Carbon market behavior decision-making method and system

Abstract

The invention discloses a carbon market behavior decision-making method and system, and relates to the technical field of carbon market behavior decision-making, and the method comprises the following steps: obtaining carbon market behavior influence data which comprises market fluctuation influence data, climate fluctuation influence data and energy fluctuation influence data; establishing a carbon market behavior risk model based on an LSTM neural network according to the carbon market behavior influence data, and generating a carbon market behavior risk coefficient; and obtaining carbon market behavior risk coefficients in different time periods, obtaining a carbon market behavior risk range based on a statistical method, obtaining a carbon market behavior risk level grade based on the carbon market behavior risk range according to the carbon market behavior risk coefficients, and determining a decision of the carbon market behavior through the carbon market behavior risk level grade. According to the invention, through carbon market behavior risk prediction, the problem of decision-making influence caused by risks generated by carbon market behaviors is solved.

Description

Technical Field

[0001] The present invention relates to the field of carbon market behavior decision-making, and more specifically, to the field of carbon market behavior decision-making methods and system technologies. Background Art

[0002] With the increase of greenhouse gas emissions, global warming and climate change have become one of the biggest environmental challenges facing the world today. Since the 1990s, global warming and climate change have become the core issues in the field of environmental protection and sustainable development. In order to meet this challenge, many countries and regions have promoted the reduction of greenhouse gas emissions through means, especially carbon pricing mechanisms. The carbon market, that is, the market through trading carbon quotas or carbon credits, is one of the important tools to achieve the combination of carbon emission control and economic incentives.

[0003] Due to the influence of various factors such as market supply and demand and economic cycles, the price fluctuations in the carbon market are large, making it difficult for companies and investors to predict carbon market behavior, thus being unable to avoid risks and affecting their long-term emission reduction decisions. At present, the decision-making on carbon market behavior is mainly focused on measuring the carbon market from a specific perspective, and has not yet considered the decision-making impact of the risks generated by carbon market behavior from an overall perspective.

[0004] In view of the above problems, the present invention proposes a solution. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a carbon market behavior decision-making method and system, which solves the problem of decision-making impact brought about by the risks generated by carbon market behavior through carbon market behavior risk prediction.

[0006] To achieve the above object, the present invention provides the following technical solutions: A carbon market behavior decision-making method and system include the following steps: obtaining carbon market behavior impact data, wherein the carbon market behavior impact data includes market fluctuation impact data, climate fluctuation impact data, and energy fluctuation impact data; establishing a carbon market behavior risk model based on an LSTM neural network according to the carbon market behavior impact data, and generating a carbon market behavior risk coefficient; obtaining the carbon market behavior risk coefficient in different time periods, and obtaining the carbon market behavior risk range based on a statistical method, obtaining the carbon market behavior risk level grade based on the carbon market behavior risk range according to the carbon market behavior risk coefficient, and determining the decision on the carbon market behavior through the carbon market behavior risk level grade.

[0007] In a preferred embodiment, the market volatility impact data includes a market volatility impact coefficient, and the specific acquisition method of the market volatility impact coefficient is as follows: obtain the production data, output data and unit carbon price data of the market volatility impact data; perform data analysis on the production data to obtain the production impact coefficient, and perform data analysis on the output data to obtain the output impact coefficient; calculate the carbon price fluctuation coefficient based on the production impact coefficient, the output impact coefficient and the unit carbon price data in combination with the carbon price sensitivity coefficient based on a preset carbon price fluctuation formula; obtain the carbon quota data of the market and the user influence factor; calculate the market volatility impact coefficient based on the preset market volatility impact formula based on the carbon price fluctuation coefficient in combination with the carbon quota data of the market and the user influence factor.

[0008] In a preferred embodiment, the climate fluctuation impact data includes a climate fluctuation impact coefficient, and the climate fluctuation impact coefficient is specifically obtained by the following method: obtaining first data of the climate fluctuation impact coefficient, the first data including the precipitation change rate and the frequency of drought and flood events; real-time monitoring and obtaining temperature change data, performing data processing and analysis on the temperature change data to obtain average temperature changes and extreme climate frequencies; calculating the temperature fluctuation amplitude based on the average temperature changes and extreme climate frequencies; obtaining historical wind speed seasonal change data, performing data analysis on the historical wind speed seasonal change data to obtain the wind speed change rate; calculating the climate fluctuation impact coefficient based on the first data of the climate fluctuation impact coefficient and the temperature fluctuation amplitude combined with the wind speed change rate based on a preset climate fluctuation impact formula.

