Power plant full life cycle carbon footprint monitoring method based on intelligent carbon tube platform

Abstract

A power plant full life cycle carbon footprint monitoring method based on an intelligent carbon tube platform relates to the field of electric digital data processing, and comprises the following steps: acquiring unit load data and fuel consumption data, and detecting desulfurization outlet gas according to a multi-layer infrared detection structure to obtain gas concentration data; integrating the gas concentration data, the unit load data and the fuel consumption data to obtain a synchronous data set; and inputting the synchronous data set into a carbon emission dynamic measurement and calculation model to obtain a carbon emission trend prediction result of a preset future time period, inputting the synchronous data set into a multi-target optimization model, and carrying out optimization calculation by taking carbon emission intensity minimization, carbon quota utilization rate optimization and power generation cost minimization as targets to obtain an optimization scheme. And generating a carbon emission monitoring result based on the carbon emission trend and the optimization scheme, and performing unit start-stop and system load adjustment control according to the carbon emission monitoring result. By implementing the method, the monitoring precision of power plant carbon footprint monitoring can be improved.

Description

Technical Field

[0001] The present application relates to the field of electrical digital data processing, and in particular to a method for monitoring the carbon footprint of a power plant throughout its life cycle based on a smart carbon tube platform. Background Art

[0002] As a key sector for carbon emissions, the power industry is increasingly important for monitoring its carbon emissions. Thermal power plants, in particular, face significant fuel consumption and high carbon emission intensity, necessitating the establishment of a systematic carbon emissions monitoring method to achieve scientific management of carbon emissions.

[0003] Currently, thermal power plants primarily monitor carbon emissions by separately collecting fuel consumption data, unit load data, and gas monitoring data. This method stores each type of data independently and performs statistical analysis on a fixed basis for carbon emissions accounting and management.

[0004] However, in actual applications, due to the independence of various types of data collection and storage, and the lack of systematic data integration and in-depth analysis mechanisms, it is difficult for power plants to grasp the changing trends of carbon emissions in a timely and accurate manner, and it is also difficult to formulate targeted optimization and control plans, which restricts the improvement of the level of refined carbon emission monitoring. Summary of the Invention

[0005] This application provides a method for monitoring the carbon footprint of a power plant throughout its life cycle based on a smart carbon tube platform, which is used to improve the monitoring accuracy of the carbon footprint of a power plant.

[0006] In the first aspect, the present application provides a method for monitoring the carbon footprint of a power plant throughout its life cycle based on a smart carbon tube platform, which is applied to the smart carbon tube platform. The method includes: obtaining unit load data and fuel consumption data, and detecting desulfurization outlet gas based on a multi-layer infrared detection structure to obtain gas concentration data; integrating the gas concentration data, unit load data and fuel consumption data to obtain a synchronized data set; inputting the synchronized data set into a carbon emission dynamic measurement model to obtain a carbon emission trend prediction result for a preset future time period. The carbon emission dynamic measurement model includes an input layer, a forget gate layer, an input gate layer, an output gate layer and an output layer; inputting the synchronized data set into a multi-objective optimization model, and performing optimization calculations with the goals of minimizing carbon emission intensity, optimizing carbon quota utilization and minimizing power generation costs to obtain an optimization scheme. The multi-objective optimization model includes initial population generation, cross-mutation, fast non-dominated sorting and congestion calculation; generating carbon emission monitoring results based on the carbon emission trend and the optimization scheme, and adjusting and controlling the start and stop of the unit and the system load according to the carbon emission monitoring results.

[0007] In the above-mentioned embodiment, a multi-layer infrared detection structure monitors desulfurization outlet gas concentration data in real time. This data is systematically integrated with unit load and fuel consumption data to form a synchronized dataset, breaking down data silos. This synchronized dataset is then fed into a dynamic carbon emissions measurement model and a multi-objective optimization model, enabling accurate prediction of carbon emission trends and optimized calculation of operational plans. Unit startup and shutdown, as well as load regulation, are performed based on the predicted results and optimized plans, establishing a closed-loop management system for data collection, analysis, optimization, and control, enabling precise monitoring and intelligent regulation of carbon emissions in power plants.

[0008] In combination with some embodiments of the first aspect, in some embodiments, the step of inputting the synchronized data set into the carbon emission dynamic calculation model to obtain the carbon emission trend prediction results for a preset future time period specifically includes: obtaining the unit operating parameters in the synchronized data set, the unit operating parameters including unit load data, fuel consumption data and equipment operating efficiency data; inputting the unit operating parameters into the prediction model for model training to obtain a carbon emission dynamic calculation model; using the carbon emission dynamic calculation model to predict the carbon emission trend for a preset future time period to obtain a carbon emission trend prediction result.

[0009] In the above example, a complete set of operating parameters, including unit load, fuel consumption, and equipment operating efficiency, was acquired from a synchronized data set to train a prediction model to generate a dynamic carbon emissions estimation model. This model leverages the relationships between various operating parameters for feature learning. This model predicts future carbon emissions trends based on in-depth analysis of multidimensional data, transcending the limitations of traditional single-parameter predictions and achieving accurate predictions of carbon emissions trends.

[0010] In combination with some embodiments of the first aspect, in some embodiments, the synchronous data set is input into the multi-objective optimization model, and optimization calculations are performed with the goals of minimizing carbon emission intensity, optimizing carbon quota utilization, and minimizing power generation costs to obtain the optimization scheme. The steps specifically include: establishing constraints for the multi-objective optimization problem based on the synchronous data set, and the constraints include unit operation constraints, carbon emission constraints, and cost constraints; determining decision variables according to the constraints, and the decision variables include unit output, start and stop status, and operation mode; constructing the objective function of the multi-objective optimization problem, and the objective function includes the objective function of minimizing carbon emission intensity, optimizing carbon quota utilization, and minimizing power generation cost; generating an initial feasible solution population that meets the constraints; iteratively optimizing the feasible solution population, including population stratification, individual selection, cross-mutation, and elite retention; in the iterative optimization process, dynamically adjusting the objective function weights, and maintaining the diversity of solutions; when the optimization iteration meets the preset convergence conditions, selecting a solution that meets the multi-objective balance from the optimal solution set as the optimization scheme.

[0011] In the above example, minimizing carbon emission intensity, optimizing carbon quota utilization, and minimizing power generation costs were modeled as a multi-objective optimization problem. By setting constraints and decision variables, the optimization objectives were quantified. Iterative optimization methods such as population stratification and crossover mutation were employed to dynamically adjust weights and maintain solution diversity, enabling the optimization solution to find the optimal balance between emission reduction effectiveness and economic efficiency. This approach overcomes the limitations of traditional single-objective optimization and achieves global optimization of power plant operation plans.

[0012] In combination with some embodiments of the first aspect, in some embodiments, the step of generating carbon emission monitoring results based on carbon emission trends and optimization schemes, and regulating and controlling unit start-up and shutdown and system load according to the carbon emission monitoring results, specifically includes: determining the unit start-up and shutdown sequence according to the carbon emission trend prediction results, and the unit start-up and shutdown sequence includes determining the start-up and shutdown unit combination, start-up and shutdown time and process control parameters; setting the system operating load based on the optimization scheme, and the system operating load includes the output distribution of each unit, the load change rate and the operating mode; collecting and storing carbon emission data during the operation of the unit, and calculating the carbon emission intensity and carbon quota usage based on the carbon emission monitoring results; when the carbon emission monitoring results deviate from expectations, dynamically adjusting the unit operating parameters, and the unit operating parameters include adjusting the unit output, optimizing the operating mode and updating the control strategy.

[0013] In the above-mentioned embodiment, the unit start-up and shutdown sequences are determined based on carbon emission trend forecasts, the system operating load is set based on an optimization plan, and the forecast and optimization results are converted into specific control instructions. During operation, carbon emission data is collected and monitored in real time. When the monitoring results deviate from expectations, the unit operating parameters are dynamically adjusted. This forms a closed-loop control system of prediction-optimization-control-monitoring-adjustment, solving the problems of delayed response and inaccurate regulation of carbon emission control in power plants.

