Real-time electricity market transaction method based on Shapley value

Abstract

The invention discloses a real-time electricity market transaction method based on a Shapley value. The real-time electricity market transaction method comprises the steps of S1, collecting various data in an electricity market in real time; s2, calculating the marginal contribution of each participant under different cooperation combinations, and evaluating the contribution of each participant to the total transaction income of the market by calculating a Shapley value to obtain a fair distribution scheme; s3, performing real-time income distribution based on the Shapley value of each participant; s4, according to the Shapley value of each participant and the market demand, real-time income payment is carried out, settlement operation is completed, and a transaction result is fed back in real time; and S5, by accumulating market data and carrying out dynamic optimization, carrying out regular adjustment on a Shapley value calculation and income distribution algorithm. Through scientific modeling and an optimization algorithm, the purpose of fairly and efficiently distributing transaction shares and earnings of each participant in the electricity market is achieved.

Description

Technical Field

[0001] The invention relates to an electricity market transaction mechanism, in particular to a real-time electricity market transaction method based on Shapley value. Technical Background

[0002] Power market reforms are gradually driving the reform of power pricing mechanisms, moving towards marketization. Pricing mechanisms such as cost-weighted average pricing and a dual-track system are being implemented, gradually guiding electricity prices to adjust dynamically in response to changes in market supply and demand.

[0003] In electricity markets with a high proportion of renewable energy, the use of the LMP mechanism for system surplus distribution faces some special challenges, while the Shapley value mechanism provides a fair and theoretically superior surplus distribution method.

[0004] Under the LMP mechanism, the electricity market determines marginal electricity prices at different nodes through market clearing, and participants trade electricity based on these prices. System surplus typically comes from producer surplus and consumer surplus. Producer surplus is the difference between the revenue generated by electricity producers and their production costs. Consumer surplus is the difference between the price consumers are willing to pay and the price they actually pay. However, with a high proportion of renewable energy, LMP prices can fluctuate more dramatically due to the volatile production of renewable energy sources such as wind and solar. This can lead to price uncertainty, especially during peak demand periods and when renewable energy generation is insufficient. Fluctuations in renewable energy generation can cause market supply and demand imbalances, further exacerbating LMP pricing instability. In this scenario, the calculation of system surplus under the LMP mechanism is similar to that of traditional electricity markets, but it must take into account several key characteristics: the marginal cost of renewable energy is near zero, but its intermittent and unpredictable nature can complicate grid scheduling. Power markets with a high proportion of renewable energy can experience price volatility and divergence, especially when wind and solar output is low. Traditional generators become the price setters in the electricity market, leading to significant fluctuations in LMP prices.

[0005] As renewable energy sources like distributed photovoltaics gain popularity, current electricity market trading mechanisms often face challenges with unfair resource allocation, conflicts of interest, and low market efficiency when handling multi-party transactions. Traditional trading mechanisms often rely on fixed rules to distribute market benefits, lacking flexibility and fairness. A market trading mechanism approach that effectively balances the interests of all participants and promotes collaboration and optimization is urgently needed. Shapley value, a game-theoretic allocation method, can provide fair distribution based on contribution, making it particularly suitable for dynamic, real-time electricity market environments. Summary of the Invention

[0006] The present invention aims to overcome the defects of the prior art and provide a real-time electricity market trading method based on Shapley value, which achieves the purpose of fairly and efficiently allocating the trading share and benefits of each participant in the electricity market through scientific modeling and optimization algorithm.

[0007] The present invention provides a real-time electricity market transaction method based on Shapley value, comprising the following steps:

[0008] Step S1: Real-time collection of various data in the power market, including total market demand, power supply of each participant, load changes, transaction prices, grid status, and power consumption of participants;

[0009] Step S2: Calculate the marginal contribution of each participant under different cooperation combinations, and evaluate its contribution to the overall market transaction revenue by calculating the Shapley value to arrive at a fair distribution plan;

[0010] Step S3: real-time profit distribution based on each participant's Shapley value;

[0011] Step S4: Based on each participant's Shapley value and market demand, the proceeds will be paid in real time, settlement operations will be completed, and real-time feedback on the transaction results will be provided;

[0012] Step S5: Regularly adjust the Shapley value calculation and profit distribution algorithm by accumulating market data and performing dynamic optimization.

