A method for carbon emission accounting and quota allocation of power grid from the perspective of carbon emission trading

Abstract

This application is applicable to the field of information technology, and provides a method for carbon emission accounting and quota allocation of power grids from the perspective of carbon emission trading, including: calculating the transmission loss rate of different transmission lines; establishing a clean energy output prediction model; predicting the emission reduction potential of the target area power grid under a preset clean energy scale, the emission reduction potential is the ability to reduce the transmission loss rate and improve the clean energy utilization rate; generating a scheduling plan for the transmission line; based on the scheduling plan of the transmission line, establishing an intelligent scheduling model, generating the optimal clean energy scale, generating the evaluation results of the emission reduction effect of different transmission lines based on the market economic value, determining the transmission line for priority scheduling; generating a carbon emission quota allocation plan. The present invention effectively supports the sustainable development and environmental protection goals of the power system, promotes the reduction of carbon emissions, and maximizes economic benefits.

Description

Technical Field

[0001] The present application belongs to the field of information technology, and in particular, relates to a method for carbon emission accounting and quota allocation for power grids from the perspective of carbon emission trading. Background Art

[0002] There are many technical contradictions and challenges in predicting the output of renewable energy and the effect of emission reduction. In order to reduce the transmission loss of electricity from power plants to end users, it is necessary to strengthen the maintenance and upgrading of transmission equipment. However, this will increase the cost and difficulty of maintenance. In addition, while reducing the transmission loss of transmission lines, it is also necessary to accurately calculate and account for the carbon emissions of the power grid. The introduction of efficient transmission equipment and the elimination of old equipment require balancing costs and benefits, which is extremely challenging in actual operation. Therefore, scientifically accounting for carbon emissions of the power grid and reasonably allocating quotas is a major challenge in carbon emissions trading. This involves how to accurately calculate the carbon emissions of the power grid and allocate quotas fairly and effectively on this basis. In the allocation of carbon emission quotas, it is necessary to comprehensively consider the conflicts of interest between power generation companies and power grid operators.

[0003] Existing technologies cannot guarantee fairness and effectiveness in carbon emission accounting and quota allocation, leading to conflicts of interest among multiple parties. In the process of reducing power transmission losses, accurately calculating grid carbon emissions, and formulating a scientific and reasonable carbon emission quota allocation plan, how to find the best balance between cost, benefit and fairness is a single problem that we need to solve urgently. Summary of the invention

[0004] The embodiment of the present application provides a method for carbon emission accounting and quota allocation of power grids from the perspective of carbon emission trading, which can solve the problem of how to find the best balance between cost, benefit and fairness in the process of reducing power transmission losses, accurately calculating the carbon emissions of the power grid and formulating a scientific and reasonable carbon emission quota allocation plan.

[0005] The present application embodiment provides a method for calculating carbon emissions and allocating quotas for a power grid from the perspective of carbon emissions trading, including:

[0006] Acquire transmission line data of the target area power grid, and calculate the transmission loss rate of different transmission lines according to the transmission line data, wherein the transmission line data of the target area power grid includes line length, transmission capacity, load fluctuation and ambient temperature information;

[0007] Acquire historical output data and real-time environmental data of each clean energy source, and establish a clean energy output prediction model based on the historical output data and the real-time environmental data, wherein the clean energy output prediction model is used to predict the total output of each clean energy source in the next cycle, wherein the clean energy sources include solar energy, wind energy, and hydropower;

[0008] Based on the predicted total output of clean energy and the transmission loss rates of different transmission lines, the emission reduction of clean energy on each transmission line is obtained, and based on the emission reduction, the emission reduction potential of the target area power grid under the preset clean energy scale is predicted, and the emission reduction potential is the ability to reduce the transmission loss rate and improve the utilization rate of clean energy;

[0009] Generate a transmission line dispatch plan based on the transmission loss rates of different transmission lines and asset data information of different transmission lines;

[0010] Based on the dispatch plan of the transmission line, a deep reinforcement learning algorithm is used to establish an intelligent dispatch model to generate an optimal clean energy scale, where the optimal clean energy scale is a clean energy scale that minimizes the transmission loss rate and improves the clean energy utilization rate;

[0011] Based on the emission reduction potential under the optimal clean energy scale and the historical carbon price data of the carbon emission trading market, generate the evaluation results of the emission reduction effects of different transmission lines based on the market economic value, and determine the transmission lines for priority scheduling;

[0012] Based on the evaluation results and the carbon emission data of different transmission lines and target areas, a carbon emission quota allocation plan is generated.

[0013] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0014] The present invention discloses a method for calculating carbon emissions and allocating quotas for power grids from the perspective of carbon emissions trading, which accurately evaluates and optimizes the transmission losses of each transmission line, and accordingly prioritizes the maintenance and upgrading of those transmission lines with high potential for reducing carbon emissions; the present invention not only optimizes the transmission efficiency of the power grid and the utilization rate of clean energy, but also increases the emission reduction potential of the transmission lines, while improving the economic returns in the carbon emissions trading market. Overall, the present invention effectively supports the sustainable development and environmental protection goals of the power system, promotes the reduction of carbon emissions, and maximizes economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0016] Figure 1 It is a flow chart of a method for carbon emission accounting and quota allocation for a power grid from the perspective of carbon emission trading provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0018] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0019] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0020] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0021] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0022] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0023] See also Figure 1As shown, the present invention is a method for calculating carbon emissions and allocating quotas for power grids from the perspective of carbon emissions trading, comprising the following steps:

[0024] S100, acquiring transmission line data of a target area power grid, and calculating transmission loss rates of different transmission lines according to the transmission line data, wherein the transmission line data of the target area power grid includes line length, transmission capacity, load fluctuation, and ambient temperature information;

[0025] In one embodiment, the step S100 includes:

[0026] A data acquisition system is used to monitor and record the transmission line data in real time, wherein the transmission line data includes line length, transmission capacity, load fluctuation, and ambient temperature information;

[0027] Cleaning and preprocessing the transmission line data to generate usable transmission line data;

[0028] Based on the available transmission line data, a transmission loss rate calculation model is established using a power flow calculation method;

[0029] According to the transmission loss rate calculation model, the current distribution and power loss of the transmission line are calculated to obtain the actual transmission loss rate of the transmission line;

[0030] Based on the historical operation data and environmental data of the transmission lines, a transmission loss rate prediction model is established using a machine learning algorithm;

[0031] The transmission loss rate of the transmission line is predicted according to the transmission loss rate prediction model.

[0032] In this embodiment, when acquiring the transmission line data of the target area power grid, the data acquisition system is used for real-time monitoring and recording. Specifically, the current, voltage, power and other parameters of the transmission line are monitored in real time by using intelligent power sensors, and the data are uploaded to the cloud server at intervals of 1 minute. At the same time, the temperature, humidity, wind speed and other information of the environment where the transmission line is located are collected by meteorological sensors, and the data are uploaded to the cloud server at intervals of 5 minutes.

[0033] After that, the collected transmission line data needs to be cleaned and preprocessed to remove outliers and invalid data to ensure the accuracy and completeness of the data, laying the foundation for the subsequent transmission loss rate calculation and line evaluation. Specifically, the sliding window algorithm is used to smooth the data and remove outliers that exceed 20% of the normal range. The Kalman filter algorithm is used to filter the data to eliminate high-frequency noise and improve the signal-to-noise ratio of the data.