[0014] In combination with some embodiments of the first aspect, in some embodiments, after generating carbon emission monitoring results based on carbon emission trends and optimization schemes, and performing unit start-up and shutdown and system load adjustment and control according to the carbon emission monitoring results, it specifically includes: collecting carbon emission monitoring results, which include real-time collected CO2 concentration, flue gas flow rate data and CO2 emissions calculated according to the formula; automatically packaging the carbon emission monitoring results to generate blocks according to preset time periods, which record the carbon emission monitoring results, the carbon emission calculation formula version number and the sensor calibration status; submitting the blocks to the blockchain network to complete consensus verification, and writing the blocks that pass the consensus verification into the blockchain.

[0015] In the above example, real-time monitoring data, including CO2 concentration and flue gas flow rate, is collected. The carbon emission monitoring results are packaged into blocks according to preset time periods, and key information such as the calculation formula version and sensor calibration status are recorded. Consensus verification and permanent storage are achieved through the blockchain network, establishing a tamper-proof trust system for carbon emission data, addressing the low credibility and traceability issues of traditional carbon emission data.

[0016] In combination with some embodiments of the first aspect, in some embodiments, after generating carbon emission monitoring results based on carbon emission trends and optimization plans, and performing unit start-up and shutdown and system load regulation and control according to the carbon emission monitoring results, it specifically includes: converting the carbon emission monitoring results into a standard data format, which standard data format includes carbon emissions, power generation, operating time and fuel consumption data; calculating the carbon emission intensity per kilowatt-hour based on the carbon emission monitoring results, and generating carbon emission labels including the production stage and the operation stage; automatically generating a carbon emission report according to the carbon emission monitoring results at preset time intervals; calculating the carbon quota balance based on the carbon emission report, and triggering a trading instruction when the carbon quota balance reaches a preset threshold.

[0017] In the above example, carbon emission monitoring results are converted into a standard data format containing carbon emissions and power generation data, carbon emission intensity per kilowatt-hour is calculated, and a carbon emission label is generated, resulting in an automated carbon emission report. Based on this report, carbon quota balances are calculated and a trading trigger mechanism is set up. This organically integrates carbon emission management with carbon market transactions, addressing the issue of passive and lagging carbon asset management in power plants.

[0018] In combination with some embodiments of the first aspect, in some embodiments, the multi-layer infrared detection structure includes a first detection chamber, a second detection chamber, and a third detection chamber arranged in sequence. The first detection chamber, the second detection chamber, and the third detection chamber are connected by a micro-flow sensor. The first detection chamber absorbs energy in the middle position of the spectrum, and the second detection chamber and the third detection chamber absorb energy in the edge band. The micro-flow sensor detects the pulsating airflow generated by infrared absorption and converts it into an electrical signal. The infrared absorption of the second detection chamber and the third detection chamber is changed by adjusting the sliding switch.

[0019] In the above embodiment, three detection chambers are sequentially arranged in a multi-layer infrared detection structure. The first detection chamber absorbs energy in the middle of the spectrum, while the second and third detection chambers absorb energy at the edges. A micro-flow sensor is used to detect the pulsating airflow generated by infrared absorption. By adjusting the sliding switch to vary the infrared absorption of the second and third detection chambers, differential absorption measurement of infrared energy in different bands is achieved. This fundamentally addresses the issue of traditional single-layer infrared detection being susceptible to interference gases and resulting in unstable measurement accuracy.

[0020] In a second aspect, an embodiment of the present application provides a smart carbon tube platform, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the smart carbon tube platform to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0021] In a third aspect, an embodiment of the present application provides a computer program product comprising instructions. When the computer program product is run on a smart carbon tube platform, the smart carbon tube platform executes the method described in the first aspect and any possible implementation of the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium comprising instructions. When the instructions are executed on a smart carbon tube platform, the smart carbon tube platform executes the method described in the first aspect and any possible implementation of the first aspect.

[0023] It is understood that the smart carbon tube platform provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the methods provided in the embodiments of this application. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects of the corresponding methods and will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. This application utilizes a multi-layer infrared detection structure to monitor desulfurization outlet gas concentration data in real time. This data is systematically integrated with unit load and fuel consumption data to form a synchronized dataset, breaking down data silos. This synchronized dataset is then fed into a dynamic carbon emissions measurement model and a multi-objective optimization model, enabling accurate prediction of carbon emission trends and optimized calculation of operational plans. Based on the predicted results and optimized plans, units are started and shut down, and load adjusted. This establishes a closed-loop management system for data collection, analysis, optimization, and control, enabling precise monitoring and intelligent regulation of carbon emissions in power plants.

[0025] 2. This application obtains a complete set of operating parameters from a synchronized data set, including unit load, fuel consumption, and equipment operating efficiency, to train a prediction model to develop a dynamic carbon emissions estimation model. This model leverages the relationships between various operating parameters for feature learning. The prediction of future carbon emissions trends is based on in-depth analysis of multidimensional data, transcending the limitations of traditional single-parameter predictions and achieving accurate predictions of carbon emissions trends.

[0026] 3. This application models minimizing carbon emission intensity, optimizing carbon quota utilization, and minimizing power generation costs as a multi-objective optimization problem. By setting constraints and decision variables, the optimization objectives are quantified. Iterative optimization methods such as population stratification and cross-mutation are used to dynamically adjust weights and maintain solution diversity, enabling the optimization scheme to find the optimal balance between emission reduction effectiveness and economic efficiency. This overcomes the limitations of traditional single-objective optimization and achieves global optimization of power plant operation plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a flow chart of a method for monitoring the carbon footprint of a power plant throughout its life cycle based on the smart carbon tube platform in an embodiment of the present application; Figure 2 This is a schematic diagram of a multi-layer infrared absorption concentration detection structure of a power plant full life cycle carbon footprint monitoring method based on a smart carbon tube platform in an embodiment of the present application; Figure 3 This is a schematic diagram of a model structure of a method for monitoring the carbon footprint of a power plant throughout its life cycle based on the smart carbon tube platform in an embodiment of the present application; Figure 4 This is a model iteration flow chart of a method for monitoring the carbon footprint of a power plant throughout its life cycle based on the smart carbon tube platform in an embodiment of the present application; Figure 5 This is another flow chart of the method for monitoring the carbon footprint of a power plant throughout its life cycle based on the smart carbon tube platform in an embodiment of the present application; Figure 6 This is a blockchain transaction flow chart of a method for monitoring the carbon footprint of a power plant throughout its life cycle based on the smart carbon pipe platform in an embodiment of the present application; Figure 7 A schematic diagram of the entire carbon trading process of a power plant's full life cycle carbon footprint monitoring method based on the smart carbon pipe platform in an embodiment of the present application; Figure 8 This is a schematic diagram of the physical device structure of the smart carbon tube platform in the embodiment of the present application. DETAILED DESCRIPTION

[0028] The terms used in the following examples of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification of the present application, the singular expressions "a", "an", "above", "the", and "this" are intended to include plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations of one or more of the listed items.

[0029] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to imply or suggest relative importance or implicitly indicate the number of the technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of this application, unless otherwise specified, "plurality" means two or more.

[0030] For ease of understanding, the application scenarios of the embodiments of the present application are introduced below.

[0031] A large thermal power plant with an installed capacity of 2 × 1000 MW generates approximately 10 billion kWh of electricity annually. The plant is required to strictly control its carbon emissions intensity, but faces multiple challenges: First, there are blind spots in the collection of CO2 emissions data at the desulfurization outlet, making it difficult to obtain real-time carbon emissions data; second, operational data such as unit load and fuel consumption is dispersed across multiple systems, including the DCS and EMS, making data integration difficult; and third, it is difficult to accurately predict carbon emission trends and adjust operating methods in response to load fluctuations and changes in fuel quality. These issues hinder the plant's ability to accurately grasp the changing patterns of carbon emissions, hindering the effective implementation of emission reduction measures.