[0009] In a preferred embodiment, the energy fluctuation impact data includes an energy fluctuation impact coefficient, and the specific acquisition method of the energy fluctuation impact coefficient is as follows: obtain energy risk impact data of the energy fluctuation impact coefficient, and the energy risk impact data includes a natural gas risk impact coefficient and an oil risk impact coefficient; monitor and obtain the coal-fired power generation cost and carbon emissions in real time to calculate the coal risk impact coefficient; obtain the historical natural gas risk impact coefficient and the historical oil risk impact coefficient as well as the historical coal risk impact coefficient, and obtain the change weights of the natural gas risk impact coefficient, the oil risk impact coefficient and the coal risk impact coefficient based on regression analysis; collect the enterprise technology carbon emissions and output efficiency within G time, and calculate the production technology transformation impact degree coefficient; calculate the energy fluctuation impact coefficient based on the energy risk impact data and the coal risk impact coefficient in combination with the production technology transformation impact degree coefficient based on a preset energy fluctuation impact formula.

[0010] In a preferred embodiment, the carbon market behavior risk model is established based on the LSTM neural network according to the carbon market behavior impact data to generate a carbon market behavior risk coefficient, and the specific steps are as follows: collect historical carbon market behavior impact data, and the historical carbon market behavior impact data includes historical market fluctuation impact data, historical climate fluctuation impact data, and historical energy fluctuation impact data; clean and preprocess the historical carbon market behavior impact data, extract features from the processed historical carbon market behavior impact data and divide them into a training set and a test set; input the training set into the LSTM neural network input layer for training, and capture the long-term dependency between the carbon market behavior impact data and the carbon market behavior risk in the LSTM layer to obtain a preliminary carbon market behavior risk model; add a fully connected layer to the preliminary carbon market behavior risk model to construct a carbon market behavior risk function to generate a complete carbon market behavior risk model; input the test set into the complete carbon market behavior risk model, evaluate the model, and improve the prediction accuracy and generalization ability of the model; input the newly acquired carbon market behavior impact data into the complete carbon market behavior risk model to generate a carbon market behavior risk coefficient.

[0011] In a preferred embodiment, the carbon market behavior risk range is obtained based on a statistical method, and the carbon market behavior risk level is obtained based on the carbon market behavior risk range according to the carbon market behavior risk coefficient. The specific steps are as follows: collect the carbon market behavior risk coefficient in different historical time periods, and calculate the mean and standard deviation of the carbon market behavior risk coefficient in different historical time periods based on a statistical method; take the mean of the carbon market behavior risk coefficient as the minimum value of the carbon market behavior risk range ; Take L times the standard deviation of the carbon market behavior risk coefficient as the maximum value of the carbon market behavior risk range , where L is a constant used to measure the risk level of carbon market behavior; obtain the latest carbon market behavior risk coefficient ;like <, at this time, the carbon market behavior risk level is at the first level risk level; if < < , at this time, the carbon market behavior risk level is at the secondary risk level; if >, at this time, the risk level of carbon market behavior is at the third risk level.

[0012] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. By quantifying the risks of carbon market behavior and predicting market fluctuations, we can help companies identify potential risks in advance, optimize carbon emission management and investment decisions, and improve cost control and compliance.

[0013] 2. The carbon market behavior risk prediction model is constructed through the LSTM neural network algorithm, which can effectively capture long-term dependencies and nonlinear relationships, improve prediction accuracy, and help companies better cope with the complex fluctuations and risks of the carbon market. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A structural diagram of the carbon market behavior decision-making method provided in an embodiment of the present application.

[0015] Figure 2 A schematic diagram of the structure of the carbon market behavior decision-making system 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 structural diagram of the carbon market behavior decision-making method provided in the embodiment of the present application includes the following steps: S1, obtaining carbon market behavior impact data, wherein the carbon market behavior impact data includes market fluctuation impact data, climate fluctuation impact data and energy fluctuation impact data.

[0018] In this example, the market volatility impact data includes the market volatility impact coefficient. The market volatility impact coefficient refers to the fluctuation of carbon prices in the carbon trading market and the impact of some user behaviors, which will cause changes in the carbon market. These changes will have a direct or indirect impact on the decision-making, cost structure, risk management and other behaviors of market participants.