[0013] In the above-mentioned real-time electricity market trading method based on Shapley values, in step S1, various types of data are collected in real time through smart meters, sensors and market data platforms, and the electricity trading behavior of each participant is monitored in real time; through high-frequency data collection, the electricity trading behavior and status of market participants including each power plant, load user, and grid operator are obtained in real time; through monitoring of the emission coefficient of each power plant, carbon emission information in the market is obtained, forming a dynamic electricity market data model to provide accurate input data for subsequent Shapley value calculations.

[0014] In the above-mentioned real-time electricity market trading method based on Shapley values, step S1 also includes short-term forecasting of market demand, combining historical data, weather forecasts, grid status and external environmental factors, and predicting future electricity demand fluctuations through data mining and machine learning algorithms.

[0015] In the above-mentioned real-time electricity market trading method based on Shapley values, in step S2, the marginal contribution of each participant under different cooperation combinations is calculated. That is, by traversing all possible cooperation combinations, the calculated Shapley value can reflect how different participants in the market interact with each other, as well as the factors that affect the matching of electricity supply and demand, and ultimately contribute to the total market revenue. In this way, the deviations existing in traditional allocation methods can be avoided to ensure that the contributions of market participants are reasonably evaluated.

[0016] In the above-mentioned real-time electricity market trading method based on Shapley values, in step S3, the basis for profit distribution is the Shapley value of each participant, ensuring that they obtain a fair share of profits based on their marginal contributions. At the same time, further optimization is performed based on the market's electricity demand, market prices, and the carbon emissions of each participant, that is, by adjusting the electricity trading price and dispatching the power generation of power plants to maximize the overall benefits of the market. During the optimization process, its goal is not only to maximize profits, but also to include a comprehensive consideration of market fairness, carbon emission control, and grid load balance.

[0017] In the above-mentioned real-time electricity market trading method based on Shapley values, step S4 includes: making real-time payments of proceeds and completing settlement operations, while feeding back the transaction results to each participant to ensure the transparency and fairness of the transaction process. Among them, through the feedback mechanism, the market can respond to changes in supply and demand and price fluctuations in a timely manner and adjust trading strategies, thereby ensuring that the market remains fair and balanced.

[0018] In the above-mentioned real-time electricity market trading method based on Shapley value, in step S5, the Shapley value calculation and profit distribution algorithm are regularly adjusted by analyzing historical trading data, user feedback and market fluctuations, wherein the dynamic optimization process includes adjusting the contribution evaluation method of each participant according to market demand, participant behavior and rule changes.

[0019] The above-mentioned real-time electricity market trading method based on Shapley values also includes long-term optimization, during which the electricity price and power generation in the electricity market need to be adjusted to cope with changes in long-term market trends.

[0020] In the above-mentioned real-time electricity market trading method based on Shapley value, in the long-term optimization, the price strategy, load dispatching rules and electricity trading mechanism are also adjusted according to the long-term market demand fluctuations to achieve the sustainable development of the electricity market.

[0021] In the above-mentioned real-time electricity market trading method based on Shapley values, during long-term optimization, it was found through long-term tracking and feedback that the price fluctuations of participants were large. Based on this feedback, the relevant parameters were adjusted to further optimize the profit distribution and trading rules.

[0022] The beneficial effects of the present invention are: the present invention utilizes scientific modeling and optimization methods to achieve fair and transparent distribution of transaction shares and benefits of each participant in the market, providing fair distribution based on contribution for all parties, and is particularly suitable for dynamic and real-time electricity market environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a flow chart of the real-time electricity market transaction method of the present invention. DETAILED DESCRIPTION

[0024] The present invention will be further described below with reference to the accompanying drawings.