[0034] When establishing the transmission loss rate calculation model, the improved power flow calculation method is used, and the resistance, reactance, capacitance and other parameters of the transmission line, as well as the influence of ambient temperature on the resistivity of the transmission line, need to be considered. By solving the power flow equation, the current distribution and power loss of the transmission line are calculated, and the transmission loss rate is obtained. Specifically, the technical parameters of each transmission line of the target area power grid are obtained, including line length, voltage level, and conductor material. The technical parameters are used as input conditions for calculating the transmission loss rate, and the Newton-Raphson method is used for power flow calculation. This method iteratively solves the power flow equation group to calculate the voltage amplitude and phase angle of each node in the target area power grid, and then calculates the current distribution and power loss of the transmission line. It is worth noting that the input data for power flow calculation include the topological structure of the target area power grid, line parameters (resistance, reactance, susceptance), node type (PQ node, PV node, balance node), and active power and reactive power of the node. Through power flow calculation, the active power, reactive power, current and loss of each transmission line, as well as the power factor and voltage drop of the transmission line can be obtained, the transmission loss rate of clean energy on different transmission lines is analyzed, and the influence of line parameters such as resistance, reactance and capacitance on the loss is considered to obtain the transmission efficiency and loss coefficient of each line.

[0035] By adopting the support vector machine (SVM) algorithm and taking the historical operation data and environmental data of the transmission line as features, a transmission loss rate prediction model is established to estimate the actual loss rate of the transmission line, and the prediction error is controlled within 1%.

[0036] S200, acquiring historical output data and real-time environmental data of each clean energy, and establishing a clean energy output prediction model based on the historical output data and the real-time environmental data, wherein the clean energy output prediction model is used to predict the total output of each clean energy in the next cycle, wherein the clean energy includes solar energy, wind energy, and hydropower;

[0037] In one embodiment, the above step S200 includes:

[0038] Collect real-time environmental data through IoT sensors, including sunlight intensity, wind speed and direction, and water flow speed;

[0039] Obtaining historical output data of clean energy through various clean energy power generation facilities, wherein the historical output data is clean energy power generation data;

[0040] Matching and aligning the real-time environment data and the historical output data to establish a time series data set, wherein the time series data set includes a training set, a validation set, and a test set;

[0041] Establishing a clean energy output prediction model, and training and optimizing the clean energy output prediction model through the time series data set;

[0042] The total output of each clean energy in the next cycle is predicted by the clean energy output prediction model;

[0043] Within the preset time, the historical output data and real-time environmental data of each clean energy are updated, and the clean energy output prediction model is updated and optimized.

[0044] In this embodiment, real-time environmental data such as solar intensity, wind speed and direction, and water flow velocity are collected in real time through IoT sensors. For the collection of solar intensity, light intensity sensors such as photoresistors and photodiodes can be used, which are arranged near solar panels to measure light intensity in real time; for the collection of wind speed and direction, wind speed sensors and wind direction sensors such as three-cup anemometers, ultrasonic anemometers, and wind vanes can be used, which are arranged near wind turbines to measure wind speed and direction in real time; for the collection of water flow velocity, flow velocity sensors such as electromagnetic flow meters and ultrasonic flow meters can be used, which are arranged at the water inlet or tailwater channel of the hydroelectric generator set to measure the water flow velocity in real time, and the data is transmitted to a database for storage and management, and historical output data is obtained from various clean energy power generation facilities at the same time, wherein the historical output data is the power generation data of clean energy, thereby forming a complete clean energy output data set.

[0045] After that, the collected real-time environmental data is preprocessed and feature-engineered to remove outliers and noise data, and the missing values ​​are reasonably filled. The correlation between the real-time environmental data and the historical output data of clean energy is analyzed through data visualization technology. Python's Matplotlib, Seaborn and other visualization libraries can be used to draw scatter plots, correlation heat maps, etc. between the real-time environmental data and the historical output data to intuitively display the relationship between them. Business intelligence tools such as Tableau and PowerBI are used to create interactive dashboards to display environmental data and energy output data in real time for easy monitoring and analysis, and select environmental features that have a greater impact on clean energy output as input to the clean energy output prediction model.

[0046] In the process of establishing the clean energy output prediction model, the historical output data of clean energy is matched and aligned with the real-time environmental data to construct a time series data set, and the data set is divided into a training set, a validation set, and a test set. The training set is used for training the clean energy output prediction model, the validation set is used for parameter adjustment and selection of the clean energy output prediction model, and the test set is used to evaluate the performance of the clean energy output prediction model.

[0047] Select machine learning models suitable for time series prediction, such as long short-term memory network (LSTM), temporal convolutional network (TCN), etc. LSTM is a recurrent neural network that can effectively handle long-term dependencies in time series data by introducing a gating mechanism. The input of LSTM includes the feature vector at the current moment and the hidden state at the previous moment, and the output is the hidden state and predicted value at the current moment; TCN is a time series model based on convolutional neural network. Through causal convolution and residual connection, it can learn the local and global features of time series in parallel. The input of TCN is time series data, and the output is the predicted value for each time step. Use the above machine learning model to establish a clean energy output prediction model, and use the training set data to train the model, and continuously optimize the parameters of the clean energy output prediction model through the back propagation algorithm.

[0048] In the training process of the clean energy output prediction model, cross-validation and grid search techniques are used to tune the hyperparameters of the clean energy output prediction model. Cross-validation is to divide the training set data into k subsets, select one of the subsets as the validation set each time, and the remaining subsets as the training set. The clean energy output prediction model is trained k times, and the average of the k validation results is taken as the performance indicator of the clean energy output prediction model. Commonly used methods are k-fold cross-validation and leave-one-out cross-validation; grid search is to set the hyperparameters that need to be tuned to a discrete value range, traverse all possible parameter combinations, train the clean energy output prediction model for each combination and evaluate the performance, select the parameter combination with the best performance as the final clean energy output prediction model hyperparameters, such as the number of hidden layers, the number of hidden units, the learning rate, etc. of LSTM, and select the clean energy output prediction model with the best performance on the validation set as the final clean energy output prediction model.

[0049] The trained clean energy output prediction model is tested through the test set data to evaluate the prediction performance of the model. Specifically, indicators such as root mean square error (RMSE) and mean absolute error (MAE) are used to measure the prediction accuracy of the model. If the model performance meets the requirements, it can be deployed in the production environment for real-time prediction.

[0050] Through the clean energy output prediction model, combined with real-time environmental data and historical output data, the total output of solar energy, wind energy and hydropower in the next cycle is predicted to provide decision support for power grid dispatch and energy management. At the same time, the prediction model is continuously updated and optimized, and the prediction model is retrained regularly (such as monthly) using the latest real-time environmental data and corresponding energy output data to enable the model to adapt to changes in data distribution.