[0019] Obtaining data on market volatility has the following advantages for analyzing carbon market behavior risks: Timely reflection of market dynamics: The immediacy of market volatility data. The carbon market itself is affected by many factors, such as changes, international climate agreements, economic development, and industry technological progress. By obtaining market volatility data in real time, we can timely understand the changing trends of carbon prices, predict potential market adjustments and risks, and respond quickly. Obtaining market volatility data can help analyze the sensitivity of carbon prices to different market events or changes, and help decision makers make timely adjustments.

[0020] Identify sources of market risk: Fluctuations in the carbon market are often closely related to changes in carbon quotas, such as possible adjustments to carbon quotas or changes to carbon pricing mechanisms. By analyzing volatility data, it is possible to clearly identify which factors have the greatest impact on market volatility, helping regulators predict the potential impact of changes. Macroeconomic factors such as economic recession and growth expectations can also affect carbon price fluctuations. By obtaining relevant volatility data, it is possible to deeply analyze the impact of these macroeconomic factors on the carbon market.

[0021] Improve market transparency: Market participant behavior analysis can reveal the behavior patterns of market participants by analyzing the fluctuations in carbon market prices and trading volumes. For example, whether there are problems such as speculative behavior, information asymmetry, or market manipulation. This helps to enhance market transparency and improve market fairness and efficiency. Liquidity analysis: Market volatility data can help assess the liquidity of the carbon market and determine whether the market can cope with large-scale buying and selling activities or whether there are liquidity risks.

[0022] Support long-term strategic planning: predict market trends. By analyzing historical market volatility data, we can better understand the cyclical characteristics of the carbon market, predict future price trends and market demand changes, and help companies formulate long-term carbon quota management and carbon emission reduction strategies. Investment decision support. For carbon exchanges, investors and energy companies, market volatility data is an important basis for investment decisions. By understanding market fluctuations and potential risks, investors can reasonably allocate capital and reduce the risk of loss.

[0023] Help decision makers identify systemic risks: Market volatility data helps identify systemic risks in the carbon market, for example, whether certain factors (such as international economic recession, energy price fluctuations, etc.) may have a chain reaction on the entire market. By identifying these systemic risks, relevant parties can take measures to reduce their threats to market stability. Market chain effect analysis: the carbon market is often a global market, so fluctuations in other markets around the world may affect the carbon market. By tracking market fluctuations, this chain reaction can be better understood.

[0024] The specific method for obtaining the market volatility impact coefficient is as follows: Obtaining production data, output data and unit carbon price data of the market fluctuation impact data; Performing data analysis on the production data to obtain a production impact coefficient, and performing data analysis on the output data to obtain an output impact coefficient; The carbon price fluctuation coefficient is calculated based on the production impact coefficient, output impact coefficient and unit carbon price data combined with the carbon price sensitivity coefficient based on the preset carbon price fluctuation formula; Obtain market carbon quota data and user impact factors; The market volatility impact coefficient is calculated based on the carbon price volatility coefficient, the market carbon quota data and the user influence factor based on the preset market volatility impact formula.

[0025] The specific calculation formula of the carbon price fluctuation coefficient is as follows:

[0026] The specific method for obtaining the market volatility impact coefficient is as follows:

[0027] In the formula, is the market volatility impact coefficient, For the carbon quota data of the market, is the carbon price volatility coefficient, is the household influence factor, is the production impact coefficient of the i-th market, where i=1,2,3,..., R, R is an integer, N is the number of markets, is the unit carbon price data of the ith market, is the output impact coefficient of the i-th market, is the carbon price sensitivity coefficient, and Q is the total carbon demand.

[0028] It should be noted that changes in the production impact coefficient and the output impact coefficient will affect the balance of the market, thus causing changes in carbon prices. In the supply impact, when new carbon market participants enter the market or existing participants increase carbon quotas, the supply and demand of the market may change, thus affecting changes in the entire market. Therefore, obtaining these data is important for analyzing market fluctuations.

[0029] The climate fluctuation impact data includes the climate fluctuation impact coefficient. The climate fluctuation impact coefficient means that when the climate is at low temperatures, users tend to increase the use of air conditioning and heating levels, which will lead to a rapid increase in actual carbon dioxide emissions and quota demand in the short term, which will have a certain risk impact on the behavior of the carbon market and cause changes in the decisions of market participants.