[0025] See also Figure 1 The present invention provides a real-time electricity market transaction method based on Shapley value, comprising the following steps:

[0026] Step S1: First, the system needs to collect various data in the power market in real time, including total market demand, power supply of each participant, load changes, transaction prices, grid status, etc. Through smart meters, sensors and market data platforms, the power trading behavior of each participant is monitored in real time;

[0027] Step S2: The system calculates the marginal contribution of each participant under different cooperation combinations. By traversing all possible cooperation combinations, the Shapley value can take into account how different participants in the market interact with each other, how they affect the matching of electricity supply and demand, and ultimately contribute to the total market revenue. In the electricity market, the core goal of calculating the Shapley value is to come up with a fair distribution plan by considering the contribution of each market participant, such as power plants, electricity consumers, and trading platforms, to the overall market transaction revenue;

[0028] Step S3: Once the Shapley value calculation is completed, the system will distribute the revenue in real time based on the Shapley value of each participant. By using the Shapley value as the basis for revenue distribution, the system ensures that each participant in the electricity market receives a fair share of revenue based on their marginal contribution.

[0029] Step S4: The system will pay the proceeds in real time and complete the settlement operation based on each participant's Shapley value and market demand. In addition, the system will provide real-time feedback on the transaction results and feedback it to each market participant to ensure the transparency and fairness of the transaction process.

[0030] Step S5: As the market operates, the system continuously accumulates market data and performs dynamic optimization. By analyzing historical transaction data, user feedback, and market fluctuations, the system regularly adjusts the Shapley value calculation and profit distribution algorithm. During the optimization process, the system adjusts the contribution evaluation method for each participant based on market demand, participant behavior, and changes in rules, such as policy changes, to ensure the fairness and efficiency of market transactions.

[0031] In the above invention, the specific implementation of step S1 is as follows:

[0032] Step S11: First, various data from the electricity market are collected in real time through smart meters, sensors, and the market data platform. This data includes information such as total market demand, power supply provided by each participant, load changes, market transaction prices, grid status, and participant power consumption. Through high-frequency data collection, the system can obtain real-time information on the power trading behavior and status of each market participant, including power plants, load users, and grid operators. Furthermore, the system also monitors carbon emissions information within the market, including the emission coefficients of each power plant and carbon credit incentive policies. This information is crucial for subsequent Shapley value calculations and revenue distribution. Using this collected real-time data, the system can form a dynamic electricity market data model, providing accurate input data for subsequent Shapley value calculations. For example, at a given moment, assuming total market demand is 500MW, the system collects real-time power generation information from each power plant through smart meters—for example, Power Plant A provides 100MW, Power Plant B provides 200MW, and so on—as well as grid load changes. This data forms the basic input for calculating the Shapley value. Real-time data acquisition provides the real-time input data required in power market transactions, ensuring that the system can respond to market changes in real time, thereby providing accurate and timely information for subsequent Shapley value calculation and revenue distribution.

[0033] Step S12: Based on real-time data collection, the system also performs short-term market demand forecasts. This process combines historical data, weather forecasts, grid conditions, and external environmental factors such as seasonal variations and temperature to predict electricity demand fluctuations within a certain timeframe. Using data mining and machine learning algorithms, the system analyzes historical load data and combines it with weather forecasts, such as temperature and humidity, and other environmental data to predict electricity demand for the next few hours or days. This forecast not only helps grid operators better dispatch power resources but also provides more accurate market demand information for subsequent Shapley value calculations. For example, if market demand is predicted to increase to 600MW within the next few hours, the system will prepare for power dispatch in advance based on the predicted load fluctuations and provide accurate market demand data for Shapley value calculations. This demand forecast helps the system estimate participants' contributions to overall market demand, ensuring fair distribution of benefits. Market demand forecasting and data processing accurately predict future market demand fluctuations, providing a basis for system optimization decisions. Forecasting can help grid operators dispatch resources in advance and provide future market demand data for Shapley value calculations, reducing uncertainty and risk and ensuring the fairness and efficiency of market transactions.

[0034] In the present invention described above, the specific implementation of step S2 is as follows:

[0035] Step S21, Game theory, also known as strategy theory, is an important branch of modern mathematics and has been widely used in the fields of power system planning, scheduling and power market. The classic cooperative game problem can be represented by (N, v), where N refers to the participating members and v refers to the characteristic function. In the problem studied in this paper, N is all the generators or loads in the system, and v(N) represents the total surplus generated by the system when the entire alliance exists in the system. Let S represent the sub-alliance, that is, a non-empty true subset of N, and there is v(S) represents the surplus generated by the system when only the participating members in the alliance S exist in the system, and accordingly It is worth noting that for the power system, only when there is at least one generator set and one load in the sub-alliance can the system have power generation behavior and v(S) may not be 0. This also reflects the interdependent relationship between generator sets and loads in the power system.