[0032] Currently, the power plant monitors carbon emissions using an indirect accounting method based on fuel consumption. A single-layer infrared detector is installed at the desulfurization outlet to measure CO2 concentrations, while simultaneously collecting unit load and fuel consumption data from the DCS system. This data is stored in an independent storage system and statistically analyzed on a monthly basis. However, this approach has significant shortcomings: the single-layer infrared detection structure is susceptible to interference from other gases, resulting in low measurement accuracy; the independent storage of various data types and the lack of correlation analysis make it impossible to identify the inherent relationship between carbon emissions and operating parameters; and the fixed statistical analysis method is difficult to adapt to changing operating conditions, resulting in poor carbon emission prediction and optimization control.

[0033] After adopting the solution of the present invention, the power plant deployed a multi-layer infrared detection structure at the desulfurization outlet. The differential absorption principle of the three-layer detection chamber improves the accuracy of CO2 concentration measurement. The system automatically collects and integrates multi-source data such as gas concentration, unit load, and fuel consumption to form a synchronized data set. Future carbon emission trends are predicted based on the LSTM model, with a prediction error of less than 5%. Simultaneously, a multi-objective optimization model comprehensively considers carbon emission intensity, carbon quota utilization, and power generation costs to dynamically generate optimization solutions. For example, during periods of low load, the system adjusts the unit start-up and shutdown sequence, prioritizing low-carbon emission units; and when carbon prices rise, it automatically triggers a quota trading strategy. Through these measures, the power plant achieves precise monitoring and intelligent optimization control of carbon emissions, reducing annual carbon emission intensity by approximately 8%.

[0034] For ease of understanding, the following describes the process of the method provided by this implementation in combination with the above scenario. Figure 1, which is a flow chart of a method for monitoring the carbon footprint of a power plant throughout its life cycle based on a smart carbon tube platform in an embodiment of the present application.

[0035] S101 , acquiring unit load data and fuel consumption data, and detecting desulfurization outlet gas using a multi-layer infrared detection structure to obtain gas concentration data.

[0036] Among them, the unit load data represents the real-time output status of the generator set during operation, including parameters such as active power, reactive power, and unit efficiency; the fuel consumption data refers to the real-time usage data of coal, natural gas and other fuels consumed during the operation of the unit; the multi-layer infrared detection structure is used to represent the gas detection device composed of the first detection chamber, the second detection chamber and the third detection chamber, and each detection chamber is connected by a micro-flow sensor; the gas concentration data represents the volume fraction of gas components such as CO2 obtained by infrared spectroscopy analysis.

[0037] This step is executed after the system is initialized and is used to obtain the basic data required for carbon emission calculations. Specifically, the system first establishes a real-time data connection with the DCS system through the OPC-UA protocol to collect unit load data, including operating parameters such as the output and efficiency of the generator set; at the same time, it obtains the real-time consumption of fuels such as coal and natural gas from the fuel management system. At the desulfurization outlet, a multi-layer infrared detection structure detects the concentration of CO2 in the flue gas through the principle of infrared spectroscopy analysis. The first detection chamber absorbs energy in the middle position of the spectrum, and the second and third detection chambers absorb energy in the edge bands. The pulsating airflow generated by infrared absorption is detected by a micro-flow sensor and converted into an electrical signal to achieve high-precision gas concentration measurement.

[0038] In some embodiments, data acquisition and gas detection can be achieved through a variety of methods: optionally, a distributed data acquisition architecture is adopted, edge computing nodes are deployed at the field control layer, and unit DCS data is collected via the ModbusTCP protocol. At the same time, the intelligent fuel metering system is connected to obtain fuel consumption data in real time. Gas detection uses a single-beam alternating infrared analysis method, and the infrared light source is modulated by an electronic chopper. Optionally, based on the industrial Internet of Things architecture, the OPCUA protocol is used to uniformly collect unit operation data, and a data acquisition service is configured to push the data to the cloud. Gas detection uses Fourier transform infrared spectroscopy, and a modulated signal is generated by an interferometer. It is understandable that other data acquisition protocols and gas detection methods can also be used to achieve data acquisition and concentration measurement, which are not limited here.

[0039] In some embodiments, the multi-layer infrared detection structure includes a first detection chamber, a second detection chamber, and a third detection chamber arranged in sequence. The first detection chamber, the second detection chamber, and the third detection chamber are connected by a micro-flow sensor. The first detection chamber absorbs energy in the middle position of the spectrum, and the second detection chamber and the third detection chamber absorb energy in the edge band. The micro-flow sensor detects the pulsating airflow generated by infrared absorption and converts it into an electrical signal. The infrared absorption of the second detection chamber and the third detection chamber is changed by adjusting the sliding switch.

[0040] See also Figure 2 ,like Figure 2 As shown, Figure 2 This is a schematic diagram of a multi-layer infrared absorption concentration detection structure of a power plant full life cycle carbon footprint monitoring method based on a smart carbon tube platform in an embodiment of the present application.

[0041] Figure 2The structure of multi-layer infrared absorption gas detection is shown, which is the core component of CO2 concentration monitoring at the desulfurization outlet of the smart carbon tube platform. Among them, the capillary (1) is responsible for stably introducing the flue gas to be detected into the sample gas chamber (4), and the sample gas flow rate is controlled by the thin diameter design to ensure the stability of the detection environment; the second detector layer (2) is the middle layer of the multi-layer detection structure, which mainly absorbs the edge band energy of the infrared spectrum to assist in identifying interference signals; the slide (10) can change the infrared light absorption path of the second and third detector layers by mechanical sliding, adjust its absorption intensity of the edge band infrared light, and weaken the influence of interfering gases; the micro flow sensor (13) connects each detection layer, and contains two nickel grids heated to 120℃. It forms a Wheatstone bridge with the resistor and can convert the pulsating airflow generated by infrared absorption into an electrical signal. The signal strength is positively correlated with the CO2 concentration; the sample gas chamber (4) is the core area where the sample gas flows. When the infrared light beam passes through here, the CO2 molecules will absorb the characteristic infrared band energy, and the absorption amount varies with the concentration; the modulation disk (5) rotates between the infrared light source and the sample gas chamber, converting the continuous infrared light into an electrical signal. The beam is modulated into an alternating light signal of 81 / 3Hz to avoid interference from ambient light; the chopper motor (6) provides rotational power for the modulation disk to ensure the stability of the modulation frequency; the infrared source (7) is heated to about 600℃ and emits a broad spectrum infrared light including the characteristic absorption band of CO2, which serves as the basis of the light source for detection; the window (9) is made of infrared transparent material and is installed at the optical intersection of the sample gas chamber and the detection layer, which allows infrared light to pass through while isolating the environment and protecting the optical components; the reflector (8) is used to change the propagation direction of the infrared light to ensure that the light beam enters each detection structure accurately; the first detector layer (11) is filled with CO2 gas of known concentration, which mainly absorbs energy in the middle position of the infrared spectrum and is the main signal source for concentration measurement; the third detector layer (12) is similar to the second detector layer, focusing on capturing interference signals at the edge of the pole to further eliminate cross interference; the sample gas inlet receives the pre-treated flue gas, and the sample gas outlet discharges the detected sample gas to achieve continuous monitoring. Overall, the structure achieves high-precision real-time monitoring of CO2 concentration at the desulfurization outlet through multi-layer division of labor absorption, signal conversion and interference adjustment, providing core data support for carbon footprint management.

[0042] S102: Integrate the gas concentration data, unit load data, and fuel consumption data to obtain a synchronized data set.

[0043] Among them, the synchronized data set refers to a standardized data set formed by integrating data from different sources and different sampling periods according to a unified time label; data integration refers to preprocessing operations such as time alignment, unit conversion, and outlier processing on the data; the time label is used to indicate the specific moment of data collection, accurate to milliseconds.

[0044] This step, performed after acquiring various raw data sources, is used to establish a standardized foundation for data analysis. Specifically, the system first aligns the timestamps of data from different sources, converting millisecond-level gas concentration data, second-level unit load data, and minute-level fuel consumption data to a second-level sampling period. Data preprocessing then occurs, including unit conversion (e.g., converting ppm to mg / m³), outlier identification and processing, and data compensation. Finally, the processed data is organized according to a unified data structure, forming a standardized dataset containing fields such as timestamps, parameter values, and quality codes.