[0030] Obtaining the climate fluctuation impact coefficient has the following advantages for analyzing carbon market behavior risks: Quantifying the impact of climate change on the carbon market: The climate fluctuation impact coefficient can help quantify the uncertainty and volatility brought about by climate change, especially in markets such as the carbon market that are closely related to energy and industrial production. Through the quantitative analysis of climate fluctuations, we can more clearly understand the impact that climate change may have on carbon emissions, energy demand and price fluctuations, so as to more accurately predict market behavior and risks.

[0031] Help assess the stability of carbon pricing: Extreme weather events caused by climate change (such as extreme cold, heat waves, droughts, etc.) often have a significant impact on multiple areas such as energy supply, industrial production, and transportation. By analyzing the impact of climate fluctuations on the market, the stability and price volatility of the carbon market can be assessed. For example, climate change may cause sharp fluctuations in energy prices, which will directly affect carbon prices. Understanding this can help policymakers and investors better manage the price volatility risks of the carbon market.

[0032] Supporting risk management and strategy formulation in the carbon market: Carbon market participants need to develop risk management strategies to cope with market fluctuations and changes. The climate volatility impact coefficient can provide specific information on how climate change affects carbon emissions, carbon trading and carbon pricing. In this way, participants can more effectively plan and optimize their carbon trading strategies, investment portfolios, hedging mechanisms, etc., thereby reducing potential risks brought about by market uncertainties.

[0033] Improve the accuracy of carbon market forecasts: By incorporating the impact of climate fluctuations into the behavioral model of the carbon market, the forecast accuracy of the carbon market can be improved. As a variable, the climate fluctuation coefficient can be combined with other economic, technical and other factors to form a more comprehensive and accurate market analysis framework. This enables researchers and decision makers to better predict future trends and possible risk events in the carbon market, helping to develop a more forward-looking carbon market.

[0034] Enhance transparency and compliance in the carbon market: As global climate change becomes increasingly serious, transparency and compliance in the carbon market have become a focus of attention for organizations. By systematically analyzing the impact of climate fluctuations on the carbon market, more transparent data support can be provided to the market, thereby promoting the healthy development of the market. This will not only help enhance the confidence of market participants, but also help improve the efficiency and effectiveness of global climate governance.

[0035] Promoting interdisciplinary cooperation and coordination: The study of climate fluctuations usually requires cooperation and integration of multiple disciplines such as climatology, economics, environmental science, and energy. This interdisciplinary cooperation can not only enhance the deep understanding of the interactive relationship between climate change and carbon market, but also promote the coordination of organizations in climate change, environmental protection, carbon pricing, etc. This has positive significance for the overall global action to address climate change.

[0036] Promoting long-term investment decisions and sustainable development: Through in-depth analysis of climate fluctuations, carbon market participants can identify long-term climate risks and adjust their investment strategies to promote the development of a low-carbon economy. This helps companies and investors make long-term plans and avoid the impact of short-term market fluctuations on their sustainable development strategies.

[0037] The specific method for obtaining the climate fluctuation influence coefficient is as follows: Acquiring first data of climate fluctuation impact coefficient, wherein the first data includes precipitation change rate and frequency of drought and flood events; Real-time monitoring and acquisition of temperature change data, and data processing and analysis of the temperature change data to obtain average temperature changes and extreme climate frequencies; The temperature fluctuation amplitude is calculated based on the average temperature change and the frequency of extreme climate; Obtain historical wind speed seasonal change data, and perform data analysis on the historical wind speed seasonal change data to obtain the wind speed change rate; The climate fluctuation influence coefficient is calculated based on the first data of the climate fluctuation influence coefficient and the temperature fluctuation amplitude combined with the wind speed change rate based on a preset climate fluctuation influence formula.

[0038] The specific calculation steps of the climate fluctuation influence coefficient are as follows:

[0039] In the formula, is the climate fluctuation influence coefficient, is the frequency of drought events, H is the frequency of flood events, is the wind speed change rate, is the precipitation change rate, is the temperature fluctuation range.

[0040] It should be noted that temperature change data is the most intuitive indicator. The degree of global warming or local climate change may affect energy demand and production costs, and thus affect the carbon market. The frequency of drought and flood events can be understood as these events may cause companies to stop production or make emergency adjustments, which in turn brings risks to the carbon market and affects the volatility of the carbon market.