[0051] Specifically, when collecting clean energy environmental data, you can choose the DS18B20 digital temperature sensor to collect the ambient temperature near the solar panel. The sensor has a measurement range of -55°C to +125°C, with an accuracy of ±0.5°C. It communicates with the data acquisition terminal through a single bus protocol and collects temperature data once a minute. For the collection of wind speed and direction, you can use an RS485 wind speed and direction sensor, and use the ModbusRTU communication protocol to collect wind speed and direction data at intervals of 1 second. At the water inlet of the hydropower generator set, use an E8000 electromagnetic flowmeter to measure the water flow velocity, with a flow measurement range of 0-100000m 3 / h, with an accuracy of ±0.5%, and flow rate data is collected in a cycle of 10 seconds. The collected data is transmitted to the InfluxDB time series database in the cloud through the MQTT protocol, and the amount of data stored every day is about 500MB. In the data preprocessing stage, Python's Pandas library is used to clean the data and remove abnormal values ​​that exceed the physical threshold, such as temperature exceeding 80℃ and wind speed exceeding 50m / s. For missing values, linear interpolation is used to fill them. For data that is missing for more than 30 minutes, the mean of the previous and next time periods is used to fill them. Then, Matplotlib is used to draw a scatter plot of environmental factors such as temperature, wind speed, wind direction, and flow rate with power generation, and the correlation between them is analyzed. It is found that temperature and photovoltaic power generation show a significant positive correlation, and the correlation coefficient reaches above 0.8. The Pearson correlation coefficient and Spearman rank correlation coefficient are used to quantify the correlation between each environmental factor and power generation, and the features with a correlation coefficient greater than 0.6 are selected as the input of the clean energy output prediction model.

[0052] When constructing the time series dataset, the environmental data and power generation data are aligned in hours. Each sample contains data from the past 168 hours (a week), forming a 7×24 two-dimensional matrix. The dataset is divided into training set, validation set, and test set in a ratio of 8:1:1. For photovoltaic power generation, the LSTM algorithm is used to build a clean energy output prediction model and predict it. The input features include environmental data such as temperature and light intensity in the past week, and the output is the photovoltaic power generation in the next day. After 50 epochs of training, the mean absolute error (MAE) of the model on the validation set is 20kWh, and the relative error does not exceed 5%. The grid search method is used for hyperparameter optimization, and the search range is: number of hidden units [32, 64, 128], learning rate [0.001, 0.01, 0.1], batch size [16, 32, 64]. Finally, the model with 64 hidden units, 0.01 learning rate, and 32 batch size is selected as the optimal model. This model is used to predict the test set data, and the mean absolute error is further reduced to 18kWh. The model is fine-tuned monthly using the environmental and power generation data of the most recent month to adapt to changes in data distribution.

[0053] S300, based on the predicted total output of clean energy and the transmission loss rate of different transmission lines, obtain the emission reduction of clean energy on each transmission line, and according to the emission reduction, predict the emission reduction potential of the target area power grid under the preset clean energy scale, wherein the emission reduction potential is the ability to reduce the transmission loss rate and improve the utilization rate of clean energy;

[0054] In one embodiment, obtaining the emission reduction of clean energy on each transmission line based on the predicted total output of clean energy and the transmission loss rate of different transmission lines includes:

[0055] allocating the predicted total clean energy output according to a transmission line allocation method of a target regional power grid;

[0056] According to the transmission loss rate of the transmission line, determine the amount of clean energy lost during the transmission process, and obtain the actual amount of clean energy used on each transmission line;

[0057] Based on the actual electricity consumption of clean energy, the carbon emission factor method is used to calculate the emission reduction of clean energy on each transmission line.

[0058] Specifically, the transmission line allocation method includes an allocation method based on node electricity price and an allocation method based on line congestion. When the allocation method based on node electricity price is adopted, clean energy is allocated to nodes with higher electricity prices first according to the electricity price level of each node to maximize the economic benefits of clean energy; when the allocation method based on line congestion is adopted, clean energy is allocated to lines with lower congestion and surplus transmission capacity first to reduce the transmission congestion and power abandonment risks of clean energy. In addition, when allocating clean energy output, it is also necessary to consider the geographical location and access voltage level of clean energy power stations, and reasonably arrange the grid connection and transmission path of clean energy. According to the transmission loss rate of the transmission line, the power loss of clean energy during the transmission process is determined, and the actual power utilization of clean energy on each line is obtained.

[0059] Based on the actual amount of electricity used by clean energy, the carbon emission factor method is used to calculate the amount of carbon emissions reduced by replacing traditional fossil energy power generation with clean energy. This method calculates the amount of carbon emissions reduced by replacing fossil energy power generation with clean energy based on the carbon dioxide emissions per unit of power generation of different types of generators. The calculation formula is emission reduction = actual electricity used × (carbon emission factor of replaced fossil energy - carbon emission factor of clean energy), where the carbon emission factor of fossil energy can be the average value of the country or region, or a weighted average based on the emission factor of the specific fossil energy type replaced (such as coal, natural gas). The carbon emission factor of clean energy can be estimated based on its life cycle emission intensity. For example, the carbon emission factors of hydropower, wind power, and photovoltaic power generation are approximately 4, 15, and 50 grams of CO2 / kWh, respectively.

[0060] In one embodiment, the step of predicting the emission reduction potential of the target regional power grid under a preset clean energy scale based on the emission reduction amount includes:

[0061] Based on the emission reduction of different transmission lines, the overall emission reduction of the target area power grid is determined through data visualization technology;

[0062] Using scenario analysis method, different clean energy development scenarios are set, and different clean energy development scenarios correspond to different clean energy scales. The clean energy development scenarios include baseline scenario, positive scenario and radical scenario;

[0063] For different clean energy development scenarios, based on the predicted total clean energy output and the predicted total social electricity consumption, the emission reduction potential of the target area power grid is predicted under the preset clean energy scale.

[0064] Specifically, we compared and analyzed the emission reductions of different transmission lines, and through data visualization technology, generated emission reduction ranking charts and spatial distribution maps, intuitively showing the performance of each transmission line in emission reduction, and comprehensively analyzing the emission reductions of each transmission line to obtain the reduction in the overall carbon emissions of the power grid due to the use of clean energy, and analyzed the marginal effect of increasing the proportion of clean energy on power grid emission reduction.

[0065] In addition, the scenario analysis method is used to set different clean energy development scenarios, such as the baseline scenario, the positive scenario and the radical scenario, which correspond to different clean energy scales. In each scenario, the emission reduction potential of the target area power grid under the preset clean energy scale is predicted based on the predicted total output of clean energy and the predicted total social electricity consumption. Specifically, energy scenario analysis models such as LEAP (Long-range Energy Alternatives Planning System) can be used for quantitative research. At the same time, it is also necessary to consider the impact of external factors such as grid planning, power market reform, carbon trading and carbon pricing on the emission reduction potential of clean energy, conduct sensitivity analysis, and predict the emission reduction potential when the scale of clean energy is expanded in the future. According to the prediction results of the emission reduction effect, technical solutions for optimizing transmission lines are proposed, such as increasing clean energy grid connection points, upgrading the conductor material of transmission lines, optimizing power flow scheduling strategies, etc., and by reducing transmission losses and improving the utilization rate of clean energy, the emission reduction benefits of clean energy are further improved, providing a decision-making basis for the low-carbon transformation of the power grid.

[0066] S400, generating a transmission line dispatching plan based on transmission loss rates of different transmission lines and asset data information of different transmission lines;

[0067] In one embodiment, the step S400 includes:

[0068] Obtain asset data information of different transmission lines, the asset data information including construction time, design parameters, operating status and historical fault data of each line;

[0069] Based on the electrical parameters, mechanical parameters and environmental parameters of the transmission line, the health index of the transmission line is calculated using the fuzzy comprehensive evaluation method;

[0070] Based on the health index and historical fault data, a remaining life prediction model of the transmission line is established, wherein the remaining life prediction model is used to predict the remaining service life of the transmission line;

[0071] Establishing a multi-classification model of the state of the transmission line, wherein the multi-classification model is used to generate potential failure risks of the transmission line;

[0072] Based on the remaining service life and potential failure risk of the transmission line, a dispatch plan for the transmission line is generated, wherein the dispatch plan includes a maintenance and upgrade plan for the transmission line.