[0045] In some embodiments, data integration can be achieved through a variety of methods: Optionally, a time series database storage architecture can be used to write data from different sources into a unified time series database, enabling data association queries and time alignment through SQL statements, while configuring data quality assessment rules to handle outliers. Optionally, a distributed storage system can be used based on a data lake architecture to uniformly manage raw data, and a real-time computing engine can be used to perform data cleaning and feature extraction to achieve data standardization. It is understood that other data storage and processing solutions can also be used to achieve data integration, and this is not limited here.

[0046] S103: Input the synchronized data set into a carbon emission dynamic calculation model to obtain a carbon emission trend prediction result for a preset future period.

[0047] Among them, the dynamic carbon emission measurement model represents a time series prediction model built based on the LSTM deep learning algorithm, including an input layer, a forget gate layer, an input gate layer, an output gate layer, and an output layer; the preset future period refers to the predicted time span, which can be different scales such as hours, days, weeks, or months; the carbon emission trend prediction results are used to represent the changing trend of carbon emissions in the future period, including the predicted values ​​of total emissions and emission intensity.

[0048] This step, performed after obtaining a standardized synchronized dataset, is used to predict future carbon emission trends. Specifically, the system first divides the synchronized dataset into a training set and a validation set in an 80:20 ratio, normalizing the data. This processed data is then fed into an LSTM model. The model's input features include historical carbon emission data, unit load curves, fuel characteristics, and other parameters. A multi-layer neural network is used to extract features and perform time series modeling. Once the model is trained, current operating data is input to predict future carbon emission trends at different timescales.

[0049] In some embodiments, carbon emission prediction can be achieved in a variety of ways: optionally, using an LSTM model based on the Pytorch framework, setting the input dimension to 24 (one data point per hour), the hidden layer dimension to 128, and training the model through the Adam optimizer. The loss function uses the mean square error (MSE), the number of training iterations is 1000 rounds, and the final prediction error on the validation set is controlled within 5%; optionally, based on the ensemble learning method, LSTM is combined with other time series prediction models (such as Prophet, ARIMA, etc.), and the prediction results of multiple models are fused by weighted voting to improve the stability and accuracy of the prediction. It is understandable that other deep learning algorithms or ensemble methods can also be used to achieve carbon emission prediction, which is not limited here.

[0050] In some embodiments, this step specifically includes: obtaining the unit operating parameters in the synchronized data set, the unit operating parameters including unit load data, fuel consumption data and equipment operating efficiency data; inputting the unit operating parameters into the prediction model for model training to obtain a carbon emission dynamic measurement model; using the carbon emission dynamic measurement model to predict the carbon emission trend in a preset future time period to obtain a carbon emission trend prediction result.

[0051] First, a real-time data connection is established with the DCS system via the OPC-UA protocol to collect unit operating parameters. Unit load data represents the actual unit output level and includes unit output and auxiliary power consumption in megawatts (MW). Fuel consumption data includes coal feeder speed, calorific value (MJ / kg) from coal quality test reports, ash content, moisture content, and other characteristic parameters. Equipment operating efficiency data is obtained through the online efficiency calculation module, including real-time calculations of boiler thermal efficiency, turbine thermal efficiency, pipeline heat loss, and other indicators. The system standardizes the collected data, removes outliers, and standardizes time tags to form a training dataset.

[0052] The processed data was then fed into the LSTM prediction model for training. First, an LSTM network structure was constructed, consisting of an input layer (with a parameter dimension equal to the number of unit operating parameters), two LSTM hidden layers (128 neurons per layer), and a fully connected output layer. The Adam optimizer was used, with a learning rate set to 0.001 and a mean squared error (MSE) loss function. The dataset was divided into a training set and a validation set in an 8:2 ratio, and iterative training was performed for 1,000 rounds using a mini-batch approach (batch size of 32). During training, an early stopping mechanism was initiated when the validation set loss showed no significant decrease for 10 consecutive rounds. The resulting trained model for dynamic carbon emissions estimation achieved a prediction error of less than 5% on the validation set.

[0053] Finally, the trained model is used to perform predictions. The model inputs the current unit operating parameters and, leveraging the time series modeling capabilities of the LSTM network, predicts carbon emission trends for future periods. Predictions are divided into two scales: short-term predictions (within 24 hours) using 15-minute intervals to guide unit optimization and regulation; medium- and long-term predictions (7-30 days) using hourly intervals to inform carbon quota management decisions. The system also calculates confidence intervals for the prediction results and dynamically adjusts the prediction window length based on historical prediction errors to ensure prediction accuracy. The prediction results are output as a time series, containing the carbon emission values ​​at each prediction moment and the upper and lower confidence limits of the confidence interval. Through these three steps, accurate predictions of the power plant's carbon emission trends are achieved, providing data support for subsequent optimization and control.

[0054] See also Figure 3 ,like Figure 3 As shown, Figure 3 This is a schematic diagram of a model structure of a method for monitoring the carbon footprint of a power plant throughout its life cycle based on a smart carbon tube platform in an embodiment of the present application.

[0055] As a core component of the dynamic carbon emissions measurement model, it deeply supports the power plant carbon monitoring process. In this structure, the input Xt=[xt, 1, xt, 2, …, xt, n] at time t corresponds to minute-level real-time production data such as desulfurization outlet CO2 concentration and unit load collected via the OPC-UA protocol. The hidden state Ht-1=[ht-1, 1, ht-1, 2, …, ht-1, d] at time t-1 conveys key timing information from previous operations, and the memory Ct-1=[ct-1, 1, ct-1, 2, …, ct-1, d] stores sequence patterns accumulated over long periods of operation. The forget gate Ft=σ(Wf·[Ht-1, Xt]+bf) uses the Sigmoid function to select long-term memory content to be retained or discarded, which can filter out invalid historical data. The input gate It=σ(Wi·[Ht-1, Xt]+bi) controls the inclusion of new input information, and the candidate memory C̃t=tanh(Wc·[Ht-1, Xt]+bc) generates new information to be stored. The two are combined to adapt to real-time working condition changes. Memory update is achieved through Ct=Ft⊙Ct-1+It⊙C̃t. The output gate Ot=σ(Wo·[Ht-1, Xt]+bo) selects the memory information for prediction, and finally generates the hidden state output through Ht=Ot⊙tanh(Ct). This LSTM network integrates time-series production characteristics with business cycle patterns, supports daily to semi-annual carbon emission trend forecasts, and has an error rate that meets industry standards. It can identify carbon quota exceeding risks in advance, guide production strategy adjustments and carbon trading decisions, and adapt to the dynamic carbon emission measurement model's requirements for real-time, long-term forecasting, and decision-making support, helping to build a closed-loop carbon footprint monitoring and management system for the entire life cycle of power plants.

[0056] S104: Input the synchronized data set into a multi-objective optimization model, perform optimization calculations with the goals of minimizing carbon emission intensity, optimizing carbon quota utilization, and minimizing power generation costs, and obtain an optimization solution.

[0057] In some embodiments, this step specifically includes: establishing constraints for a multi-objective optimization problem based on a synchronized data set, the constraints including unit operation constraints, carbon emission constraints, and cost constraints; determining decision variables based on the constraints, the decision variables including unit output, start / stop status, and operation mode; constructing an objective function for the multi-objective optimization problem, the objective function including a carbon emission intensity minimization objective function, a carbon quota utilization optimization objective function, and a power generation cost minimization objective function; generating an initial population of feasible solutions that meet the constraints; iteratively optimizing the population of feasible solutions, including population stratification, individual selection, cross-mutation, and elite retention; dynamically adjusting the objective function weights during the iterative optimization process, and maintaining the diversity of solutions; when the optimization iteration meets the preset convergence conditions, selecting a solution that meets the multi-objective balance from the optimal solution set as the optimization solution.

[0058] Decision variables are determined based on constraints, primarily including unit output, start / stop status, and operating mode. Specifically, unit output refers to the power generated by each unit at different times (corresponding to Pgen,i,t in the formula). It must meet "unit output upper and lower limits" (Pmin,i ≤ Pgen,i,t ≤ Pmax,i) to ensure that output is within the equipment's safe operating range. Start / stop status refers to the unit's startup / shutdown schedule, for example, prioritizing gas-fired units during off-peak periods. By adjusting the start / shutdown mix of high- and low-carbon-emitting units, marginal carbon emissions can be reduced. Operating modes include fuel type adjustments (such as switching between coal and gas) and load adjustments for the carbon capture and use system (CCUS) (such as automatically increasing CCUS load when the carbon price exceeds 80 yuan / ton), directly impacting carbon emission intensity and costs.

