Electricity and carbon transaction risk management method and system based on machine learning

Abstract

The invention discloses an electricity-carbon transaction risk management method and system based on machine learning, and relates to the technical field of electricity-carbon transaction risk management, and the method comprises the following steps: obtaining a carbon quota initial value and historical data, carrying out the feature extraction to obtain historical standard carbon emission, and judging whether there is a transaction carbon quota based on the carbon quota initial value; if the carbon quota of the transaction exists, obtaining power carbon demand influence data and emission reduction capability influence data in a plurality of regions; constructing an electricity-carbon transaction risk assessment model based on machine learning by using the electricity-carbon demand influence data and the emission reduction capability influence data; and comparing the input of the electricity-carbon transaction risk assessment model with preset electricity-carbon transaction stability data, dividing the risks of the electricity-carbon transaction into high risk degrees, medium risk degrees and low risk degrees, and making corresponding adjustments according to the three risk degrees. According to the invention, the electricity-carbon transaction risk is judged through machine learning, so that the problem that the market lacks the flexible adjustment capability for specific market conditions is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of electricity-carbon trading risk management, and more specifically, to the technical field of electricity-carbon trading risk management methods and systems based on machine learning. Background Art

[0002] As the global need to address climate change increases, carbon emissions trading has become an important means of reducing greenhouse gas emissions. The electricity carbon market provides companies with a flexible carbon emissions control mechanism by converting carbon emission quotas into tradable assets. Risk management in the electricity carbon market relies on a combination of technology and tools, aiming to improve market transparency, enhance market stability, and help companies achieve maximum benefits in carbon trading.

[0003] Traditional risk management technologies mostly use fixed hedging strategies or trading rules, lack the ability to flexibly adjust to specific market conditions, and lack transparent real-time data to judge the degree of risk, which makes it difficult for traders to accurately grasp market dynamics. In view of the above problems, the present invention proposes a solution. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an electricity carbon trading risk management method and system based on machine learning, which judges the electricity carbon trading risks through machine learning to solve the problem of lack of flexible adjustment capabilities for specific market conditions in the market.

[0005] To achieve the above object, the present invention provides the following technical solutions: The method for risk management of electric carbon trading based on machine learning includes the following steps: obtaining an initial value and historical data of carbon quotas, performing feature extraction on the historical data to obtain historical standard carbon emissions, and judging whether there are carbon quotas to be traded based on the initial value of carbon quotas according to the historical standard carbon emissions; if there are carbon quotas to be traded, obtaining electric carbon demand impact data and emission reduction capacity impact data in several regions; training the electric carbon demand impact data and emission reduction capacity impact data based on machine learning to determine the degree of impact on electric carbon trading risks, and constructing an electric carbon trading risk assessment model; comparing the input of the electric carbon trading risk assessment model with preset electric carbon trading stability data, and dividing the risks of electric carbon trading into three risk levels of high, medium and low according to the comparison results, and making corresponding adjustments according to the three risk levels.

[0006] In a preferred embodiment, the feature extraction is performed on the historical data to obtain the carbon quota demand value, and the judgment on whether there is a carbon quota to be traded is made based on the carbon quota initial value according to the carbon quota demand value. The specific steps are as follows: historical data is obtained, and the historical data is preprocessed and feature extracted to obtain the media and equipment that affect the carbon emissions; data analysis is performed on the media and equipment that affect the carbon emissions to obtain the total carbon emissions, and the carbon quota demand value is obtained based on the total carbon emissions; the carbon quota demand value is judged based on the carbon quota initial value; if the carbon quota demand value is greater than or equal to the carbon quota initial value, there is no carbon quota to be traded; if the carbon quota demand value is less than the carbon quota initial value, there is a carbon quota amount that can be traded.