[0073] In this embodiment, the asset data information of the transmission line is obtained through the power grid company, wherein the asset data information includes the construction time, design parameters, operating status, historical fault records and other information of each line. At the same time, online monitoring and fault diagnosis technology is used to collect the electrical parameters, environmental parameters and mechanical parameters of the transmission line in real time. When evaluating the health status of the line, a fuzzy comprehensive evaluation method is adopted, considering the electrical parameters (such as resistance, reactance, capacitance, etc.), mechanical parameters (such as conductor sag, tension, etc.) and environmental parameters (such as temperature, humidity, wind speed, etc.) of the transmission line, so as to calculate the health index of the transmission line.

[0074] Through the health index and historical fault data of the transmission line, a remaining life prediction model of the transmission line based on Weibull distribution is established to predict the remaining service life of the transmission line, with a prediction accuracy of more than 90%. When formulating line maintenance and upgrade plans, priority is given to lines with a health index lower than 0.7 and a remaining life of less than 5 years, and work tasks and resource allocation are arranged reasonably to improve overall transmission efficiency and reliability.

[0075] Through big data analysis and machine learning algorithms, a multi-classification model based on machine learning is established, such as random forests and support vector machines. The input features of the multi-classification model include the voltage, current, temperature, fault frequency, etc. of the line. Through feature engineering and feature selection methods, such as correlation analysis and recursive feature elimination, the key features that have the greatest impact on the line status are screened out. The multi-classification model is trained and tuned using methods such as cross-validation and grid search, and the classification accuracy and robustness of the multi-classification model are continuously iterated and optimized. When applying the multi-classification model, various parameters of the transmission line are collected in real time and input into the trained multi-classification model to obtain the potential fault risk of the transmission line.

[0076] Based on the remaining service life and potential failure risk of the transmission line, a transmission line dispatch plan is generated. Specifically, according to the aging degree, blocking level, fault type and other characteristics of the line, suitable transmission technology and equipment are matched. For example, for seriously aged transmission lines, key components such as conductors and insulators are replaced; for transmission lines with frequent overloads, capacity increase and splitting are adopted to improve the transmission capacity and reliability of the transmission line; for lines across different terrains and climate zones, special conductor materials such as high temperature resistance and anti-icing are used to improve the adaptability and reliability of the line. For long-distance and large-capacity transmission lines, flexible AC transmission technology is given priority. Through controllable high-voltage power electronic devices such as static VAR compensators (SVCs) and static synchronous compensators (STATCOMs), the reactive power and voltage of the line are dynamically adjusted to improve the stability and transmission capacity of the line. For transmission lines prone to lightning strikes and icing, measures such as insulator series resistors and high-damping conductors are adopted to improve the anti-interference and reliability of the line. By optimizing the reactive power and harmonic characteristics of the line, reducing the reactance and susceptance of the line, reducing the power loss and voltage fluctuation of the line, and improving the transmission efficiency and stability of the line.

[0077] In one embodiment, the step S400 further includes:

[0078] According to the emission reduction potential and dispatching cost of each transmission line, a project management system is established and quantitative key performance indicators are formulated, including transmission line availability, transmission loss rate, clean energy utilization rate, and return on investment;

[0079] Use the balanced scorecard to establish a performance evaluation system for the key performance indicators;

[0080] Based on the project management system and performance evaluation system, a performance evaluation is conducted on the dispatched transmission lines to evaluate the cost-effectiveness and implementation value of the transmission line dispatch.

[0081] In this embodiment, according to the emission reduction potential and dispatching cost of each line, by establishing a project management system and performance evaluation system for transmission line dispatching, economic evaluation methods such as life cycle cost analysis and investment return analysis are adopted to weigh the cost-effectiveness of transmission line dispatching, and transmission lines with high emission reduction potential and short investment return period are preferentially selected for dispatching. At the same time, the impact of the dispatching plan on the overall benefits and social benefits of the power grid is considered, such as improving the safety of the power grid, promoting the consumption of clean energy, and reducing carbon emissions, and the feasibility and necessity of transmission line dispatching are comprehensively evaluated. Specifically, a project management system is established, and quantitative key performance indicators (KPIs) such as transmission line availability, transmission loss rate, clean energy utilization rate, and investment return rate are formulated to conduct a full life cycle performance evaluation of the dispatched transmission line. In addition, a performance evaluation indicator system is established from multiple dimensions such as finance, customers, internal processes, and learning and growth by using tools such as the balanced scorecard (BSC). Based on the project management system and performance evaluation system, the cost-effectiveness and implementation value of transmission line dispatching are comprehensively evaluated.

[0082] Specifically, when obtaining the asset data information of the power grid's transmission lines, ETL tools can be used to extract key information such as the laying time, conductor model, number of towers, and fault frequency of the transmission lines from different data sources, such as ERP systems, EAM systems, and SCADA systems. Then, through data cleaning and conversion, a unified format of the transmission line asset data information library is generated. Association rule mining algorithms, such as Apriori and FP-growth, are used to analyze the correlation between line faults and line attributes, and to find out the key factors that cause line failures. For example, the probability of failure of lines with a running time of more than 20 years is more than twice that of lines within 5 years. A multi-classification model of line status is established using the random forest algorithm. The input features include 30 parameters such as line voltage, current, and temperature. The model hyperparameters are optimized through grid search to obtain a line status classification with an accuracy of more than 95%. For long-distance lines spanning more than 100 kilometers, flexible AC transmission technology is used for upgrading and transformation. By installing 10 static VAR compensation devices, each with a capacity of 50Mvar, the line transmission capacity can be increased by more than 20% and the line loss can be reduced by 1.5 percentage points. Using deep reinforcement learning algorithms to establish intelligent scheduling models, 500 line operation scenarios were designed, each of which contained 24-hour operation data of 8 lines. Through 10,000 training iterations, the optimal transmission line scheduling plan was generated, so that the clean energy scale was the optimal clean energy scale, which could increase the clean energy consumption ratio by more than 10%. At the same time, using the two indicators of IRR and NPV, assuming that the investment in the scheduling of transmission lines is 50 million yuan, the operation period is 10 years, and the annual average power loss is saved by 20 million kWh. According to the electricity price of 0.5 yuan / kWh and the discount rate of 5%, the IRR of the project is 12%, the NPV is 30 million yuan, and the investment recovery period is 6 years, which has good economic benefits and emission reduction benefits.

[0083] S500, based on the dispatch plan of the transmission line, using a deep reinforcement learning algorithm, establishing an intelligent dispatch model, and generating an optimal clean energy scale, wherein the optimal clean energy scale is a clean energy scale that minimizes the transmission loss rate and improves the clean energy utilization rate;

[0084] In this embodiment, deep reinforcement learning algorithms, such as DeepQ-Network (DQN) and PolicyGradient, are used to establish an intelligent dispatching model for transmission lines. Through a large amount of simulation training and self-learning of measured data, the dispatching model is continuously optimized to achieve safe and economical operation of transmission lines. According to the new characteristics of the transmission lines after maintenance and upgrades, the intelligent dispatching model is retrained and adaptively learned to continuously improve the control performance and robustness of the model, find the optimal dispatching plan for transmission lines and the grid connection method for clean energy, and minimize transmission losses and improve the utilization rate of clean energy while ensuring the safe and stable operation of the power grid.