[0059] Based on the carbon management requirements of power plants throughout their life cycle, the multi-objective optimization problem constructed includes three core objective functions: The objective function for minimizing carbon emission intensity is as follows: minC=∑i=1(Efuel,i,t·EFCO2,i+Egrid,t·EFgrid), which means minimizing the total carbon emissions of the entire process by calculating the product of the fuel consumption of the computer group (Efuel,i,t) and the corresponding emission factor (EFCO2,i) and adding the carbon emissions of purchased electricity (Egrid,t·EFgrid).

[0060] The objective function for optimizing carbon quota utilization is as follows: maxU = (Qallocated - Qemitted) / Qallocated. This means maximizing the difference between the total annual carbon quota (Qallocated) and the actual emissions (Qemitted) to improve quota utilization efficiency and avoid quota waste or excessive shortfalls.

[0061] The objective function for minimizing power generation cost is as follows: minCost=∑i=1(Cfuel,i·Efuel,i,t+CoM,i)+Pgrid,t·Egrid,t, which means comprehensively calculating the fuel cost (Cfuel,i·Efuel,i,t), fixed operation and maintenance cost (CoM,i) and purchased electricity cost (Pgrid,t·Egrid,t) to achieve a balance between economy and emission reduction.

[0062] When generating an initial population of feasible solutions that satisfy the constraints, the population size is set to 200-500 to cover a sufficiently broad solution space and avoid premature convergence. Based on historical operating data and real-time monitoring data (such as current fuel supply and maximum unit output), solutions that satisfy the fuel supply constraint (∑Efuel,i,t≤Fsupply,t) and the carbon emission quota (Qemitted,t≤Qquota,t+α·Qtrade,in,t) are randomly generated, ensuring that each initial solution is a viable operational plan.

[0063] The NSGA-II algorithm is used for iterative optimization. Its core steps include: population stratification (fast non-dominated sorting), which stratifies the population based on the "dominance relationship" of solutions: if solution A outperforms solution B on all objectives, then A dominates B. This stratification identifies "non-dominated solutions" (Pareto frontiers) and uses them as optimization targets. The computational complexity is kept within O(MN²) (M is the number of objectives, N is the population size). Individual selection is based on "crowding degree," prioritizing solutions with a uniform distribution. Crossover and mutation uses a crossover probability of 0.8-0.9 and a mutation probability of 0.01-0.05 to maintain population diversity through genetic recombination and random mutation. Elite retention merges the parent and child populations, retaining the best N solutions as the parent population for the next generation.

[0064] During the iterative optimization process, dynamic crowding weighting is introduced to assign higher weights to high-dimensional objectives, ensuring that the optimization process is not biased towards a single objective. Crowding calculations maintain solution diversity and prevent excessive clustering of solutions in the objective space. Iterations are terminated when the number of iterations reaches 300-500, or when the target change between two generations is less than 1%.

[0065] See also Figure 4 ,like Figure 4 As shown, Figure 4This is a model iteration flow chart of a method for monitoring the carbon footprint of a power plant throughout its life cycle based on the smart carbon tube platform in an embodiment of the present application; This deeply supports the iterative logic of the "Building a Multi-Objective Optimization Model" phase. The process begins with the generation of an initial population. Based on the power plant's carbon management requirements, it randomly generates a set of candidate solutions, P0 = {s1, s2, …, sN}, containing unit operation strategies and carbon reduction measures (where si is the i-th candidate solution and N is the population size). When the "Generation of the First Generation Subpopulation" is determined, the evolutionary generation number Gen = 2 is initialized, and iteration begins. Each evolutionary generation begins by merging parent and child individuals, fusing the parent population PGen-1 and the child population QGen-1 into a temporary population of size 2N, RGen-1 = PGen-1 ∪ QGen-1, to preserve historically high-quality solutions. If "generating a new parent population" is required, a fast non-dominated sorting and crowding calculation is performed: RGen-1 is first divided into a non-dominated hierarchy Fm={s∈RGen-1|no other solution dominates s} (F1 is the optimal Pareto frontier) using a multi-objective dominance relationship (e.g., optimizing "lowest carbon emission reduction cost, highest carbon quota utilization rate, and lowest power generation cost"). The individual crowding degree I(s)=∑m=1M(Δfm,max-Δfm,min) / Δfm(s) is then calculated (M is the number of targets, and Δfm(s) is the difference between adjacent values ​​on target m) to measure the diversity of solutions. A new parent population is then generated through selection, crossover, and mutation: Individuals with high hierarchy and high crowding are prioritized to form PGen = argmaxs∈RGen-1 {level(Fm), I(s)}. Crossover (schild = cross(sparent1,sparent2,pc)) with probability pc (0.8-0.9) and mutation (smutant = mutate(sparent,pm)) with probability pm (0.01-0.1) yield the offspring generation QGen. When Gen reaches the maximum number of iterations (300-500), the final Pareto frontier solution set Ffinal is output. This process, using the NSGA-II framework, achieves multi-objective collaborative optimization, retaining diverse solutions to adapt to different production scenarios. This directly supports power plants in dynamically adjusting unit operations and optimizing fuel mixes. This model, from theoretical construction to engineering implementation, provides a decision-making basis for carbon footprint management that balances emission reduction costs, quota utilization, and power generation benefits.

[0066] S105. Generate carbon emission monitoring results based on the carbon emission trend and optimization plan, and perform unit start-up and shutdown and system load adjustment and control according to the carbon emission monitoring results.

[0067] Among them, the carbon emission monitoring results refer to the comprehensive monitoring report generated based on real-time data and forecast results, including indicators such as carbon emissions, emission intensity, and quota usage; unit start and stop refers to the start-up and shutdown operation sequence of the generator set; system load is used to represent the output distribution plan of each unit; regulation control refers to the real-time optimization and adjustment process of the unit operating parameters.

[0068] This step is performed after obtaining the carbon emission trend forecast and optimization plan, and is used to achieve closed-loop control of carbon emissions. Specifically, the system first generates monitoring results including real-time carbon emission monitoring data, forecast trends and optimization based on the predicted carbon emission trend and the optimal plan obtained by multi-objective optimization. Then, based on the monitoring results, the system automatically calculates the optimal unit start and stop sequence, including determining the start and stop unit combination, start and stop time and process control parameters. At the same time, the output distribution plan of each unit is set based on the optimization plan, and the load change rate and operation mode are dynamically adjusted. When the actual carbon emission monitoring results deviate from expectations, the system will automatically trigger the control strategy and adjust the unit operating parameters in real time.

[0069] In some embodiments, monitoring and control can be achieved in a variety of ways: optionally, a hierarchical control architecture is adopted, carbon emission monitoring reports and trend analysis charts are generated at the monitoring layer, the control layer executes the unit start and stop control instructions, and the control instructions are sent to the DCS system through the OPC interface. At the same time, a real-time feedback mechanism is configured to automatically adjust the control strategy when carbon emissions exceed the warning threshold; optionally, based on an intelligent control method, a model predictive control (MPC) algorithm is adopted to embed the carbon emission prediction model into the controller, optimize the control sequence in real time, achieve precise control through a feedforward-feedback combination, and set a multi-level alarm mechanism to ensure control safety. It is understandable that other control algorithms or architectures can also be used to achieve carbon emission monitoring and regulation, which is not limited here.

[0070] In conjunction with some embodiments of the first aspect, in some embodiments, after generating carbon emission monitoring results based on carbon emission trends and optimization plans, and adjusting and controlling unit startup and shutdown and system load based on the carbon emission monitoring results, two additional branches are implemented: one is collecting carbon emission monitoring results and using blockchain technology to implement data notarization; the other is converting the monitoring results into a standard format and connecting them to the carbon trading market. The specific implementation methods of these two branches will be described in detail later.