[0007] In a preferred embodiment, the electricity carbon demand impact data includes an electricity carbon demand impact coefficient and an electricity carbon trading fluctuation impact coefficient. The electricity carbon demand impact coefficient is specifically obtained as follows: obtain initial carbon quota data and electricity carbon demand data of several regions; draw a curve chart based on the initial carbon quota data and the electricity carbon demand data; use the initial carbon quota data as the horizontal axis and the electricity carbon demand data as the vertical axis, and perform data analysis based on the curve chart to obtain the electricity carbon demand fluctuation coefficient; obtain carbon quota introduction data, market carbon quota fluctuation data and regional carbon emission factors; calculate the electricity carbon consumption impact factor based on the carbon quota introduction data, market carbon quota fluctuation data and regional carbon emission factors based on a preset electricity carbon consumption impact formula; calculate the electricity carbon demand impact coefficient based on the electricity carbon consumption impact factor combined with the electricity carbon demand fluctuation coefficient based on the preset electricity carbon demand impact formula.

[0008] In a preferred embodiment, the electricity carbon trading fluctuation impact coefficient is specifically obtained by the following method: obtaining electricity carbon price data of several trading days; calculating the average electricity carbon price by averaging the electricity carbon price data of the several trading days; obtaining first electricity carbon trading fluctuation data, the first data including the total trading days, electricity carbon trading volume and the average carbon price in the carbon market; calculating the electricity carbon trading fluctuation impact coefficient based on a preset electricity carbon trading fluctuation impact formula according to the first electricity carbon trading fluctuation data combined with the average electricity carbon price.

[0009] In a preferred embodiment, the emission reduction capacity impact data is specifically acquired by the following method: acquiring emission reduction capacity change data of several regions; obtaining the maximum potential coefficient of emission reduction capacity and the time span of emission reduction capacity change based on regression analysis of the emission reduction capacity change data; obtaining the rate coefficient of emission reduction capacity change by monitoring the efficiency of emission reduction capacity of several regions within a period of L; and calculating the emission reduction capacity impact data based on a preset emission reduction capacity impact formula according to the emission reduction capacity change data combined with the rate coefficient of emission reduction capacity change.

[0010] In a preferred embodiment, the electricity carbon demand impact data and the emission reduction capacity impact data are trained based on machine learning to determine the degree of impact on electricity carbon trading risks, and an electricity carbon trading risk assessment model is constructed. The specific steps are as follows: historical electricity carbon demand impact data and historical emission reduction capacity impact data are obtained, and the data are preprocessed; the processed data is divided into a training set and a test set, and the training set is used based on a machine learning algorithm to train the degree of impact on electricity carbon trading risks through historical electricity carbon demand impact data and historical emission reduction capacity impact data, and based on automatic adjustment, a weight coefficient for electricity carbon trading risks is obtained to construct an electricity carbon trading risk assessment model; the test set is used to test and optimize the electricity carbon trading risk assessment model to obtain a final electricity carbon trading risk assessment model, and based on the final electricity carbon trading risk assessment model, an electricity carbon trading risk assessment coefficient is generated.

[0011] In a preferred embodiment, the input of the electricity carbon trading risk assessment model is compared with the preset electricity carbon trading stability data, and the risk of electricity carbon trading is divided into three risk levels of high, medium and low according to the comparison results, and corresponding adjustments are made according to the three risk levels. The specific steps are as follows: obtaining the electricity carbon trading risk assessment coefficient through the input of the electricity carbon trading risk assessment model; comparing and analyzing the electricity carbon trading risk assessment coefficient with the preset electricity carbon trading stability data, and making a judgment based on the comparison and analysis results; if the electricity carbon trading risk assessment coefficient is less than the preset electricity carbon trading stability data, it is a low level of electricity carbon trading risk; if the electricity carbon trading risk assessment coefficient is equal to the preset electricity carbon trading stability data, it is a medium level of electricity carbon trading risk; if the electricity carbon trading risk assessment coefficient is greater than the preset electricity carbon trading stability data, it is a high level of electricity carbon trading risk.

[0012] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. The technical advantage of using machine learning to build a risk model to judge the risks of electricity-carbon trading is that it can efficiently process massive data, identify complex nonlinear relationships, and dynamically adjust the prediction model, thereby improving the accuracy and real-time performance of risk prediction.

[0013] 2. By obtaining data on the impact of electricity carbon demand and emission reduction capacity to make risk assessments, we can comprehensively capture changes in market supply and demand and changing factors, and improve the accuracy and flexibility of risk assessments. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A schematic diagram of the structure of the electric carbon trading risk management method based on machine learning provided in an embodiment of the present application.