[0041] Energy fluctuation impact data include energy fluctuation impact coefficient, where energy fluctuation impact coefficient refers to fluctuations related to energy market prices, supply, demand, and industrial technological transformation and upgrading. These fluctuations have a direct or indirect impact on the performance of the carbon market and the behavior of participants, affecting not only the price trend of the carbon market, but also the behavior and decision-making of traders.

[0042] Obtaining the energy fluctuation impact coefficient has the following advantages for analyzing carbon market behavior risks: Quantifying the transmission effect of the energy market on the carbon market: There is a close interconnection between the energy market and the carbon market. Factors such as energy price fluctuations, supply chain changes, and demand fluctuations will directly affect the supply and demand relationship and price trends in the carbon market. By obtaining the energy fluctuation impact coefficient, it is possible to quantify the specific impact of these fluctuations on the carbon market. For example, fluctuations in the price of a certain energy source may lead to changes in the demand for the carbon market, or change the price of carbon quotas. Quantifying these impacts can help predict risk changes and possible price fluctuations in the carbon market.

[0043] Improve the accuracy of the prediction model: In the prediction of behavioral risk in the carbon market, the use of energy fluctuation impact coefficients can establish a mathematical relationship between changes in the energy market and the behavior of the carbon market, making the prediction model more accurate. For example, if the emissions of a carbon market participant are closely related to energy prices, then by introducing the energy fluctuation impact coefficient, the fluctuations in its carbon quota demand can be more accurately predicted, and based on this, the overall price trend and risk of the carbon market can be predicted.

[0044] Help decision makers identify key risk factors: Obtaining energy volatility impact coefficients helps analyze which energy factors have a significant impact on the behavior of the carbon market. By analyzing these coefficients, policymakers, companies, and investors can identify the most important sources of risk in the carbon market. For example, certain energy prices (such as coal and natural gas) may have a greater impact on carbon market price fluctuations, while other energy sources (such as renewable energy) may have a smaller impact. Mastering this information can help relevant parties take more targeted countermeasures, such as hedging strategies and risk avoidance.

[0045] Optimize risk management strategies: For enterprises or investors in the carbon market, the volatility of the energy market is an important factor in determining their carbon trading decisions and risk exposure. By obtaining and analyzing the energy volatility impact coefficient, enterprises can better predict the risks that may be brought about by energy price fluctuations and adjust their carbon quota purchase or sale strategies accordingly. For example, if energy prices rise, resulting in an increase in carbon emission costs, enterprises may purchase carbon quotas in advance or adjust production plans to reduce emissions, thereby avoiding the risks brought about by price fluctuations.

[0046] Enhanced sensitivity to market fluctuations: Obtaining the impact coefficient of energy fluctuations can improve the sensitivity of carbon market analysis, allowing analysts to promptly identify and respond to potential abnormal fluctuations in the market. Price fluctuations in the carbon market may be affected by unexpected events in the energy market (such as oil price surges, tight energy supply, etc.). By understanding the impact coefficients of these fluctuations, the magnitude and direction of price fluctuations in the carbon market can be predicted, helping analysts to respond in a timely manner.

[0047] Better understand the price formation mechanism of the carbon market: Energy prices are one of the important driving factors of carbon market prices. Obtaining the energy fluctuation impact coefficient can help analyze the internal mechanism of carbon market price formation and deeply understand the impact of energy price changes on carbon prices. For example, when energy prices fluctuate significantly, changes in carbon prices may lag or amplify. Understanding its fluctuation impact coefficient can help explain the transmission effect of this price and provide an important basis for the carbon market pricing model.

[0048] Enhanced cross-market analysis capabilities: The interaction between the carbon market and the energy market is very complex, and the energy volatility impact coefficient can help with cross-market analysis. For example, rising natural gas prices may lead to increased carbon emissions in the power industry, as some electricity production may shift to using energy sources with higher carbon emissions (such as coal). By analyzing these impact coefficients, a cross-market risk prediction model can be established to understand the interactive relationship between different markets and provide data support for cross-market investment and trading decisions.