[0071] In some embodiments, this step specifically includes: The start-up and shutdown sequence of the units is determined based on the carbon emission trend forecast results. The start-up and shutdown sequence of the units includes determining the start-up and shutdown unit combination, start-up and shutdown time and process control parameters; the system operating load is set based on the optimization plan. The system operating load includes the output distribution of each unit, the load change rate and the operating mode; the carbon emission data during the operation of the unit is collected and stored, and the carbon emission intensity and carbon quota usage are calculated based on the carbon emission monitoring results; when the carbon emission monitoring results deviate from expectations, the unit operating parameters are dynamically adjusted. The unit operating parameters include adjusting the unit output, optimizing the operating mode and updating the control strategy.

[0072] Based on the carbon emission trends derived from the predictive model, the system optimizes unit start-up and shutdown. A dynamic programming algorithm is used to determine the unit start-up and shutdown combinations within a 24-hour rolling cycle. The unit start-up and shutdown sequence refers to the specific schedule for generator unit startup and shutdown. The start-up and shutdown combinations represent the number and type of units to be operated during the same period. Start-up and shutdown times include specific startup and shutdown times. Process control parameters include heating rate, steam parameters, and grid synchronization. For a 300MW fleet, units with low carbon emission intensity are prioritized when determining which units to start and stop. In practice, the system plans the control parameters for the start-up and shutdown processes based on the minimum start-up and shutdown interval constraints (e.g., a minimum of four hours of continuous operation). The boiler heating rate is controlled within 3°C / minute, and the excitation voltage and speed are controlled during the unit grid connection process to ensure successful first-time grid connection. Furthermore, the control strategies for cold start, warm start, and hot start are dynamically adjusted based on factors such as ambient temperature and load forecasts.

[0073] Based on the NSGA-II multi-objective optimization algorithm, the optimal load allocation plan is calculated by simultaneously considering the two objectives of minimizing carbon emission intensity and power generation costs. The system operating load refers to the overall power plant and individual unit output levels. Unit output allocation represents the specific method for allocating the total load demand to each operating unit. The load change rate refers to the speed limit for unit output changes. Operating modes include peak shaving, frequency regulation, and automatic gearing (AGC). The algorithm uses a population size of 300, 500 iterations, a crossover probability of 0.9, and a mutation probability of 0.05. For operating units, the system calculates each unit's energy consumption and carbon emission characteristics in real time, prioritizing load allocation to units with low energy consumption and low carbon emissions. The load change rate is controlled within the unit's design range (e.g., 2% of rated load / minute) to avoid drastic adjustments that could worsen carbon emissions. The system automatically switches unit operating modes based on grid demand during different periods of time, such as adopting peak shaving during off-peak hours at night and maintaining full load during peak hours.

[0074] The system collects CO2 concentration data once per second using an infrared CO2 analyzer deployed at the desulfurization outlet and flue gas flow rate using a Pitot tube velocity measurement device. Carbon emission data includes real-time monitoring data such as CO2 concentration and flue gas flow rate. Carbon emission intensity refers to carbon emissions per unit of power generation, and carbon quota usage includes indicators such as quota consumption and balance. Blockchain technology is used to store monitoring data, generating a block every 15 minutes to record carbon emission data, calculation formula version number, and sensor calibration status. Carbon emission intensity is calculated based on monitoring data: CO2 emissions are divided by power generation to obtain carbon emission intensity per kilowatt-hour (gCO2 / kWh). Carbon quota usage is also calculated: cumulative carbon emissions are compared with the annual quota to calculate the quota balance rate. The system uses Hyperledger Fabric to build a blockchain network, and the PBFT consensus mechanism ensures that data cannot be tampered with.

[0075] When carbon emission monitoring results are detected to deviate from the predicted range, the system initiates a dynamic adjustment mechanism. A deviation from the predicted range refers to actual carbon emissions exceeding the predicted range. Adjusting unit output involves changing the unit load level. Optimizing operation involves adjusting operating parameters such as coal feed and air volume. Updating the control strategy involves revising the control parameters of the automatic control system. The cause of the deviation is first analyzed. If the deviation is due to improper load distribution, load optimization is re-executed and unit output is adjusted. If the deviation is due to decreased combustion efficiency, operating optimization is implemented, such as adjusting the primary / secondary air ratio and coal feed. The specific adjustment strategy is as follows: When carbon emission intensity exceeds expectations by 5%, the load of high-carbon emission units is reduced to below 70%, and the output of low-carbon units is increased. When the deviation exceeds 10%, DCS control parameters are adjusted, such as reducing the excess air coefficient and optimizing the burner swing angle. The system evaluates the effectiveness of these adjustments hourly to ensure that carbon emissions remain within the expected range. If the adjustment results are poor, the system initiates an online update of the deep learning model, retraining the prediction model based on new data to improve prediction accuracy.

[0076] The following is a more detailed description of the process of the method provided by this implementation. Figure 5 , which is another flow chart of the method for monitoring the carbon footprint of a power plant throughout its life cycle based on the smart carbon tube platform in an embodiment of the present application.

[0077] S501. Collect carbon emission monitoring results, which include real-time collected CO2 concentration, flue gas flow rate data, and CO2 emissions calculated according to a formula.

[0078] Among them, CO2 concentration refers to the volume fraction of carbon dioxide in the flue gas, with the unit of mg / m³; flue gas flow rate data refers to the flow velocity of flue gas in the pipeline, with the unit of m / s; CO2 emissions are calculated based on gas concentration, flue gas flow rate, pipeline cross-sectional area and operating time, with the unit of ton.

[0079] Carbon emission data is collected in real time based on a continuous monitoring system. A multi-layer infrared CO2 analyzer is arranged in the flue of the desulfurization outlet to collect CO2 concentration data once a second; at the same time, the flue gas flow rate is collected through a pitot tube velocity measuring device. According to the calculation method specified in the "Technical Specifications for Continuous Monitoring of Flue Gas Emissions from Fixed Pollution Sources" (HJ75-2017), CO2 emissions are equal to the product of flue gas flow rate and CO2 concentration multiplied by the operating time. The specific calculation formula is: CO2 emissions = CO2 concentration × flue gas flow rate × pipe cross-sectional area × operating time × correction coefficient. The correction coefficient takes into account the influence of operating parameters such as temperature and pressure, and is corrected by the molar volume of the gas under standard conditions.

[0080] S502: Automatically package the carbon emission monitoring results into blocks according to preset time periods, and the blocks record the carbon emission monitoring results, the version number of the carbon emission calculation formula, and the sensor calibration status.

[0081] Among them, the preset time period represents the time interval for block generation, which is usually set to 15 minutes; a block refers to a data packet containing multiple transaction records; the carbon emission calculation formula version number is used to trace the update record of the calculation method; the sensor calibration status includes calibration parameters such as zero drift and range drift.

[0082] Continuously collected carbon emissions data is automatically packaged into blocks at preset 15-minute intervals. Each block contains all raw data for CO2 concentration and flue gas flow rate during that period, as well as the CO2 emissions calculated using the calculation formula. The calculation formula version number is also recorded to facilitate tracking of changes to the calculation method. The block also contains sensor calibration status information, such as zero-point calibration value, span calibration coefficient, and calibration time. This data is organized according to a unified data structure and forms the main content of the block. Each block is uniquely identified using a cryptographic hash algorithm and is linked to the previous block.

[0083] S503: Submit the block to the blockchain network to complete consensus verification, and write the block that passes the consensus verification into the blockchain.

[0084] Among them, the blockchain network refers to a distributed ledger system composed of multiple nodes; consensus verification refers to the process of confirming the authenticity of block data by nodes in the network; writing to the blockchain means permanently storing the verified blocks in the distributed ledger.

[0085] The resulting blocks are submitted to a consortium blockchain network built on Hyperledger Fabric. Nodes in the network first perform format and integrity checks on the block data. Verification is then performed using the PBFT (Practical Byzantine Fault Tolerance) consensus mechanism, requiring consensus from more than two-thirds of the nodes for verification to pass. The specific process is as follows: A block proposal is first sent to an endorsing node, which simulates the transaction and generates an endorsement signature. The signed transaction is then sent to an ordering node for sorting and packaging. Finally, each ledger node verifies and submits the block. Blocks that pass verification are added to the chain, creating an immutable data record. The entire process typically completes within three seconds, ensuring real-time and reliable data.