[0085] S600, based on the emission reduction potential under the optimal clean energy scale and the historical carbon price data of the carbon emission trading market, generate the evaluation results of the emission reduction effects of different transmission lines based on the market economic value, and determine the transmission lines for priority scheduling;

[0086] In one embodiment, the step S600 includes:

[0087] Based on the emission reduction potential of each transmission line after scheduling and the historical carbon price data of the carbon emission trading market, a time series prediction model is used to establish a carbon price prediction model, and the carbon price prediction model is used to generate a carbon price prediction result;

[0088] According to the carbon price forecast results, the scenario analysis method is used to set different carbon price scenarios, and different carbon price scenarios correspond to different carbon price levels; the carbon price scenarios include optimistic scenarios, neutral scenarios and pessimistic scenarios;

[0089] Evaluate the economic value of emissions reductions for each transmission line under different carbon price scenarios;

[0090] Calculating the dispatching cost of each transmission line and calculating the economic feasibility evaluation value of the dispatching plan of the transmission line, wherein the economic feasibility evaluation value includes the financial net present value and the internal rate of return;

[0091] By comparing and analyzing the economic value of emission reduction and dispatching costs, the transmission lines for priority dispatch are determined.

[0092] In this embodiment, the emission reduction potential of each transmission line after scheduling is obtained, and at the same time, a carbon price prediction model is established through the historical data of the carbon emission trading market and time series prediction models such as ARIMA and Prophet. The input features of the carbon price prediction model may include key factors affecting carbon prices, such as the supply and demand relationship of carbon emission quotas, the level of economic development, and the energy structure. By training and verifying the carbon price prediction model, a carbon price prediction model with higher prediction accuracy is selected to obtain the trend and fluctuation range of carbon prices in the future.

[0093] According to the carbon price forecast results, the scenario analysis method is used to set up three scenarios: optimistic, neutral, and pessimistic, corresponding to different carbon price levels. When setting carbon price scenarios, you can refer to the price levels and trends of domestic and foreign carbon markets, as well as the carbon emission management policies and emission reduction targets in the region. For example, you can set three scenarios: baseline scenario (current price), optimistic scenario (price increase of 20%), and pessimistic scenario (price decrease of 20%). Estimate the emission reduction benefits of transmission line upgrades separately, and evaluate the economic value of emission reduction for each line under different carbon price scenarios.

[0094] Calculate the dispatching cost of each transmission line, including the initial investment cost, operation and maintenance cost, and financial cost. Calculate the financial net present value (NPV) and internal rate of return (IRR) of the dispatched transmission line by cash flow discounting method to judge the economic feasibility of the dispatching plan. Specifically, when selecting the discount rate, you can refer to the average capital cost (WACC) of the power grid industry or the benchmark interest rate. You can choose a 5% discount rate to discount the cash flow of the upgrade project and calculate the net present value and internal rate of return. At the same time, you can perform a sensitivity analysis on the discount rate to test the impact of changes in the discount rate on the financial indicators of the project. Compare and analyze the economic value of emission reduction and dispatching cost, and use indicators such as benefit-cost ratio (BCR) to evaluate the economic rationality of dispatching each transmission line. When BCR is greater than 1, it indicates that the upgrade plan is economically feasible. The higher the BCR, the greater the economic return. Considering factors such as emission reduction benefits, economic returns, and upgrade urgency, prioritize the dispatching of transmission lines, determine the dispatching schedule and resource arrangements, and give priority to dispatching transmission lines with great emission reduction potential, high economic returns, and low technical risks.

[0095] In this embodiment, it also includes sensitivity analysis and risk assessment of the transmission lines with priority scheduling. Specifically, key parameters that affect the scheduling benefits of transmission lines, such as carbon price, electricity price, project cost, operation and maintenance cost, etc., are obtained. Then, on the basis of the benchmark scheme, each key parameter is changed by a certain positive or negative ratio (such as ±10%) to calculate the change in scheduling benefits. Through sensitivity analysis, the key parameters that have the greatest impact on scheduling benefits are obtained, and the prediction and estimation of related parameters are further refined. On the basis of sensitivity analysis, scenario analysis can be used to set three scenarios of optimism, neutrality and pessimism to comprehensively evaluate the impact of changes in external conditions on scheduling benefits. Through sensitivity analysis and scenario analysis, the main risk factors affecting transmission line scheduling are obtained, and corresponding risk response strategies are formulated, such as locking carbon prices through contracts and hedging electricity price risks. The scheduling of transmission lines is monitored and evaluated, the progress and actual benefits of the implementation of transmission line scheduling are timely adjusted and optimized, the scheduling plan and management process are continuously improved, the emission reduction efficiency and economic value of transmission lines are continuously improved, and support is provided for the low-carbon transformation and sustainable development of power grid companies.

[0096] Specifically, when obtaining the emission reduction potential of transmission lines, online monitoring of energy consumption is carried out, and parameters such as current, voltage, and power before and after the transmission line is dispatched are collected and recorded every 15 minutes. The reduction in the transmission loss rate after the transmission line is dispatched is calculated through data analysis, and then the carbon dioxide emission reduction corresponding to the reduction in power loss per kilowatt-hour is estimated based on the average power supply coal consumption and emission factor of the power grid. For example, the annual average line loss rate of a 110 kV line was 5% before the transformation, and the line loss rate dropped to 3% after the transformation. The annual transmission capacity is 500 million kWh, the average power supply coal consumption of the power grid is 320 grams / kilowatt-hour, and the carbon dioxide emission factor is 0.8 tons / ton of standard coal. After the transformation of the line, the power loss can be reduced by 10 million kWh per year, which is equivalent to reducing carbon dioxide emissions by 2,560 tons. At the same time, the Prophet model is used to predict carbon prices. Taking the carbon market price data of the past three years as samples, considering factors such as economic growth rate, coal prices, and renewable energy installed capacity, the trend of carbon prices in the next five years is predicted, and three scenarios are set: optimistic, neutral, and pessimistic, with carbon prices of 80 yuan, 60 yuan, and 40 yuan per ton, respectively. According to the carbon price and annual emission reduction under different scenarios, it can be calculated that the annual emission reduction benefits of line transformation are 205,000 yuan, 154,000 yuan, and 102,000 yuan, respectively. When evaluating the economic feasibility of line upgrades, the cash flow discounting method is used. Assuming that the investment in upgrading and transformation is 10 million yuan, the operation and maintenance cost is 1.2 times that before the transformation, according to a discount rate of 5% and an operating period of 10 years, the financial net present value of the project is calculated to be 6.2 million yuan, the internal rate of return is 8%, and the investment payback period is 6.5 years. Through the benefit-cost ratio analysis, the BCR of the line upgrade is 1.62, which is greater than 1, so it is economically feasible. Through sensitivity analysis, when the parameters such as carbon price, electricity price and investment cost changed by ±10%, the change rates of the project's financial net present value were ±15%, ±8% and ±12% respectively, indicating that carbon price and investment cost are key factors affecting project benefits and require special attention and control.

[0097] S700: Generate a carbon emission quota allocation plan based on the evaluation results and the carbon emission data of different transmission lines and target areas.