[0049] The specific method for obtaining the energy fluctuation influence coefficient is as follows: Obtaining energy risk impact data of energy fluctuation impact coefficients, wherein the energy risk impact data includes a natural gas risk impact coefficient and an oil risk impact coefficient; Real-time monitoring and acquisition of coal-fired power generation costs and carbon emissions to calculate the coal risk impact coefficient; Obtain the historical natural gas risk impact coefficient, the historical oil risk impact coefficient and the historical coal risk impact coefficient. Based on regression analysis, obtain the change weights of the natural gas risk impact coefficient, the oil risk impact coefficient and the coal risk impact coefficient respectively; Collect the enterprise's technical carbon emissions and output efficiency within the G time and calculate the production technology reform impact coefficient; The energy fluctuation impact coefficient is calculated based on the energy risk impact data and the coal risk impact coefficient combined with the production technology transformation impact coefficient based on the preset energy fluctuation impact formula.

[0050] The specific calculation formula of the energy fluctuation impact coefficient is as follows:

[0051] In the formula, is the energy fluctuation impact coefficient, is the coal risk impact coefficient, is the change weight of the coal risk impact coefficient, is the change weight of the oil risk impact coefficient, is the change weight of the natural gas risk impact coefficient, is the oil risk impact coefficient, is the natural gas risk impact coefficient, is the coefficient of impact degree of production technological reform.

[0052] It should be noted that the coefficient of the impact degree of production technology transformation can be understood as the adjustment of enterprises' production technology transformation for equipment with large carbon emissions to reduce the carbon emissions of the equipment. Production technology transformation of a large number of equipment will affect the carbon market, resulting in corresponding changes in the decision-making of carbon market participants or enterprises.

[0053] S2, based on the carbon market behavior impact data, establishes a carbon market behavior risk model based on the LSTM neural network to generate a carbon market behavior risk coefficient. The specific steps are as follows: Collect historical carbon market behavior impact data, including historical market fluctuation impact data, historical climate fluctuation impact data, and historical energy fluctuation impact data; Clean and preprocess the historical carbon market behavior impact data, extract features from the processed historical carbon market behavior impact data and divide them into training sets and test sets; The training set is input into the LSTM neural network input layer for training, and the long-term dependency between carbon market behavior impact data and carbon market behavior risk is captured in the LSTM layer to obtain a preliminary carbon market behavior risk model; Add a fully connected layer to the preliminary carbon market behavior risk model to construct the carbon market behavior risk function and generate a complete carbon market behavior risk model. Input the test set into the complete carbon market behavior risk model to evaluate the model and improve its prediction accuracy and generalization ability; The newly acquired carbon market behavior impact data is input into the complete carbon market behavior risk model to generate the carbon market behavior risk coefficient.

[0054] The specific calculation method of the carbon market behavior risk model is as follows:

[0055] In the formula, is the carbon market behavior risk coefficient, is the market volatility impact coefficient, is the climate fluctuation influence coefficient, is the energy fluctuation impact coefficient, , , is the weight coefficient.

[0056] S3, obtain the carbon market behavior risk coefficient in different historical time periods, and obtain the carbon market behavior risk range based on statistical methods, obtain the carbon market behavior risk level grade based on the carbon market behavior risk coefficient and the carbon market behavior risk range, and determine the decision on carbon market behavior through the carbon market behavior risk level grade. The specific steps are as follows: Collect the carbon market behavior risk coefficients in different historical time periods, and calculate the mean and standard deviation of the carbon market behavior risk coefficients in different historical time periods based on statistical methods; The mean value of the carbon market behavior risk coefficient is taken as the minimum value of the carbon market behavior risk range. ; Take L times the standard deviation of the carbon market behavior risk coefficient as the maximum value of the carbon market behavior risk range , where L is a constant used to measure the risk level of carbon market behavior; like <, at this time, the carbon market behavior risk level is at the first level risk level; like < < , at this time, the carbon market behavior risk level is at the secondary risk level; like >, at this time, the carbon market behavior risk level is at the third risk level; When the carbon market behavior risk level is at the first-level risk level, it means that the carbon market risk is low at this time. Participants can increase their market participation, take advantage of the lower market volatility and risk, directly participate in spot market transactions, and obtain better market prices; When the carbon market behavior risk level is at the secondary risk level, it means that the carbon market risk is at the intermediate level. Enterprises can take more emission reduction measures to reduce the demand for purchasing carbon quotas in the future and avoid cost pressure caused by rising market prices. When the carbon market behavior risk level is at the third risk level, it means that the carbon market risk level is relatively high at this time. Enterprises or participants can choose to lock in prices in the futures market to reduce the uncertainty caused by price fluctuations. They can also formulate emergency plans and set stop-loss points in advance to reduce potential losses.