[0036] Step S22: The Shapley value is a fair distribution method in game theory that fairly calculates each participant's share of revenue based on their marginal contribution to the collaboration. In the electricity market, the core goal of Shapley value calculation is to arrive at a fair distribution plan by considering the contribution of each market participant (such as power plants, electricity consumers, and trading platforms) to the overall market transaction revenue.

[0037] The Shapley value is calculated as follows:

[0038] Consider a cooperative game (N,v), where N = {1,2,…,n, represents the set of participants, is the characteristic function that assigns a value to each coalition v(S) represents the surplus generated by the system when only the participating members in the alliance S exist in the system.

[0039] It is defined as the weighted average of the marginal contributions of member i, i∈N in all possible coalitions:

[0040]

[0041] where S is the subset excluding participant i, |S| is the size of the coalition, v(S∪{i})-v(S) is the marginal contribution of participant i, and |S|!(n-|S|-1)! / n! is the probability of random ordering of participants.

[0042] The Shapley value is unique, that is, it can allocate a unique surplus to each participating member. At the same time, it takes into account the possibility of participating members appearing in each sub-alliance and the order of their appearance in each sub-alliance.

[0043] Specifically, the system calculates the marginal contribution of each participant under different cooperation combinations. By traversing all possible cooperation combinations, the Shapley value takes into account how different market participants interact, how they influence the matching of electricity supply and demand, and ultimately how they contribute to the total market revenue. In this way, the system avoids the potential biases of traditional allocation methods and ensures that the contributions of market participants are reasonably assessed.

[0044] In the above invention, the specific implementation of step S3 is as follows:

[0045] Step S31: Once the Shapley value calculation is complete, the system will distribute revenue in real time based on each participant's Shapley value. By using the Shapley value as the basis for revenue distribution, the system ensures that each participant in the electricity market receives a fair share of revenue based on their marginal contribution.

[0046] Assume that the total market revenue is R total , calculated based on the market electricity price and power consumption, the income of each participant i is distributed according to its Shapley value:

[0047]

[0048] Where: R iis the payoff that participant i should receive based on his Shapley value, It is the weighted average of the marginal contributions of all members.

[0049] Step S32. At this time, the system will further optimize according to the market's electricity demand, market price and the carbon emissions of each participant. Specifically, the system will maximize the overall benefits of the market by adjusting the electricity transaction price and dispatching the power generation of power plants. During the optimization process, the system considers not only the maximization of benefits, but also the comprehensive consideration of market fairness, carbon emission control and grid load balance. For example, if the market electricity price is 0.5 yuan per kilowatt-hour in a certain period of time, and the Shapley value of a power plant is higher, the system will adjust the electricity price or power generation distribution to ensure that the power plant obtains the due benefits according to its contribution. At the same time, the system will further adjust the price strategy in combination with the carbon credit incentive mechanism.

[0050] In the above invention, the specific implementation of step S4 is as follows:

[0051] Step S41: After the power transaction is completed, the system will settle the transaction based on the Shapley value calculation results. Specifically, the system will pay the proceeds in real time based on each participant's Shapley value and market demand, and complete the settlement operation. In addition, the system will provide real-time feedback on the transaction results and feedback to each market participant to ensure the transparency and fairness of the transaction process.

[0052] Step S42: The market feedback mechanism enables the system to adjust based on real-time trading results. This feedback mechanism allows the market to promptly respond to changes in supply and demand, price fluctuations, and other issues. For example, if the power supply of certain participants changes, the system can adjust revenue distribution based on real-time data, thereby ensuring a fair and balanced market. For example, if a power plant's power generation decreases due to weather during a transaction, the system will adjust its revenue distribution in real time based on market feedback to ensure that its contribution to the market is reasonably compensated.

[0053] The purpose of the market feedback mechanism is to adjust trading strategies based on market conditions. Assuming the feedback factor is λ, the adjustment of the feedback mechanism can be expressed as follows:

[0054]

[0055] Where: p i (t+1) is the electricity price of participant i at the next moment, p i (t) is the electricity price of participant i at the current moment, is the expected payoff of participant i (predicted value), R i(t) is the actual payoff of participant i.