[0015] Figure 2A schematic diagram of the structure of an electricity carbon trading risk management system based on machine learning provided in an embodiment of the present application. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] Embodiment 1, Figure 1 The schematic diagram of the structure of the electric carbon trading risk management method based on machine learning provided in the embodiment of the present application includes the following steps: S1, obtaining the initial value of the carbon quota and historical data, extracting features from the historical data to obtain the historical standard carbon emissions, and judging whether there is a traded carbon quota based on the initial value of the carbon quota according to the historical standard carbon emissions.

[0018] In this example, the initial value of the carbon quota can be understood as the initial available carbon quota allocated to the enterprise at the beginning. The feature extraction of the historical data and the acquisition of the carbon quota demand value are performed, and the determination of whether there is a carbon quota to be traded is based on the initial value of the carbon quota according to the carbon quota demand value. The specific steps are as follows: Obtain historical data, pre-process and extract features from the historical data, and obtain the media and equipment that affect carbon emissions; Performing data analysis based on the media and equipment that affect carbon emissions to obtain the total amount of carbon emissions, and obtaining a carbon quota demand value based on the total amount of carbon emissions; Determining the carbon quota demand value based on the carbon quota initial value; If the carbon quota demand value is greater than or equal to the carbon quota initial value, there is no trading amount of carbon quota; If the carbon quota demand value is less than the carbon quota initial value, there is a carbon quota amount that can be traded.

[0019] S2: If there are carbon quotas to be traded, obtain the impact data on electricity carbon demand and emission reduction capacity in several regions.

[0020] In this example, the electricity carbon demand impact data includes the electricity carbon demand impact coefficient and the electricity carbon trading fluctuation impact coefficient. The electricity carbon demand impact coefficient refers to the demand for carbon quotas in the electricity price trading market. The electricity carbon demand impact coefficient reflects the demand changes caused by various factors in the carbon market demand side in a certain period of time in the future. Obtaining the electricity carbon demand impact coefficient has the following advantages for electricity carbon trading risk management: Accurately predict electricity-carbon market fluctuations: The electricity-carbon demand impact coefficient reflects the sensitivity of electricity and carbon emission demand. By obtaining these coefficients, market participants can better understand the relationship between electricity demand and carbon emissions, and then predict market price fluctuations. This helps to identify potential market risks and price fluctuations in advance, making it easier to formulate response strategies.

[0021] Risk assessment and quantitative management: The impact coefficient of electricity carbon demand can provide a quantitative indicator to help analyze the potential impact of changes in market demand on electricity carbon trading prices. By simulating changes in demand under different scenarios, companies can quantitatively assess risk exposure under different market conditions. This quantitative analysis method can help companies better formulate risk management strategies, such as adjusting carbon quotas and implementing price hedging.

[0022] Optimize carbon quota allocation and trading strategies: By understanding the impact of changes in electricity carbon demand on the carbon emissions trading market, companies can optimize their carbon quota purchase and sale strategies. If the demand impact coefficient indicates that electricity demand or carbon emissions are likely to increase in the future, companies can purchase carbon quotas in advance when prices are low to reduce future carbon emission costs. Conversely, if the demand coefficient predicts a decrease in demand, companies can choose to sell excess quotas.

[0023] Improve the scientific nature of long-term decision-making: The electricity carbon demand impact coefficient helps companies consider potential changes in the carbon emissions trading market in their long-term planning. Through in-depth analysis of the relationship between electricity demand changes and carbon emissions, companies can more scientifically formulate long-term development plans, such as carbon emission reduction measures, while avoiding losses caused by large fluctuations in market prices.

[0024] Optimize resource allocation: In electricity-carbon trading, enterprises need to dynamically adjust resource allocation according to changes in electricity-carbon demand. By obtaining the electricity-carbon demand impact coefficient, enterprises can more accurately assess the risks and opportunities in carbon trading, make more reasonable decisions in the purchase, storage and sale of carbon quotas, and avoid excessive waste of resources.