[0086] S504: Convert the carbon emission monitoring results into a standard data format, where the standard data format includes carbon emission, power generation, operating time, and fuel consumption data.

[0087] Among them, the standard data format represents a unified data structure that complies with relevant carbon emission trading regulations; carbon emissions are measured in tons of CO2 equivalent; power generation is measured in megawatt-hours (MWh); operating time records the duration of the unit's start and stop status; fuel consumption data includes the consumption of different fuels and their calorific value.

[0088] Raw monitoring data is converted to a standard format. First, carbon emissions data is converted from milligrams per cubic meter to tons of CO2 equivalent. Power generation data is accumulated and summarized, with minute-by-minute data converted to hourly power generation. Operating hours are calculated by analyzing unit start and stop records. Fuel consumption data includes coal, natural gas, and other fuels, along with fuel characteristics such as calorific value and carbon content. The converted data is stored in the format specified in the "Carbon Emission Trading Management Rules" to facilitate integration with the carbon trading market. An automatic verification mechanism is implemented during the data conversion process to ensure the accuracy of the conversion results.

[0089] S505: Calculate the carbon emission intensity per kilowatt-hour based on the carbon emission monitoring results, and generate carbon emission labels including the production stage and the operation stage.

[0090] Among them, the carbon emission intensity per kilowatt-hour of electricity refers to the carbon emissions corresponding to each kilowatt-hour of electricity generation; carbon emissions in the production stage include carbon emissions from equipment manufacturing and installation processes; carbon emissions in the operation stage refer to direct and indirect carbon emissions during the operation of the power plant; and the carbon emission label is a quantitative description of the carbon footprint of the product throughout its life cycle.

[0091] The power plant's carbon emission intensity index is calculated based on monitoring data. Carbon emission intensity per kilowatt-hour is equal to total carbon emissions divided by power generation, calculated using the formula: Carbon emission intensity = carbon emissions / power generation. A life cycle assessment approach is also used, dividing carbon emissions into two phases: production and operation. Carbon emissions in the production phase are calculated using the equipment and materials bill and a carbon emission factor. Carbon emissions in the operation phase include direct emissions from fossil fuel combustion and indirect emissions from purchased electricity. A carbon emission label is generated based on the calculated results, containing information such as the carbon emission intensity per kilowatt-hour, the proportion of carbon emissions in different phases, and the potential for carbon reduction.

[0092] S506: Automatically generate a carbon emission report at preset time intervals based on the carbon emission monitoring results.

[0093] Among them, the preset time interval represents the cycle of report generation, which can be daily, weekly, monthly or quarterly; the carbon emission report refers to a standardized emission data summary file that meets regulatory requirements, including information such as emission volume, accounting method, and data source.

[0094] Carbon emissions reports are automatically generated at preset intervals. These reports strictly adhere to the requirements of the "Guidelines for Corporate Greenhouse Gas Emissions Accounting and Reporting Methodology" and include the following sections: a basic information section records company information, reporting period, and accounting boundaries; an emissions section lists specific data for direct and indirect emissions; an accounting method section explains the basis for selecting emission factors and calculation formulas; and a quality assurance section describes data collection, monitoring equipment maintenance, and other aspects. The system automatically extracts relevant data, generates report documents according to standard templates, and performs data consistency verification. The generated reports can be directly used for carbon emission quota settlement and regulatory review.

[0095] S507: Calculate the carbon quota balance based on the carbon emission report, and trigger a transaction instruction when the carbon quota balance reaches a preset threshold.

[0096] Among them, the carbon quota balance represents the difference between the actual carbon emissions of an enterprise and the quota obtained, which can be positive (surplus) or negative (gap); the preset threshold refers to the quota balance quantity standard that triggers the transaction; the transaction instructions include transaction parameters such as transaction direction, quantity, and price range.

[0097] Based on carbon emission report data, the system calculates allowance balances and executes trading operations. The formula for calculating carbon allowance balances is: Balance = Total Allowances - Actual Emissions. The total allowances include both freely allocated allowances and allowances acquired through trading, while actual emissions are derived from the accounting data in the carbon emission report. When the balance is positive and exceeds a preset surplus threshold (e.g., 20%), the system generates a sell order; when the balance is negative and exceeds a preset deficit threshold (e.g., -10%), the system generates a buy order. Trading orders specify parameters such as the transaction quantity (e.g., 50% of the balance) and the price range (e.g., ±5% of the average price over the past 30 days). The system submits trading orders through the carbon trading platform's API, enabling automated trading of allowances. During trading, the system monitors market price fluctuations in real time and automatically suspends trading when the price exceeds the set range, ensuring transaction security. Detailed information for each transaction, including transaction time, counterparty, transaction price, and transaction quantity, is recorded for subsequent statistical analysis and compliance management.

[0098] In some embodiments, quota trading management can be achieved through a variety of methods: optionally, a carbon price prediction model based on regression analysis is established to predict future price trends based on historical trading data, market supply and demand, and policy changes. Trading strategies are dynamically adjusted, and the system sets risk control measures such as transaction limits and batch trading to improve transaction efficiency through algorithmic trading. Optionally, smart contract technology is used to encode trading rules into the blockchain network, automatically executing transactions when preset conditions are met, and peer-to-peer transactions are achieved through quota tokenization. At the same time, a transaction credit rating mechanism is established to optimize counterparty selection. It is understandable that other trading strategies and technologies can also be used to achieve intelligent trading of carbon quotas, which are not limited here.

[0099] like Figure 6 As shown, Figure 6This is a blockchain transaction flow chart of the carbon footprint monitoring method for the entire life cycle of a power plant based on the smart carbon pipe platform in the embodiment of this application; the figure clearly shows the complete process of realizing carbon data notarization and transaction based on the Hyperledger Fabric alliance chain architecture. First, the application submits a transaction proposal containing carbon data related transaction parameters to the endorsement node. The endorsement node calls the chain code to simulate the execution of the proposal (without updating the ledger), generates a read-write set that records the reading and writing of status data and performs endorsement signature. The generated proposal response is then returned to the application. The application generates a transaction containing the signature endorsement and simulation execution result according to the proposal response and submits it. The transaction is handed over to the sorting service node. After sorting the transactions, the sorting service node packages a group of transactions into blocks and distributes them to the master node. The master node broadcasts the blocks to each accounting node and saves the blocks to the ledger. After each accounting node synchronously receives the blocks, it verifies all transactions in the blocks (including endorsement policies and version conflicts, etc.). After the verification is passed, the read-write set is updated to the state database and the block is submitted to the blockchain. At the same time, the application is notified of the transaction results. The entire process is coordinated by each node to ensure the security and immutability of carbon data and the standardization of transactions, providing technical support for data credibility and transaction compliance in power plant carbon footprint management.

[0100] like Figure 7 As shown, Figure 7 A schematic diagram of the entire carbon trading process of a power plant's full life cycle carbon footprint monitoring method based on the smart carbon pipe platform in an embodiment of the present application; Starting from data collection, sensors are used for real-time monitoring and PLC systems to obtain raw data related to power plant carbon emissions. After data processing, edge computing is used to clean redundant and erroneous information, and feature engineering is used to mine effective features. The data is then input into the LSTM prediction module to predict the carbon emission trend for the next 24 hours based on the long and short-term memory network. An MRV report is then generated to carry out monitoring, reporting, and verification work and obtain third-party certification. After that, the quota accounting is entered, and the remaining quota is allocated through baseline comparison. When the conditions set by the smart contract are met, the transaction trigger module automatically issues an over-quota warning or executes the transaction based on the smart contract. The blockchain transaction link is then initiated, and the distributed ledger and smart contract execution ensure that the transaction is transparent and cannot be tampered with. Finally, the performance and settlement are completed, and the quota delivery and financial settlement are realized. The entire process covers the key steps from power plant carbon footprint monitoring to carbon trading performance, building a complete and intelligent carbon management closed loop.

[0101] The following describes the smart carbon tube platform in the embodiment of the present invention from the perspective of hardware processing. Figure 8 , is a schematic diagram of the physical device structure of the smart carbon tube platform in the embodiment of this application.