[0056] In the present invention described above, the specific implementation of step S5 is as follows:

[0057] Step S51: As the market operates, the system continuously accumulates market data and performs dynamic optimization. By analyzing historical transaction data, user feedback, and market fluctuations, the system regularly adjusts the Shapley value calculation and profit distribution algorithms. During the optimization process, the system adjusts the contribution evaluation method for each participant based on market demand, participant behavior, and policy changes to ensure the fairness and efficiency of market transactions.

[0058] In long-term optimization, the system needs to adjust the electricity price and power generation in the power market to cope with changes in the long-term market trend. Assuming that the long-term goal of the power market is to maximize overall benefits, the optimization objective function can be expressed as:

[0059]

[0060] Among them: maximize is the maximum value of the optimization target, is the weighted average of all members’ marginal contributions, R i is the payoff that participant i should receive based on his Shapley value, γ i Load i is the impact of load scheduling on the revenue of participant i, γ i is the weight.

[0061] Step S52: Long-term optimization includes not only adjustments to the Shapley value algorithm but also optimization of trading rules and strengthening of carbon emission management. For example, the system may adjust pricing strategies, load dispatch rules, and power trading mechanisms based on long-term market demand fluctuations to achieve sustainable development of the power market. For example, if the system discovers through long-term tracking and feedback that certain power plants are experiencing significant price fluctuations, potentially adversely affecting market fairness, it will adjust relevant parameters based on this feedback to further optimize revenue distribution and trading rules.

[0062] In summary, the innovation of the present invention lies in the Shapley allocation mechanism and feedback mechanism.

[0063] The Shapley allocation mechanism is based on the traditional Shapley value allocation method of the present invention, and further integrates real-time multi-dimensional data of the electricity market, including load forecasting, carbon emission monitoring, price fluctuations, etc., so as to dynamically and finely evaluate the marginal contribution of each participant in the market. Unlike the traditional method of only considering fixed or historical data, this mechanism can establish a more accurate contribution measurement between multiple energy types, such as renewable energy and conventional thermal power, and diversified market entities, such as the power generation side, the load side and the scheduling side. On this basis, each participant will obtain differentiated benefit distribution according to real-time contribution, breaking the previous rigid thinking of "whoever generates electricity gets the benefit", and promoting the collaboration of multiple parties such as clean energy, load response and scheduling optimization to maximize their value. At the same time, the mechanism combines the fairness principle of game theory distribution with the actual constraints of power grid operation, so that the market can ensure the rationality and sustainability of distribution while pursuing the maximization of overall benefits, thereby improving the resilience and economic benefits of the entire power system;

[0064] The feedback mechanism goes beyond simple post-hoc adjustments. Instead, it leverages high-frequency data collection and real-time calculations to integrate the differences between actual and predicted revenues for all parties, automatically adjusting the electricity price and trading strategy for the next moment through adjustable feedback factors. For example, when a significant deviation is detected between the actual revenue of certain power plants or users and their expected contributions, the system will revise the corresponding pricing or scheduling plan in real time based on the deviation and the overall market revenue. This can quickly bridge the uncertainty caused by price fluctuations and shorten the duration of market imbalances. In this iterative cycle, the feedback mechanism not only improves the transparency and flexibility of transaction settlement, but also encourages participants to optimize their supply and demand strategies through continuous adaptive adjustments, providing a new dynamic regulatory tool for maintaining long-term fairness and efficiency in the electricity market in the context of a high proportion of renewable energy access.

[0065] The present invention has been described in detail above with reference to the embodiments of the accompanying drawings. A person skilled in the art can make various modifications to the present invention based on the above description. Therefore, certain details in the embodiments should not be construed as limiting the present invention. The scope of protection of the present invention shall be determined by the scope defined in the appended claims.

Claims

1. A real-time electricity market trading method based on Shapley value, comprising the following steps: Step S1: Real-time collection of various data in the power market, including total market demand, power supply of each participant, load changes, transaction prices, grid status, and power consumption of participants; Step S2: Calculate the marginal contribution of each participant under different cooperation combinations, and evaluate its contribution to the overall market transaction revenue by calculating the Shapley value to arrive at a fair distribution plan; Step S3: real-time profit distribution based on each participant's Shapley value; Step S4: Based on each participant's Shapley value and market demand, the proceeds will be paid in real time, settlement operations will be completed, and real-time feedback on the transaction results will be provided; Step S5: Regularly adjust the Shapley value calculation and profit distribution algorithm by accumulating market data and performing dynamic optimization.