[0025] The specific method for obtaining the electric carbon demand impact coefficient is as follows: Obtain initial carbon quota data and electricity carbon demand data for several regions; Draw a curve graph according to the initial carbon quota data and the electricity carbon demand data; Using the initial carbon quota data as the horizontal axis and the electric carbon demand data as the vertical axis, data analysis is performed according to the curve chart to obtain the electric carbon demand fluctuation coefficient; Obtain carbon quota introduction data, market carbon quota fluctuation data and regional carbon emission factors; The carbon quota introduction data, market carbon quota fluctuation data and regional carbon emission factors are used to calculate the electricity carbon consumption impact factor based on the preset electricity carbon consumption impact formula; The electricity carbon demand impact coefficient is calculated based on the preset electricity carbon demand impact formula according to the electricity carbon consumption impact factor combined with the electricity carbon demand fluctuation coefficient.

[0026] The specific calculation formula of the electric carbon consumption influencing factor is as follows:

[0027] The specific calculation formula of the electric carbon demand impact coefficient is as follows:

[0028] In the formula, is the impact coefficient of electricity carbon demand, is the initial carbon quota data of the ith region, is the electricity carbon demand data of the ith region, is the fluctuation coefficient of electricity carbon demand, is the influencing factor of carbon consumption, Introduce data for the carbon quota of the i-th region, is the regional carbon emission factor of the i-th region, where i=1,2,3,...,R, R is an integer, N is a number of regions, It is the market carbon quota fluctuation data.

[0029] It should be noted that obtaining the initial carbon quota data of several regions can be understood as the allocation of carbon quotas in several regions in the initial stage. However, due to the excessive carbon emissions of enterprises, the carbon quota for processing carbon emissions cannot be met, and the carbon quota required for the excess carbon emissions needs to be traded in the electricity carbon trading market. When the electricity carbon demand in several regions is too high, the electricity carbon trading market will form an upward trend and be in an advantageous stage.

[0030] The impact coefficient of electricity-carbon trading volatility is one of the important factors affecting the risk management of electricity-carbon trading, which represents the instability of electricity-carbon market transactions. Obtaining the impact coefficient of electricity-carbon trading volatility has the following advantages for the risk management of electricity-carbon trading: Quantifying risk exposure: The volatility of the electricity carbon trading market may have a significant impact on companies, especially when the price of carbon emission quotas fluctuates greatly. By obtaining the volatility impact coefficient, it is possible to quantify the risk exposure of market fluctuations to specific assets or portfolios. This helps companies understand and manage the potential impact of price fluctuations in the carbon trading market on their operations and financial status.

[0031] Accurate risk assessment: The volatility impact coefficient can accurately reflect the risk level of the market, thereby helping managers assess the impact of market price fluctuations on asset value. For example, in electricity carbon trading, the volatility coefficient can help predict the magnitude of carbon quota price changes and help companies conduct more scientific risk assessments. In this way, companies can make more rational decisions and avoid unnecessary losses caused by price fluctuations.

[0032] Optimize risk hedging strategies: By identifying the volatility coefficient of the electricity carbon trading market, companies can design more effective risk hedging strategies. If companies have identified which factors have a greater impact on carbon price fluctuations, they can hedge through derivative tools. For example, if the volatility coefficient of carbon prices is high during a certain period, companies can lock in prices to avoid the risk of future price fluctuations.

[0033] Improve decision-making efficiency: Obtaining the volatility impact coefficient can provide an objective reference for decision makers, helping them make faster and more accurate decisions when dealing with market fluctuations. Through quantitative analysis of fluctuations in the electricity carbon trading market, companies can quickly identify potential risk points and opportunities, and take corresponding measures in market changes.

[0034] Dynamically adjust risk management strategies: The volatility of the electricity carbon trading market may change over time and with changes in the market environment. Real-time updates of the volatility impact coefficient can help companies dynamically adjust risk management strategies in different market environments. By continuously tracking and analyzing market fluctuations, companies can ensure that their risk management measures are always in line with the actual market situation.

[0035] The specific method for obtaining the electricity-carbon trading fluctuation impact coefficient is as follows: Obtain electricity carbon price data for several trading days; The average value of the electricity carbon price is calculated by averaging the electricity carbon price data of the several trading days; Obtaining first data on fluctuations in electricity-carbon trading, the first data including total trading days, electricity-carbon trading volume, and average carbon price in the carbon market; The electricity carbon trading fluctuation impact coefficient is calculated based on the first data of electricity carbon trading fluctuation and the average electricity carbon price based on the preset electricity carbon trading fluctuation impact formula.