[0098] In one embodiment, the step S700 includes:

[0099] Collect and organize carbon emission data through a data collection system, including historical emissions, emission intensity and emission reduction potential;

[0100] Based on the carbon emission data, the K-means clustering algorithm is used to divide the carbon emission areas, and the Apriori association analysis algorithm is used to obtain the key factors affecting carbon emissions;

[0101] Through data visualization technology, we can obtain the regional distribution of carbon emissions and display data of key factors;

[0102] A multi-objective optimization model is established based on influencing factors, including historical carbon emission levels, emission reduction potential, technological level, and economic development;

[0103] A heuristic optimization algorithm is used to solve the multi-objective optimization model and generate a carbon emission quota allocation plan.

[0104] In this embodiment, based on the carbon emission data of each transmission line and target area, including historical emissions, emission intensity, emission reduction potential, etc., the data acquisition system automatically collects and organizes carbon emission related data to form complete carbon emission data, providing basic data support for quota allocation. Statistical analysis and mining of carbon emission data are performed, and the K-means clustering algorithm is used to group transmission lines and target areas according to indicators such as carbon emission intensity and emission reduction potential, and identify key areas and transmission lines with high emissions and low emission reduction. The Apriori association analysis algorithm is used to mine the association rules between carbon emissions and factors such as line aging and equipment failure, and find out the key factors affecting carbon emissions. Then, through data visualization techniques such as heat maps and scatter plots, the regional distribution and key factors of carbon emissions are intuitively displayed to provide a decision-making basis for quota allocation.

[0105] In the process of establishing the carbon emission quota allocation model, the historical emission level, emission reduction potential, technological level, economic development and other influencing factors are comprehensively considered to establish a multi-objective optimization model. Among them, the objective function of the target optimization model includes minimizing quota allocation and maximizing quota allocation. Minimizing the Gini coefficient of quota allocation is used to measure the fairness of quota allocation, and the adjustment range is used to measure the feasibility of quota allocation; maximizing quota allocation is the coverage of emission reduction potential, which is used to measure the effectiveness of quota allocation. Constraints include that the total quota does not exceed the total emission reduction target, the quota amount of each line and region does not exceed its emission reduction potential, and the quota adjustment range does not exceed a certain proportion. Heuristic optimization algorithms such as genetic algorithms and particle swarm algorithms are used to solve the multi-objective optimization model to obtain the Pareto optimal solution set, and then according to the decision-making preferences of decision makers, the minimum quota allocation or the maximum quota allocation is selected to determine the final carbon emission quota allocation plan.

[0106] In addition, it also includes dynamic adjustment of the carbon emission quota allocation plan, and regular evaluation and adjustment of quota allocation according to the actual carbon emission performance of the transmission lines and target areas. Specifically, an evaluation index system for quota adjustment is established, including emission reduction effects, cost-effectiveness, technological progress and other aspects, to conduct a comprehensive evaluation of the quota implementation of each transmission line and target area. Set the trigger threshold for quota adjustment, such as emission reduction 20% lower than expected, maintenance and upgrade investment less than 50% of the plan, etc., to trigger the quota adjustment process. The quota implementation is evaluated annually by rolling adjustment, and the carbon emission quota adjustment plan for the next year is determined based on the evaluation results. For transmission lines and regions with significant emission reduction effects and active maintenance and upgrading, the quota reward is appropriately increased; for transmission lines and regions with ineffective emission reduction and serious equipment aging, the quota allocation is appropriately reduced to form a benign incentive mechanism and achieve dynamic optimization of quota allocation.

[0107] In addition, a quota trading market can be developed to allow quota trading between different transmission lines and regions, and optimize the allocation of quota resources through market-based means. For example, a two-way auction trading mechanism is adopted, and buyers and sellers submit quota purchase and sale declarations, and the trading system matches transactions based on the principle of price priority and time priority. Set the price range and quantity upper limit of quota transactions to prevent illegal behaviors such as malicious speculation and price manipulation. Establish an information disclosure system for quota transactions, regularly publish information such as transaction prices and transaction volumes, ensure the openness and transparency of transactions, and guide quotas to flow to transmission lines and regions with great emission reduction potential and good economic benefits, so as to achieve efficient utilization of carbon emission reduction resources and achieve overall emission reduction goals.

[0108] In addition, we will strengthen the monitoring, evaluation and performance assessment of quota allocation, establish a scientific evaluation indicator system, use machine learning algorithms such as support vector machines and random forests, establish a carbon emission monitoring model, and conduct real-time monitoring of quota implementation and abnormal warning.

[0109] Specifically, when obtaining carbon emission data, smart meters and environmental sensors can be used to collect data such as current, voltage, power and temperature of transmission lines every 15 minutes, and the carbon emission intensity of each transmission line can be calculated at the same time. The data is uploaded to the database through the MQTT protocol. The K-means algorithm is used to perform cluster analysis on carbon emission intensity. When K is 3, the lines can be divided into three emission levels: high, medium and low, of which high emission lines account for 15%, medium emission lines account for 35%, and low emission lines account for 50%. The Apriori algorithm is used to mine association rules with line aging degree and equipment failure rate as antecedents and carbon emission intensity as the consequent. It is found that when the line aging degree exceeds 20 years and the equipment failure rate exceeds 5%, the probability of carbon emission intensity exceeding 100 tons / km reaches 80%.

[0110] In the multi-objective optimization model of the carbon emission quota allocation model, the Gini coefficient, emission reduction coverage and adjustment range are used as the objective function, and the total quota, emission reduction potential and adjustment upper limit are used as constraints. The NSGA-II genetic algorithm is used to solve it. After 500 generations of evolution, 30 non-dominated solutions are obtained. Then, according to the decision maker's preference, the solution with the smallest Gini coefficient is selected as the final carbon emission quota allocation plan. When dynamically adjusting the quota, six evaluation indicators such as emission reduction and upgrade investment are set, and a weight coefficient is set for each indicator. The weighted average method is used to calculate the comprehensive score. When the comprehensive score is lower than 0.6, the quota adjustment is triggered. The adjustment range is proportional to the negative deviation of the score, and it is adjusted dynamically once a year. In the quota trading market, a continuous two-way auction method is adopted. Buyers and sellers submit applications every day, and the energy management system matches once an hour. The matching rules are the highest bid priority and the earliest application priority. The price of a single transaction shall not exceed ±10% of the average price of the previous day, and the cumulative daily trading volume of a single account shall not exceed 1 million tons. The support vector machine algorithm is used to establish a quota monitoring model, which takes the real-time carbon emission data of the line as input features, compares the quota allocation curve, and identifies abnormal deviation behaviors. The model accuracy can reach more than 95%.

[0111] In one embodiment, the present application further includes:

[0112] Collecting historical electricity consumption data of the target area, wherein the historical electricity consumption data includes total electricity consumption and power distribution structure of the power grid;

[0113] Based on the historical electricity consumption data, establish an electricity consumption prediction model;

[0114] Generate a prediction result of future electricity consumption in the target area according to the electricity consumption prediction model;

[0115] Based on the prediction results, a power production and dispatch plan is generated.