[0057] Embodiment 2, Figure 2 The structural diagram of the carbon market behavior decision-making system provided in the embodiment of the present application includes a carbon market behavior impact data acquisition module, a carbon market behavior risk model generation module, and a carbon market behavior risk assessment module. There are connections between the modules: A carbon market behavior impact data acquisition module is used to acquire carbon market behavior impact data, wherein the carbon market behavior impact data includes market fluctuation impact data, climate fluctuation impact data, and energy fluctuation impact data; The carbon market behavior risk model generation module is used to establish a carbon market behavior risk model based on the LSTM neural network according to the carbon market behavior impact data and generate the carbon market behavior risk coefficient; The carbon market behavior risk assessment module is used to obtain the carbon market behavior risk coefficient in different time periods, and obtain the carbon market behavior risk range based on statistical methods, obtain the carbon market behavior risk level grade based on the carbon market behavior risk coefficient and the carbon market behavior risk level grade, and determine the decision on carbon market behavior through the carbon market behavior risk level grade.

[0058] 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.

[0059] 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.

[0060] 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.

[0061] 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.

[0062] 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.

[0063] 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. Carbon market behavior decision-making method, characterized by: The steps include: Obtaining carbon market behavior impact data, wherein the carbon market behavior impact data includes market fluctuation impact data, climate fluctuation impact data, and energy fluctuation impact data; According to the carbon market behavior impact data, a carbon market behavior risk model is established based on the LSTM neural network to generate a carbon market behavior risk coefficient; Obtain the carbon market behavior risk coefficient in different time periods, and obtain the carbon market behavior risk range based on statistical methods, obtain the carbon market behavior risk level grade based on the carbon market behavior risk range according to the carbon market behavior risk coefficient, and determine the decision on carbon market behavior through the carbon market behavior risk level grade.

2. The carbon market behavior decision-making method according to claim 1, characterized in that: The market volatility impact data includes a market volatility impact coefficient. The specific method for obtaining the market volatility impact coefficient is as follows: Obtaining production data, output data and unit carbon price data of the market fluctuation impact data; Performing data analysis on the production data to obtain a production impact coefficient, and performing data analysis on the output data to obtain an output impact coefficient; The carbon price fluctuation coefficient is calculated based on the production impact coefficient, output impact coefficient and unit carbon price data combined with the carbon price sensitivity coefficient based on the preset carbon price fluctuation formula; Obtain market carbon quota data and user impact factors; The market volatility impact coefficient is calculated based on the carbon price volatility coefficient, the market carbon quota data and the user influence factor based on the preset market volatility impact formula.

3. The carbon market behavior decision-making method according to claim 1, characterized in that: The climate fluctuation impact data includes a climate fluctuation impact coefficient. The specific method for obtaining the climate fluctuation impact coefficient is as follows: Acquiring first data of climate fluctuation impact coefficient, wherein the first data includes precipitation change rate and frequency of drought and flood events; Real-time monitoring and acquisition of temperature change data, and data processing and analysis of the temperature change data to obtain average temperature changes and extreme climate frequencies; The temperature fluctuation amplitude is calculated based on the average temperature change and the frequency of extreme climate; Obtain historical wind speed seasonal change data, and perform data analysis on the historical wind speed seasonal change data to obtain the wind speed change rate; The climate fluctuation influence coefficient is calculated based on the first data of the climate fluctuation influence coefficient and the temperature fluctuation amplitude combined with the wind speed change rate based on a preset climate fluctuation influence formula.

4. The carbon market behavior decision-making method according to claim 1, characterized in that: The energy fluctuation impact data includes an energy fluctuation impact coefficient. The specific method for obtaining the energy fluctuation impact coefficient is as follows: Obtaining energy risk impact data of energy fluctuation impact coefficients, wherein the energy risk impact data includes a natural gas risk impact coefficient and an oil risk impact coefficient; Real-time monitoring and acquisition of coal-fired power generation costs and carbon emissions to calculate the coal risk impact coefficient; Obtain the historical natural gas risk impact coefficient, the historical oil risk impact coefficient and the historical coal risk impact coefficient. Based on regression analysis, obtain the change weights of the natural gas risk impact coefficient, the oil risk impact coefficient and the coal risk impact coefficient respectively; Collect the enterprise's technical carbon emissions and output efficiency within the G time and calculate the production technology reform impact coefficient; The energy fluctuation impact coefficient is calculated based on the energy risk impact data and the coal risk impact coefficient combined with the production technology transformation impact coefficient based on the preset energy fluctuation impact formula.