[0036] The specific calculation formula of the electricity carbon trading fluctuation impact coefficient is as follows:

[0037] In the formula, is the fluctuation impact coefficient of electricity-carbon trading, is the total number of trading days, For the Electricity carbon price data for trading days, including: =1,2,3,...,R, R is an integer, is the average carbon price of electricity, is the volume of electricity carbon trading, is the average carbon price in the carbon market.

[0038] It should be noted that the preset formula for the impact of electricity-carbon trading fluctuations quantifies the volatility of the electricity-carbon market by calculating the standard deviation of the electricity-carbon price data, and adjusts the electricity-carbon trading fluctuation impact coefficient by the ratio of the electricity-carbon market transaction volume.

[0039] Emission reduction capacity impact data refers to the low-carbon emission reduction capacity of each region, that is, the reduction of carbon emissions through technological progress. The role of emission reduction capacity impact data in the electricity carbon trading market cannot be ignored, especially the progress of carbon capture and storage technology, which plays an important role in the risk assessment of electricity carbon trading.

[0040] Analyzing the impact data of emission reduction capacity has the following advantages for risk management of electricity carbon trading: Accurately assess emission reduction potential and market fluctuations: By analyzing emission reduction capabilities, the emission reduction potential of each entity can be better estimated. Understanding the emission reduction that a company can achieve in the future helps to assess the supply and demand relationship and price fluctuations in the market.

[0041] Optimize carbon quota allocation and market price formation: Through emission reduction capacity data, a more scientific and reasonable carbon quota allocation mechanism can be formulated. Different companies may have different emission reduction capabilities. If this difference is not fully reflected, it may lead to unfair quota allocation and further affect the stability and fairness of the market.

[0042] Improve transparency of carbon market: Analysis of emission reduction capacity can improve information transparency of carbon market and reduce information asymmetry among market participants. Through open and transparent emission reduction capacity data, market participants can better assess potential market risks and make more rational decisions. Transparent emission reduction data can help reduce speculation in the market and enhance market stability and credibility. In addition, emission reduction capacity impact data can be used to predict and adjust market prices. For example, when certain regions or enterprises have strong emission reduction capacity, it may lead to an increase in carbon supply, thereby affecting market prices.

[0043] Enhance risk early warning and dynamic regulation capabilities: Analyzing the impact data of emission reduction capacity will help establish a more accurate risk early warning system. Regularly tracking changes in emission reduction capacity can provide important dynamic information for risk management. Adjusting carbon trading and quota allocation methods, or taking intervention measures when the market is overheated or overcooled, can effectively mitigate the risks brought by market fluctuations.

[0044] In the carbon trading market, companies with stronger emission reduction capabilities can obtain more quotas or sell carbon credits, while companies with weaker emission reduction capabilities may face more cost pressure. Analyzing emission reduction data helps predict these changes and identify risks in advance.

[0045] The specific method for obtaining the emission reduction capacity impact data is as follows: Obtain data on changes in emission reduction capacity in several regions; Obtaining the maximum potential coefficient of emission reduction capacity and the time span of emission reduction capacity change based on regression analysis through the emission reduction capacity change data; By monitoring the efficiency of emission reduction capacity in several areas within a period of time L, the speed coefficient of emission reduction capacity change is obtained; The emission reduction capacity impact data is calculated based on a preset emission reduction capacity impact formula according to the emission reduction capacity change data combined with a speed coefficient of the emission reduction capacity change.

[0046] The emission reduction capacity affects the data, and the specific calculation formula is as follows:

[0047] In the formula, To reduce the impact of emission capacity, For the The emission reduction capacity change data of each region, among which: =1, 2, 3, ..., R, R is an integer, N is a number of regions, is the maximum potential coefficient, is the speed coefficient of emission reduction capacity change, is the time span over which the emission reduction capacity changes.

[0048] It should be noted that the preset emission reduction capacity impact formula takes into account the progress of emission reduction capacity and the impact of the maturity of emission reduction capacity on the electricity carbon trading market, and simulates the impact of the rate coefficient of change of emission reduction capacity on the electricity carbon trading market through exponential terms.