[0116] In this embodiment, the results of the power consumption forecast can provide important input and basis for the optimization of power grid operation. By accurately predicting the power consumption in the future, the power grid dispatching department can formulate power production and dispatching plans in advance, reasonably arrange the output and spare capacity of the generator set, and optimize the operation mode of the power grid. At the same time, combined with the power consumption forecast results and the real-time operation status of the power grid, it is possible to predict the power supply bottlenecks and potential risks that may appear in the future, and take measures in advance, such as adjusting the load distribution, arranging planned maintenance, etc., to improve the power supply reliability and stability of the power grid. In addition, the power consumption forecast can also provide a reference for transactions in the power market and help power companies to reasonably arrange power purchase and sales plans and electricity price strategies. Taking into account the power consumption forecast results, the operation status of the power grid and other factors, the operation mode and resource allocation of the power grid are optimized, the power production and dispatching plan is reasonably arranged, the power supply reliability and economy of the power grid are improved, and the power demand of the target area is guaranteed to be met.

[0117] Specifically, the historical electricity consumption data of the target area is obtained through the power company's metering system and database. The historical electricity consumption data includes the total electricity consumption and the electricity consumption distribution data of each substation and distribution line. The data granularity should reach the hourly level, and the time span should cover at least 3 years.

[0118] After that, the acquired historical electricity consumption data is cleaned and preprocessed. Specifically, outliers and missing values ​​are removed. For time periods with more missing values, interpolation or regression analysis is used to supplement the data to ensure the continuity and integrity of the data. At the same time, descriptive statistical analysis methods are used to calculate statistical indicators such as the mean, variance, and peak of electricity consumption data. By drawing the time series diagram and box-and-whisker diagram of electricity consumption, the distribution characteristics and outliers of electricity consumption are intuitively analyzed. The time series decomposition technology of additive decomposition or multiplicative decomposition is used to decompose the electricity consumption time series into components such as long-term trends, seasonal factors, cyclical factors, and random noise. Among them, the long-term trend can be extracted using the moving average method. By setting different window sizes, such as a 12-month sliding window, the electricity consumption data is smoothed, short-term fluctuations are removed, and long-term trends are obtained. Seasonal factors can be extracted by calculating the average electricity consumption in each season and the difference between it and the total average electricity consumption. Cyclic factors can be extracted using methods such as Fourier transform or wavelet analysis.

[0119] After cleaning and preprocessing the historical electricity consumption data, select an appropriate time series prediction model, such as an ARIMA model, a seasonal ARIMA model, a Prophet model, etc., to establish an electricity consumption prediction model, and use the historical electricity consumption data to train and optimize the model. Specifically, in one embodiment of the present application, a Prophet model is used to establish an electricity consumption prediction model, and the Prophet model decomposes the time series into three components: a trend term, a seasonal term, and a holiday term. Among them, the trend term is used to capture the long-term trend of the time series, which can be linear or nonlinear; the seasonal term is used to capture the periodic changes of the time series, such as weekly, monthly, and annual cycles; the holiday term is used to capture the impact of special events (such as holidays) on the time series. By modeling and fitting these three components, the Prophet model can predict future time series, generate an electricity consumption prediction model, apply the established electricity consumption prediction model to the future electricity consumption prediction of the target area, generate daily, monthly, and annual electricity consumption prediction results, and compare and verify with the actual electricity consumption data. The electricity consumption forecasting model is continuously adjusted and improved. Specifically, more features related to electricity consumption, such as weather factors and economic indicators, are introduced to enrich the input features of the model and improve the forecast accuracy. In addition, multiple different forecasting algorithms, such as ARIMA, Prophet, neural network, etc., are used to combine the forecast results of multiple models by weighted average or voting to improve the robustness and accuracy of the forecast. The hyperparameters of the electricity consumption forecasting model are tuned, such as the changepoint_prior_scale, holidays_prior_scale and other parameters in the Prophet model, and the optimal parameter combination is found through grid search or Bayesian optimization. The rolling forecast method is used to regularly add new actual electricity consumption data to the training set, update the electricity consumption forecasting model parameters, and use the electricity forecasting model to adapt to the latest changes in electricity consumption trends.

[0120] Specifically, when obtaining historical electricity consumption data in the target area, we can cooperate with the power company and use the API interface to automatically extract data from its metering and database. The data granularity can reach 15 minutes, and the time span covers more than 5 years. The acquired data is cleaned and preprocessed, and the box plot method is used to identify outliers. Data points that exceed the first quartile and the third quartile by more than 1.5 times are regarded as outliers and removed. For missing values, linear interpolation can be used to supplement them. For missing values ​​that exceed 24 hours continuously, the average value of the same moment of the previous and next 7 days is used to fill them. The Pandas library of Python is used for data processing and analysis, and statistical indicators such as the daily average, weekly average, and monthly average of electricity consumption are calculated. Time series diagrams and box plots are drawn for visual analysis. The STL (Seasonal-Trend decomposition using LOESS) algorithm is used to decompose the electricity consumption time series, setting the seasonal cycle to 7 (days) and 365 (years) to extract long-term trends, seasonal factors, and random noise. The Prophet library is used to establish an electricity consumption prediction model, and the data is divided into a training set and a test set, with the training set accounting for 80% and the test set accounting for 20%. The parameters of the Prophet model are set, such as the growth mode is logistic growth, the seasonal factor is multiplicative seasonality, and the holiday effect considers major holidays such as the Spring Festival and National Day. During the training process of the electricity consumption prediction model, the grid search method is used to optimize the hyperparameters, such as changepoint_prior_scale is searched in [0.01, 0.05, 0.1, 0.5], and holidays_prior_scale is searched in [0.1, 1, 10], and the parameter combination with the smallest error on the validation set is selected. By traversing these preset combinations, each set of hyperparameter configurations is evaluated on the validation set, and the evaluation indicator is the prediction error. Finally, the set of hyperparameter combinations that leads to the smallest prediction error on the validation set is selected to ensure that the generalization ability and prediction accuracy of the electricity consumption prediction model on unknown data are optimal. The trained electricity consumption prediction model is applied to the test set to evaluate the prediction effect and calculate indicators such as mean absolute percentage error (MAPE) and root mean square error (RMSE). If MAPE is less than 5%, the prediction effect of the electricity consumption prediction model is considered to be good and can be applied to actual scenarios; if MAPE is greater than 10%, the electricity consumption prediction model needs to be further optimized. According to the prediction results of the electricity consumption prediction model and combined with the real-time operation data of the power grid, the mixed integer programming (MIP) algorithm is used to solve the power production and dispatching plan. The objective function is to minimize the power generation cost and the power purchase cost. The constraints include power balance, spare capacity requirements, upper and lower limits of unit output, etc., to obtain the optimal power generation plan and power grid operation plan for each time period in the next week.