5. The carbon market behavior decision-making method according to claim 1, characterized in that: The carbon market behavior risk model is established based on the LSTM neural network according to the carbon market behavior impact data to generate the carbon market behavior risk coefficient. The specific steps are as follows: Collect historical carbon market behavior impact data, including historical market fluctuation impact data, historical climate fluctuation impact data, and historical energy fluctuation impact data; Clean and preprocess the historical carbon market behavior impact data, extract features from the processed historical carbon market behavior impact data and divide them into training sets and test sets; The training set is input into the LSTM neural network input layer for training, and the long-term dependency between carbon market behavior impact data and carbon market behavior risk is captured in the LSTM layer to obtain a preliminary carbon market behavior risk model; Add a fully connected layer to the preliminary carbon market behavior risk model to construct the carbon market behavior risk function and generate a complete carbon market behavior risk model. Input the test set into the complete carbon market behavior risk model to evaluate the model and improve its prediction accuracy and generalization ability; The newly acquired carbon market behavior impact data is input into the complete carbon market behavior risk model to generate the carbon market behavior risk coefficient.

6. The carbon market behavior decision-making method according to claim 1, characterized in that: The steps of obtaining the carbon market behavior risk range based on the statistical method and obtaining the carbon market behavior risk level based on the carbon market behavior risk range according to the carbon market behavior risk coefficient are as follows: Collect the carbon market behavior risk coefficients in different historical time periods, and calculate the mean and standard deviation of the carbon market behavior risk coefficients in different historical time periods based on statistical methods; The mean value of the carbon market behavior risk coefficient is taken as the minimum value of the carbon market behavior risk range. ; Take L times the standard deviation of the carbon market behavior risk coefficient as the maximum value of the carbon market behavior risk range , where L is a constant used to measure the risk level of carbon market behavior; Get the latest carbon market behavior risk factor ; like <, at this time, the carbon market behavior risk level is at the first level risk level; like < < , at this time, the carbon market behavior risk level is at the secondary risk level; like >, at this time, the risk level of carbon market behavior is at the third risk level.

7. The carbon market behavior decision-making method according to claim 2, characterized in that: The specific calculation formula of the carbon price fluctuation coefficient is as follows: The specific method for obtaining the market volatility impact coefficient is as follows: In the formula, is the market volatility impact coefficient, Carbon quota data for the market, is the carbon price volatility coefficient, is the household influence factor, is the production impact coefficient of the i-th market, where i=1,2,3,..., R, R is an integer, N is the number of markets, is the unit carbon price data of the i-th market, is the output impact coefficient of the i-th market, is the carbon price sensitivity coefficient, and Q is the total carbon demand.

8. The carbon market behavior decision-making method according to claim 3 is characterized in that: The specific calculation steps of the climate fluctuation influence coefficient are as follows: In the formula, is the climate fluctuation influence coefficient, is the frequency of drought events, H is the frequency of flood events, is the wind speed change rate, is the precipitation change rate, is the temperature fluctuation range.

9. The carbon market behavior decision-making method according to claim 5, characterized in that: The specific calculation method of the carbon market behavior risk model is as follows: In the formula, is the carbon market behavior risk coefficient, is the market volatility impact coefficient, is the climate fluctuation influence coefficient, is the energy fluctuation impact coefficient, , , is the weight coefficient.

10. A system using the carbon market behavior decision-making method according to any one of claims 1 to 9, characterized in that: It includes the carbon market behavior impact data acquisition module, the carbon market behavior risk model generation module, and the carbon market behavior risk assessment module. There are connections between the modules: A carbon market behavior impact data acquisition module is used to acquire carbon market behavior impact data, wherein the carbon market behavior impact data includes market fluctuation impact data, climate fluctuation impact data, and energy fluctuation impact data; The carbon market behavior risk model generation module is used to establish a carbon market behavior risk model based on the LSTM neural network according to the carbon market behavior impact data and generate the carbon market behavior risk coefficient; The carbon market behavior risk assessment module is used to obtain the carbon market behavior risk coefficient in different time periods, and obtain the carbon market behavior risk range based on statistical methods, obtain the carbon market behavior risk level grade based on the carbon market behavior risk coefficient and the carbon market behavior risk level grade, and determine the decision on carbon market behavior through the carbon market behavior risk level grade.

Images