[0049] S3, based on machine learning, train the data on the impact of electricity-carbon demand and emission reduction capacity to determine the degree of impact on electricity-carbon trading risks, and build an electricity-carbon trading risk assessment model.

[0050] The specific steps of training the impact data of electricity carbon demand and emission reduction capacity on the risk of electricity carbon trading based on machine learning and building an electricity carbon trading risk assessment model are as follows: Obtain historical data on the impact of electricity carbon demand and historical data on the impact of emission reduction capacity, and pre-process the data; The processed data is divided into a training set and a test set. The training set is used based on a machine learning algorithm to train the impact degree of the electricity-carbon trading risk through the historical electricity-carbon demand impact data and the historical emission reduction capacity impact data. Based on automatic adjustment, the weight coefficient of the electricity-carbon trading risk is obtained to build an electricity-carbon trading risk assessment model. The electricity-carbon trading risk assessment model is tested and optimized using a test set to obtain a final electricity-carbon trading risk assessment model, and an electricity-carbon trading risk assessment coefficient is generated based on the final electricity-carbon trading risk assessment model.

[0051] The specific calculation method of the final electricity carbon trading risk assessment model is as follows:

[0052] In the formula, is the risk assessment coefficient of electricity carbon trading, is the impact coefficient of electricity carbon demand, is the fluctuation impact coefficient of electricity-carbon trading, To reduce the impact of emission capacity, is the weight coefficient of the impact coefficient of electricity carbon demand, is the weight coefficient of the impact coefficient of electricity carbon trading fluctuation, It is the data weight coefficient affecting emission reduction capacity.

[0053] It should be noted that machine learning is a subfield of artificial intelligence that enables computers to automatically improve their performance through experience without being explicitly programmed. Through algorithms, machine learning models can learn regularities from data, identify patterns, and make predictions or decisions.

[0054] S4, compare the input of the electricity-carbon trading risk assessment model with the preset electricity-carbon trading stability data, divide the risks of electricity-carbon trading into three risk levels of high, medium and low according to the comparison results, and make corresponding adjustments according to the three risk levels. The specific steps are as follows: Obtaining an electricity-carbon trading risk assessment coefficient through the input of the electricity-carbon trading risk assessment model; Compare and analyze the electricity-carbon trading risk assessment coefficient with the preset electricity-carbon trading stability data, and make a judgment based on the comparative analysis results; If the electricity-carbon trading risk assessment coefficient is less than the preset electricity-carbon trading stability data, it is a low-level electricity-carbon trading risk. Enterprises and investors can invest in the electricity-carbon trading market and conduct electricity-carbon trading; If the power-carbon trading risk assessment coefficient is equal to the preset power-carbon trading stability data, it is a medium-level power-carbon trading risk, and enterprises and investors need to wait and observe and continue to monitor; If the electricity carbon trading risk assessment coefficient is greater than the preset electricity carbon trading stability data, it is a high degree of electricity carbon trading risk. Enterprises and investors need to withdraw from the electricity carbon trading market and formulate appropriate strategies, waiting for the risk in the electricity carbon trading market to be reduced before starting to invest.

[0055] Embodiment 2, Figure 2 The schematic diagram of the structure of the electric carbon trading risk management system based on machine learning provided in the embodiment of the present application includes a carbon quota judgment module, a data acquisition module, a model building module and a risk degree classification module, and there are connections between the modules: A carbon quota judgment module is used to obtain an initial value of a carbon quota and historical data, extract features from the historical data to obtain a historical standard carbon emission, and judge whether there is a traded carbon quota based on the initial value of the carbon quota according to the historical standard carbon emission; A data acquisition module is used to obtain the impact data of electricity carbon demand and emission reduction capacity in several regions if there are traded carbon quotas; The model building module is used to train the impact data of electricity-carbon demand and emission reduction capacity on the risk of electricity-carbon trading based on machine learning, and to build an electricity-carbon trading risk assessment model; The risk level classification module is used to compare the input of the electricity carbon trading risk assessment model with the preset electricity carbon trading stability data, and divide the risks of electricity carbon trading into three risk levels: high, medium and low according to the comparison results, and make corresponding adjustments based on the three risk levels.