[0102] It should be noted that Figure 8The structure of the smart carbon tube platform shown is only an example and should not limit the functions and scope of use of the embodiments of the present invention.

[0103] like Figure 8 As shown, the smart carbon tube platform includes a central processing unit (CPU) 801, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 802 or the program loaded from the storage part 808 to the random access memory (RAM) 803, such as executing the method described in the above embodiment. Various programs and data required for system operation are also stored in RAM 803. CPU 301, ROM 802 and RAM 803 are connected to each other via a bus 804. Input / output (I / O) interface 805 is also connected to bus 804.

[0104] The following components are connected to the I / O interface 805: an input section 306 including an audio input device, push button switches, and the like; an output section 807 including a liquid crystal display (LCD), an audio output device, indicator lights, and the like; a storage section 808 including a hard disk and the like; and a communication section 809 including a network interface card such as a LAN (Local Area Network) card or a modem. The communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to the I / O interface 805 as needed. Removable media 811, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 810 as needed, so that computer programs read from the removable media can be installed in the storage section 808 as needed.

[0105] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for executing the methods illustrated in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via the communication section 809 and / or installed from removable media 811. When executed by the central processing unit (CPU) 801, the computer program performs the various functions defined in the present invention.

[0106] It should be noted that specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0107] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram can represent a module, program segment, or part of the code, and the above-mentioned module, program segment, or part of the code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings.

[0108] Specifically, the smart carbon tube platform of this embodiment includes a processor and a memory, and a computer program is stored in the memory. When the computer program is executed by the processor, the power plant full life cycle carbon footprint monitoring method based on the smart carbon tube platform provided in the above embodiment is implemented.

[0109] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the smart carbon tube platform described in the above embodiments, or may exist independently and not be incorporated into the smart carbon tube platform. The storage medium carries one or more computer programs, and when the one or more computer programs are executed by a processor of the smart carbon tube platform, the smart carbon tube platform implements the power plant lifecycle carbon footprint monitoring method based on the smart carbon tube platform provided in the above embodiments.

[0110] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0111] Those skilled in the art will appreciate that all or part of the process steps in the above-described method embodiments can be implemented by a computer program instructing the relevant hardware. The program can be stored in a computer-readable storage medium, and when executed, the program can include the process steps in the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for monitoring the carbon footprint of a power plant throughout its life cycle based on a smart carbon tube platform, characterized in that: Applied to the smart carbon tube platform, the method includes: Obtain unit load data and fuel consumption data, and detect desulfurization outlet gas using a multi-layer infrared detection structure to obtain gas concentration data; Integrating the gas concentration data, the unit load data, and the fuel consumption data to obtain a synchronized data set; Inputting the synchronized data set into a carbon emission dynamic calculation model to obtain a carbon emission trend prediction result for a preset future period, wherein the carbon emission dynamic calculation model includes an input layer, a forget gate layer, an input gate layer, an output gate layer, and an output layer; Inputting the synchronized data set into a multi-objective optimization model, performing optimization calculations with the objectives of minimizing carbon emission intensity, optimizing carbon quota utilization, and minimizing power generation costs, and obtaining an optimization solution, wherein the multi-objective optimization model includes initial population generation, crossover mutation, fast non-dominated sorting, and congestion calculation; A carbon emission monitoring result is generated based on the carbon emission trend and the optimization scheme, and the start and stop of the unit and the regulation and control of the system load are performed according to the carbon emission monitoring result.

2. The method according to claim 1, characterized in that The step of inputting the synchronized data set into the carbon emission dynamic calculation model to obtain the carbon emission trend prediction result for a preset future period specifically includes: Acquiring unit operating parameters in the synchronized data set, wherein the unit operating parameters include unit load data, fuel consumption data, and equipment operating efficiency data; Inputting the unit operating parameters into the prediction model for model training to obtain a carbon emission dynamic calculation model; The carbon emission dynamic calculation model is used to predict the carbon emission trend in the preset future period to obtain the carbon emission trend prediction result.

3. The method according to claim 1, characterized in that The step of inputting the synchronized data set into a multi-objective optimization model, performing optimization calculations with the objectives of minimizing carbon emission intensity, optimizing carbon quota utilization, and minimizing power generation costs, and obtaining an optimization solution specifically includes: establishing constraints for a multi-objective optimization problem based on the synchronized data set, wherein the constraints include unit operation constraints, carbon emission constraints, and cost constraints; Determine decision variables according to the constraint conditions, wherein the decision variables include unit output, start / stop status, and operation mode; Constructing an objective function for a multi-objective optimization problem, wherein the objective function includes a carbon emission intensity minimization objective function, a carbon quota utilization optimization objective function, and a power generation cost minimization objective function; generating an initial population of feasible solutions that satisfy the constraints; Iteratively optimizing the feasible solution population, including population stratification, individual selection, crossover mutation, and elite retention; During the iterative optimization process, the objective function weight is dynamically adjusted and the diversity of solutions is maintained; When the optimization iteration satisfies the preset convergence condition, a solution that satisfies the multi-objective balance is selected from the optimal solution set as the optimization solution.

4. The method according to claim 1, wherein The step of generating a carbon emission monitoring result based on the carbon emission trend and the optimization scheme, and performing unit start-up and shutdown and system load regulation and control according to the carbon emission monitoring result, specifically includes: Determining a unit start-stop sequence based on the carbon emission trend prediction result, wherein the unit start-stop sequence includes determining a start-stop unit combination, start-stop time, and process control parameters; Setting a system operating load based on the optimization plan, wherein the system operating load includes the output distribution of each unit, the load change rate and the operating mode; Collect and store carbon emission data during unit operation, and calculate carbon emission intensity and carbon quota usage based on carbon emission monitoring results; When the carbon emission monitoring results deviate from expectations, the unit operating parameters are dynamically adjusted, including adjusting the unit output, optimizing the operating mode and updating the control strategy.

5. The method according to claim 1, wherein After the step of generating a carbon emission monitoring result based on the carbon emission trend and the optimization scheme, and performing unit start-up and shutdown and system load adjustment and control according to the carbon emission monitoring result, the method further includes: Collecting the carbon emission monitoring results, which include real-time CO2 concentration, flue gas flow rate data, and CO2 emissions calculated according to a formula; Automatically package the carbon emission monitoring results into blocks at intervals of a preset time period, wherein the blocks record the carbon emission monitoring results, the carbon emission calculation formula version number, and the sensor calibration status; The blocks are submitted to the blockchain network for consensus verification, and the blocks that pass the consensus verification are written into the blockchain.

6. The method according to claim 1, characterized in that After the step of generating a carbon emission monitoring result based on the carbon emission trend and the optimization scheme, and performing unit start-up and shutdown and system load adjustment and control according to the carbon emission monitoring result, the method further includes: Converting the carbon emission monitoring results into a standard data format, wherein the standard data format includes carbon emission, power generation, operating time and fuel consumption data; Calculate the carbon emission intensity per kilowatt-hour based on the carbon emission monitoring results and generate a carbon emission label including the production stage and the operation stage; Automatically generate a carbon emission report at preset time intervals based on the carbon emission monitoring results; The carbon quota balance is calculated based on the carbon emission report, and a transaction instruction is triggered when the carbon quota balance reaches a preset threshold.

7. The method according to claim 6, characterized in that The multi-layer infrared detection structure includes a first detection chamber, a second detection chamber and a third detection chamber arranged in sequence. The first detection chamber, the second detection chamber and the third detection chamber are connected by a micro-flow sensor. The first detection chamber absorbs energy in the middle position of the spectrum, and the second detection chamber and the third detection chamber absorb energy in the edge band. The micro-flow sensor detects the pulsating airflow generated by infrared absorption and converts it into an electrical signal. The infrared absorption of the second detection chamber and the third detection chamber is changed by adjusting the sliding switch.

8. A smart carbon tube platform, characterized in that: The smart carbon tube platform includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the smart carbon tube platform to execute the method described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that: When the instruction is executed on the smart carbon tube platform, the smart carbon tube platform executes the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product is run on the smart carbon tube platform, the smart carbon tube platform is enabled to execute the method according to any one of claims 1 to 7.

Images