 1 to 8, characterized in that, Includes the following steps: S1: Real-time acquisition of energy consumption, production operation and environmental monitoring data through multi-source heterogeneous data acquisition modules; S2: Construct an energy-material-carbon emission correlation network using the dynamic carbon flow topology reconstruction module, and calculate the carbon emissions at each stage; S3: Based on the adaptive emission factor correction module, the emission factor is dynamically corrected according to the measured data to improve the accuracy of the calculation; S4: Through the digital twin visualization module, carbon emission status and early warning information are displayed in real time in the virtual cockpit; S5: Utilize the predictive emission reduction decision module to generate the optimal emission reduction strategy for the future production cycle and push it to the execution system; S6: Through the intelligent carbon asset management module, monitor the status of carbon quotas in real time and optimize carbon trading strategies.

10. The carbon management method based on industrial energy data according to claim 9, characterized in that, The dynamic carbon flow topology reconstruction in step S2 includes: Construct a directed graph G=(V,E) containing energy input nodes, process conversion nodes, and product output nodes; Based on the principles of material balance and energy conservation, a carbon flow transfer equation between nodes is established; By using graph neural networks to learn the carbon emission transfer weights between nodes, the precise allocation of carbon flows in complex process networks can be achieved.

Images