[0056] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0057] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0058] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0059] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0060] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0061] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A risk management method for electric carbon trading based on machine learning, characterized in that: The steps include: Obtaining an initial value of a carbon quota and historical data, extracting features from the historical data to obtain a historical standard carbon emission, and determining whether there is a traded carbon quota based on the initial value of the carbon quota according to the historical standard carbon emission; If there are traded carbon quotas, obtain the impact data on electricity carbon demand and emission reduction capacity in several regions; Based on machine learning, the impact data of electricity carbon demand and emission reduction capacity are trained to determine the degree of impact on electricity carbon trading risks, and a risk assessment model for electricity carbon trading is constructed; The input of the electricity carbon trading risk assessment model is compared with the preset electricity carbon trading stability data. Based on the comparison results, the risks of electricity carbon trading are divided into three risk levels: high, medium and low, and corresponding adjustments are made based on the three risk levels.

2. The method for risk management of electric carbon trading based on machine learning according to claim 1 is characterized in that: The feature extraction of the historical data is performed to obtain the historical standard carbon emissions, and the determination of whether there is a traded carbon quota based on the historical standard carbon emissions and the initial value of the carbon quota is performed. The specific steps are as follows: Obtain historical data, pre-process and extract features from the historical data, and obtain the media and equipment that affect carbon emissions; Performing data analysis based on the media and equipment that affect carbon emissions to obtain the total amount of carbon emissions, and obtaining a carbon quota demand value based on the total amount of carbon emissions; Determining the carbon quota demand value based on the carbon quota initial value; If the carbon quota demand value is greater than or equal to the carbon quota initial value, there is no carbon quota transaction amount; If the carbon quota demand value is less than the carbon quota initial value, there is a carbon quota amount to be traded.

3. The method for risk management of electric carbon trading based on machine learning according to claim 1 is characterized in that: The electricity carbon demand impact data includes the electricity carbon demand impact coefficient and the electricity carbon trading fluctuation impact coefficient. The specific method for obtaining the electricity carbon demand impact coefficient is as follows: Obtain initial carbon quota data and electricity carbon demand data for several regions; Draw a curve graph according to the initial carbon quota data and the electricity carbon demand data; Using the initial carbon quota data as the horizontal axis and the electric carbon demand data as the vertical axis, data analysis is performed according to the curve chart to obtain the electric carbon demand fluctuation coefficient; Obtain carbon quota introduction data, market carbon quota fluctuation data and regional carbon emission factors; The carbon quota introduction data, market carbon quota fluctuation data and regional carbon emission factors are used to calculate the electricity carbon consumption impact factor based on the preset electricity carbon consumption impact formula; The electricity carbon demand impact coefficient is calculated based on the preset electricity carbon demand impact formula according to the electricity carbon consumption impact factor combined with the electricity carbon demand fluctuation coefficient.

4. The method for risk management of electric carbon trading based on machine learning according to claim 3 is characterized in that: The specific method for obtaining the electricity-carbon trading fluctuation impact coefficient is as follows: Obtain electricity carbon price data for several trading days; The average value of the electricity carbon price is calculated by averaging the electricity carbon price data of the several trading days; Obtaining first data on fluctuations in electricity-carbon trading, the first data including total trading days, electricity-carbon trading volume, and average carbon price in the carbon market; The electricity carbon trading fluctuation impact coefficient is calculated based on the first data of electricity carbon trading fluctuation and the average electricity carbon price based on the preset electricity carbon trading fluctuation impact formula.

5. The method for risk management of electric carbon trading based on machine learning according to claim 1, characterized in that: The specific method for obtaining the emission reduction capacity impact data is as follows: Obtain data on changes in emission reduction capacity in several regions; Obtaining the maximum potential coefficient of emission reduction capacity and the time span of emission reduction capacity change based on regression analysis through the emission reduction capacity change data; By monitoring the efficiency of emission reduction capacity in several areas within a period of time L, the speed coefficient of emission reduction capacity change is obtained; The emission reduction capacity impact data is calculated based on a preset emission reduction capacity impact formula according to the emission reduction capacity change data combined with a speed coefficient of the emission reduction capacity change.