[0121] The embodiments described above 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 aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for calculating power grid carbon emissions and allocating quotas from the perspective of carbon emissions trading, characterized in that: The method comprises: Acquire transmission line data of the target area power grid, and calculate the transmission loss rate of different transmission lines according to the transmission line data, wherein the transmission line data of the target area power grid includes line length, transmission capacity, load fluctuation and ambient temperature information; Acquire historical output data and real-time environmental data of each clean energy source, and establish a clean energy output prediction model based on the historical output data and the real-time environmental data, wherein the clean energy output prediction model is used to predict the total output of each clean energy source in the next cycle, wherein the clean energy sources include solar energy, wind energy, and hydropower; Based on the predicted total output of clean energy and the transmission loss rates of different transmission lines, the emission reduction of clean energy on each transmission line is obtained, and based on the emission reduction, the emission reduction potential of the target area power grid under the preset clean energy scale is predicted, and the emission reduction potential is the ability to reduce the transmission loss rate and improve the utilization rate of clean energy; Generate a transmission line dispatch plan based on the transmission loss rates of different transmission lines and asset data information of different transmission lines; Based on the dispatch plan of the transmission line, a deep reinforcement learning algorithm is used to establish an intelligent dispatch model to generate an optimal clean energy scale, where the optimal clean energy scale is a clean energy scale that minimizes the transmission loss rate and improves the clean energy utilization rate; Based on the emission reduction potential under the optimal clean energy scale and the historical carbon price data of the carbon emission trading market, generate the evaluation results of the emission reduction effects of different transmission lines based on the market economic value, and determine the transmission lines for priority scheduling; Generate a carbon emission quota allocation plan based on the assessment results and carbon emission data of different transmission lines and target areas; Wherein, the carbon emission quota allocation scheme is generated based on the assessment results and the carbon emission data of different transmission lines and target areas, including: Collect and organize carbon emission data through a data collection system, including historical emissions, emission intensity and emission reduction potential; Based on the carbon emission data, the K-means clustering algorithm is used to divide the carbon emission areas, and the Apriori association analysis algorithm is used to obtain the key factors affecting carbon emissions; Through data visualization technology, we can obtain the regional distribution of carbon emissions and display data of key factors; A multi-objective optimization model is established based on influencing factors, including historical carbon emission levels, emission reduction potential, technological level, and economic development; A heuristic optimization algorithm is used to solve the multi-objective optimization model and generate a carbon emission quota allocation plan.

2. The method according to claim 1, characterized in that The acquiring of the transmission line data of the target area power grid and calculating the transmission loss rates of different transmission lines according to the transmission line data specifically includes: A data acquisition system is used to monitor and record the transmission line data in real time, wherein the transmission line data includes line length, transmission capacity, load fluctuation and ambient temperature information; Cleaning and preprocessing the transmission line data to generate usable transmission line data; Based on the available transmission line data, a transmission loss rate calculation model is established using a power flow calculation method; According to the transmission loss rate calculation model, the current distribution and power loss of the transmission line are calculated to obtain the actual transmission loss rate of the transmission line; Based on the historical operation data and environmental data of the transmission lines, a transmission loss rate prediction model is established using a machine learning algorithm; The transmission loss rate of the transmission line is predicted according to the transmission loss rate prediction model.

3. The method according to claim 1, characterized in that: The obtaining of historical output data and real-time environmental data of each clean energy source, and establishing a clean energy output prediction model based on the historical output data and the real-time environmental data, wherein the clean energy output model is used to predict the total output of each clean energy source in the next cycle, wherein the clean energy source includes solar energy, wind energy and hydropower, including: Collect real-time environmental data through IoT sensors, including sunlight intensity, wind speed and direction, and water flow speed; Obtaining historical output data of clean energy through various clean energy power generation facilities, wherein the historical output data is clean energy power generation data; Matching and aligning the real-time environmental data and the historical production data to establish a time series data set, wherein the time series data set includes a training set, a validation set, and a test set; Establishing a clean energy output prediction model, and training and optimizing the clean energy output prediction model through the time series data set; The total output of each clean energy in the next cycle is predicted by the clean energy output prediction model; Within the preset time, the historical output data and real-time environmental data of each clean energy are updated, and the clean energy output prediction model is updated and optimized.

4. The method according to claim 1, characterized in that: The emission reduction of clean energy on each transmission line is obtained based on the predicted total output of clean energy and the transmission loss rate of different transmission lines, including: allocating the predicted total clean energy output according to a transmission line allocation method of a target regional power grid; According to the transmission loss rate of the transmission line, determine the amount of clean energy lost during the transmission process, and obtain the actual amount of clean energy used on each transmission line; Based on the actual electricity consumption of clean energy, the carbon emission factor method is used to calculate the emission reduction of clean energy on each transmission line.

5. The method according to claim 4, characterized in that The emission reduction potential of the target regional power grid is predicted based on the emission reduction amount under the preset clean energy scale, including: Based on the emission reduction of different transmission lines, the overall emission reduction of the target area power grid is determined through data visualization technology; Using scenario analysis method, different clean energy development scenarios are set, and different clean energy development scenarios correspond to different clean energy scales. The clean energy development scenarios include baseline scenario, positive scenario and radical scenario; For different clean energy development scenarios, based on the predicted total clean energy output and the predicted total social electricity consumption, the emission reduction potential of the target area power grid is predicted under the preset clean energy scale.

6. The method according to claim 1, characterized in that: The generating of a transmission line dispatching plan based on the transmission loss rates of different transmission lines and the asset data information of different transmission lines comprises: Obtain asset data information of different transmission lines, the asset data information including construction time, design parameters, operating status and historical fault data of each line; Based on the electrical parameters, mechanical parameters and environmental parameters of the transmission line, the health index of the transmission line is calculated using the fuzzy comprehensive evaluation method; Based on the health index and historical fault data, a remaining life prediction model of the transmission line is established, wherein the remaining life prediction model is used to predict the remaining service life of the transmission line; Establishing a multi-classification model of the state of the transmission line, wherein the multi-classification model is used to generate potential failure risks of the transmission line; Based on the remaining service life and potential failure risk of the transmission line, a dispatch plan for the transmission line is generated, wherein the dispatch plan includes a maintenance and upgrade plan for the transmission line.

7. The method according to claim 6, characterized in that: The generating of a transmission line dispatch plan based on the transmission loss rates of different transmission lines and the asset data information of different transmission lines also includes: According to the emission reduction potential and dispatching cost of each transmission line, a project management system is established and quantitative key performance indicators are formulated, including transmission line availability, transmission loss rate, clean energy utilization rate, and return on investment; Use the balanced scorecard to establish a performance evaluation system for the key performance indicators; Based on the project management system and performance evaluation system, a performance evaluation is conducted on the dispatched transmission lines to evaluate the cost-effectiveness and implementation value of the transmission line dispatch.

8. The method according to claim 1, characterized in that The emission reduction potential under the optimal clean energy scale and the historical carbon price data of the carbon emission trading market are used to generate the evaluation results of the emission reduction effects of different transmission lines based on the market economic value, and determine the transmission lines for priority scheduling, including: Based on the emission reduction potential of each transmission line after scheduling and the historical carbon price data of the carbon emission trading market, a time series prediction model is used to establish a carbon price prediction model, and the carbon price prediction model is used to generate a carbon price prediction result; According to the carbon price forecast results, the scenario analysis method is used to set different carbon price scenarios, which correspond to different carbon price levels. The carbon price scenarios include optimistic scenarios, neutral scenarios and pessimistic scenarios; Evaluate the economic value of emissions reductions for each transmission line under different carbon price scenarios; Calculating the dispatching cost of each transmission line and calculating the economic feasibility evaluation value of the dispatching plan of the transmission line, wherein the economic feasibility evaluation value includes the financial net present value and the internal rate of return; By comparing and analyzing the economic value of emission reduction and dispatching costs, the transmission lines for priority dispatch are determined.

9. The method according to claim 1, characterized in that: The method further comprises: Collecting historical electricity consumption data of the target area, wherein the historical electricity consumption data includes total electricity consumption and power distribution structure of the power grid; Based on the historical electricity consumption data, establish an electricity consumption prediction model; Generate a prediction result of future electricity consumption in the target area according to the electricity consumption prediction model; Based on the prediction results, a power production and dispatch plan is generated.

Images