2. The real-time electricity market transaction method based on Shapley value according to claim 1, characterized in that: In step S1, various types of data are collected in real time through smart meters, sensors and market data platforms, and the electricity trading behavior of each participant is monitored in real time; through high-frequency data collection, the electricity trading behavior and status of market participants including each power plant, load user, and grid operator are obtained in real time; through monitoring of the emission coefficient of each power plant, carbon emission information in the market is obtained, forming a dynamic electricity market data model to provide accurate input data for subsequent Shapley value calculations.

3. The real-time electricity market transaction method based on Shapley value according to claim 1, characterized in that: The step S1 also includes short-term forecasting of market demand, combining historical data, weather forecasts, grid status and external environmental factors to predict future power demand fluctuations through data mining and machine learning algorithms.

4. The real-time electricity market transaction method based on Shapley value according to claim 2 or 3, characterized in that: In step S2, the marginal contribution of each participant under different cooperation combinations is calculated. That is, by traversing all possible cooperation combinations, the calculated Shapley value can reflect how different participants in the market interact with each other, as well as the factors that affect the matching of electricity supply and demand, and ultimately contribute to the total market revenue. In this way, the deviations existing in traditional allocation methods can be avoided to ensure that the contributions of market participants are reasonably evaluated.

5. The real-time electricity market transaction method based on Shapley value according to claim 4, characterized in that: In step S3, the basis for profit distribution is the Shapley value of each participant, ensuring that they obtain a fair share of profits based on their marginal contribution. At the same time, further optimization is performed based on the market's electricity demand, market price, and the carbon emissions of each participant. That is, by adjusting the electricity trading price and dispatching the power generation of power plants, the overall market benefits are maximized. During the optimization process, its goal is not only to maximize profits, but also to comprehensively consider market fairness, carbon emission control, and grid load balance.

6. The real-time electricity market transaction method based on Shapley value according to claim 1, characterized in that: Step S4 includes: paying the proceeds in real time and completing the settlement operation, while feeding back the transaction results to all participants to ensure the transparency and fairness of the transaction process. Through the feedback mechanism, the market can respond to changes in supply and demand, price fluctuations, and adjust trading strategies in a timely manner, thereby ensuring that the market remains fair and balanced. Assuming the feedback factor is λ, the adjustment of the feedback mechanism can be expressed by the following formula: Where: p i (t+1) is the electricity price of participant i at the next moment, p i (t) is the electricity price of participant i at the current moment, is the expected payoff of participant i, R i (t) is the actual payoff of participant i, and the total payoff of the market is R total .

7. The real-time electricity market transaction method based on Shapley value according to claim 1, characterized in that: In step S5, the Shapley value calculation and profit distribution algorithm are regularly adjusted by analyzing historical transaction data, user feedback and market fluctuations, wherein the dynamic optimization process includes adjusting the contribution evaluation method of each participant according to market demand, participant behavior and rule changes.

8. The real-time electricity market transaction method based on Shapley value according to claim 7, characterized in that: It also includes long-term optimization, during which the electricity price and power generation in the power market need to be adjusted to cope with changes in the long-term market trend. Assuming that the long-term goal of the power market is to maximize overall benefits, the optimization objective function can be expressed as: Among them: maximize is the maximum value of the optimization target, is the weighted average of all members’ marginal contributions, R i is the payoff that participant i should receive based on his Shapley value, γ i Load i is the impact of load scheduling on the revenue of participant i, γ i is the weight.

9. The real-time electricity market transaction method based on Shapley value according to claim 8, characterized in that: In long-term optimization, price strategies, load dispatching rules and power trading mechanisms are also adjusted according to long-term market demand fluctuations to achieve sustainable development of the power market.

10. The real-time electricity market transaction method based on Shapley value according to claim 8, characterized in that: During long-term optimization, we also discovered through long-term tracking and feedback that the price fluctuations of participants were large. Based on this feedback, we adjusted the relevant parameters and further optimized the profit distribution and trading rules.

Images