6. The method for risk management of electric carbon trading based on machine learning according to claim 1, characterized in that: The specific steps of training the impact data of electricity carbon demand and emission reduction capacity on the risk of electricity carbon trading based on machine learning and building an electricity carbon trading risk assessment model are as follows: Obtain historical data on the impact of electricity carbon demand and historical data on the impact of emission reduction capacity, and pre-process the data; The processed data is divided into a training set and a test set. The training set is used based on a machine learning algorithm to train the impact degree of the electricity-carbon trading risk through the historical electricity-carbon demand impact data and the historical emission reduction capacity impact data. Based on automatic adjustment, the weight coefficient of the electricity-carbon trading risk is obtained to build an electricity-carbon trading risk assessment model. The electricity-carbon trading risk assessment model is tested and optimized using a test set to obtain a final electricity-carbon trading risk assessment model, and an electricity-carbon trading risk assessment coefficient is generated based on the final electricity-carbon trading risk assessment model.

7. The method for risk management of electric carbon trading based on machine learning according to claim 1, characterized in that: The input of the electric carbon trading risk assessment model is compared with the preset electric carbon trading stability data, and the risk of the electric carbon trading is divided into three risk levels of high, medium and low according to the comparison results, and corresponding adjustments are made according to the three risk levels. The specific steps are as follows: Obtaining an electricity-carbon trading risk assessment coefficient through the input of the electricity-carbon trading risk assessment model; Compare and analyze the electricity-carbon trading risk assessment coefficient with the preset electricity-carbon trading stability data, and make a judgment based on the comparative analysis results; If the electricity-carbon trading risk assessment coefficient is less than the preset electricity-carbon trading stability data, it is a low-level electricity-carbon trading risk; If the electricity-carbon trading risk assessment coefficient is equal to the preset electricity-carbon trading stability data, it is a medium-level electricity-carbon trading risk; If the electricity-carbon trading risk assessment coefficient is greater than the preset electricity-carbon trading stability data, it is a high level of electricity-carbon trading risk.

8. The method for risk management of electric carbon trading based on machine learning according to claim 3 is characterized in that: The specific calculation formula of the electric carbon consumption influencing factor is as follows: The specific calculation formula of the electric carbon demand impact coefficient is as follows: In the formula, is the impact coefficient of electricity carbon demand, is the initial carbon quota data of the ith region, is the electricity carbon demand data of the ith region, is the fluctuation coefficient of electricity carbon demand, is the influencing factor of carbon consumption, Introduce data for the carbon quota of the i-th region, is the regional carbon emission factor of the i-th region, where i=1,2,3,...,R, R is an integer, N is a number of regions, It is the market carbon quota fluctuation data.

9. The method for risk management of electric carbon trading based on machine learning according to claim 6 is characterized in that: The specific calculation method of the final electricity carbon trading risk assessment model is as follows: In the formula, is the risk assessment coefficient of electricity carbon trading, is the impact coefficient of electricity carbon demand, is the impact coefficient of electricity-carbon trading fluctuations, To reduce the impact of emission capacity, is the weight coefficient of the impact coefficient of electricity carbon demand, is the weight coefficient of the impact coefficient of electricity carbon trading fluctuation, It is the data weight coefficient affecting emission reduction capacity.

10. A system using the method for managing electric carbon trading risk based on machine learning as described in any one of claims 1 to 9, characterized in that: It includes carbon quota judgment module, data acquisition module, model building module and risk level classification module. There are connections between the modules: A carbon quota judgment module is used to obtain an initial value of a carbon quota and historical data, extract features from the historical data to obtain a historical standard carbon emission, and judge whether there is a traded carbon quota based on the initial value of the carbon quota according to the historical standard carbon emission; A data acquisition module is used to obtain the impact data of electricity carbon demand and emission reduction capacity in several regions if there are traded carbon quotas; The model building module is used to train the impact data of electricity-carbon demand and emission reduction capacity on the risk of electricity-carbon trading based on machine learning, and to build an electricity-carbon trading risk assessment model; The risk level classification module is used to compare the input of the electricity carbon trading risk assessment model with the preset electricity carbon trading stability data, and divide the risks of electricity carbon trading into three risk levels: high, medium and low according to the comparison results, and make corresponding adjustments based on the three risk levels.

Images