Multi-microgrid P2P energy transaction method and system for improving reputation management model

Abstract

The invention discloses a multi-microgrid P2P energy transaction method and system with an improved reputation management model, and belongs to the technical field of energy transaction, and the method comprises the steps: employing the improved reputation management model to calculate the reputation score of each microgrid in real time according to the history of multi-microgrid P2P energy transaction; the improvement of the reputation management model comprises the following steps: introducing a dynamic forgetting factor calculated according to a reputation score variation, and adjusting the weights of historical information and real-time information in the reputation score calculation; calculating a priority score of each micro-grid according to the reputation score, and calculating a coupling weight of a corresponding transaction through the priority scores of the micro-grids of the two transaction parties; and weighting the distributed optimization algorithm through the coupling weight, adjusting the priority of each micro-grid transaction, and generating a multi-micro-grid P2P energy transaction strategy. According to the invention, the problem that the multi-microgrid P2P energy transaction method is not high in reliability and is easy to cause energy waste can be solved.

Description

Technical Field

[0001] This invention belongs to the field of energy trading technology, specifically relating to a multi-microgrid P2P energy trading method and system with an improved reputation management model. Background Technology

[0002] Multi-microgrid P2P energy trading refers to a model where multiple microgrids directly trade energy, without the need for a traditional centralized trading center. Each microgrid can act as an independent entity, engaging in peer-to-peer transactions with other microgrids based on its own energy supply and demand. The multi-microgrid P2P energy trading market is a decentralized trading platform where participants trade directly, lacking a central authority to guarantee fairness and trust. Therefore, reputation management models are crucial, establishing trust mechanisms by evaluating participants' historical behavior to facilitate smooth transactions.

[0003] However, traditional reputation management models in multi-microgrid P2P energy trading have several shortcomings. First, static scoring mechanisms cannot reflect participants' latest behavior in real time, leading to a disconnect between reputation scores and actual performance. Second, single-dimensional assessment ignores key factors such as the severity of events, resulting in a lack of comprehensiveness in scoring. Furthermore, traditional reputation management models are easily manipulated; for example, participants may use "whitewashing" behavior to boost their reputation scores—a practice involving multiple small, compliant transactions to quickly improve their reputation and cover up past misconduct. Finally, poor adaptability makes traditional multi-microgrid P2P energy trading methods ill-suited to cope with market changes and the dynamics of participant behavior, resulting in low reliability of energy transactions, potential energy waste, and limited application in complex energy trading environments. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide an improved reputation management model for multi-microgrid P2P energy trading methods and systems, thereby solving the problems of low reliability and easy energy waste in multi-microgrid P2P energy trading methods.

[0005] To solve the above-mentioned technical problems, the present invention is implemented using the following technical solution:

[0006] In a first aspect, the present invention provides an improved reputation management model for multi-microgrid P2P energy trading, comprising:

[0007] Based on the P2P energy trading history of the multi-microgrid, an improved reputation management model is used to calculate the reputation score of each microgrid in real time. The calculation of the reputation score of each microgrid includes: based on a pre-set dynamic forgetting factor calculated according to the change in reputation score and the P2P energy trading history of the multi-microgrid, the historical positive and negative feedback accumulation and the real-time global positive and negative feedback increment are weighted and fused respectively to calculate the real-time positive and negative feedback accumulation of each microgrid, and the reputation score of each microgrid is updated according to the real-time positive and negative feedback accumulation.

[0008] The priority score of each micronetwork is calculated based on the reputation score, and the coupling weight of the corresponding transaction is calculated through the priority scores of the micronetworks of both parties to the transaction.

[0009] By weighting the distributed optimization algorithm with the coupling weights, the priority of each microgrid transaction is adjusted, and a multi-microgrid P2P energy trading strategy is generated.

[0010] The aforementioned improved reputation management model for multi-microgrid P2P energy trading, wherein the method calculates the real-time positive and negative feedback accumulation of each microgrid based on a pre-set dynamic forgetting factor calculated according to reputation score changes and the history of multi-microgrid P2P energy trading, weighted and fused with the historical positive and negative feedback accumulation and the real-time global positive and negative feedback increment, and updates the reputation score of each microgrid according to the real-time positive and negative feedback accumulation, includes:

[0011] Based on the P2P energy trading history of multi-microgrid, a dynamic forgetting factor calculated based on the change in reputation score is introduced in the Beta reputation model reputation score calculation. The cumulative positive and negative feedback at the previous moment and the global positive and negative feedback increment at the current moment are weighted and fused to calculate the cumulative positive and negative feedback of each microgrid at the current moment.

[0012] Micronet i Current Moment Positive feedback accumulation The calculation formula is:

[0013] ,

[0014] In the formula, Indicates the current moment of microgrid i The dynamic forgetting factor; Indicates the previous moment of the microgrid i The cumulative amount of positive feedback; Indicates the current moment of microgrid i The global positive feedback increment;

[0015] Micronet i Current Moment negative feedback accumulation The calculation formula is:

[0016] ,

[0017] In the formula, Indicates the previous moment of the microgrid i The cumulative amount of negative feedback; Indicates the current moment of microgrid i The global negative feedback increment;

[0018] Update the reputation score of each micronetwork for the next time step based on the current cumulative positive and negative feedback.

[0019] Micronet iNext Moment reputation score The calculation formula is:

[0020] ,

[0021] In the formula, Indicates the current moment of microgrid i The cumulative amount of positive feedback; Indicates the current moment of microgrid i The cumulative amount of negative feedback; and This represents the preset smoothing coefficient.

[0022] The aforementioned improved reputation management model for multi-microgrid P2P energy trading calculates a dynamic forgetting factor based on changes in reputation scores, including:

[0023] Based on the P2P energy trading history of multiple microgrids, calculate the short-term reputation fluctuations and long-term reputation trends of each microgrid.

[0024] Micronet i Current Moment Short-term reputation fluctuations The calculation formula is:

[0025] ,

[0026] In the formula, Indicates the current moment of microgrid i Reputation score; Indicates the previous moment of the microgrid i Reputation score;

[0027] Micronet i Current Moment Long-term reputation trend The calculation formula is:

[0028] ,

[0029] In the formula, Indicates the first seven moments of the microgrid i Reputation score;

[0030] The dynamic forgetting factor for each micronet at the current moment is calculated based on short-term reputation fluctuations and long-term reputation trends; micronet i at the current moment... Dynamic forgetting factor The calculation formula is:

[0031] , ,

[0032] In the formula, This represents the preset balance factor that adjusts the short-term weights; This represents the preset balance factor that adjusts the long-term weights; This indicates the preset sensitivity to short-term fluctuations. ; This indicates the preset sensitivity to long-term fluctuations. .

[0033] The aforementioned improved reputation management model for multi-microgrid P2P energy trading further includes the calculation of reputation scores for each microgrid, in addition to:

[0034] The system employs a comprehensive transaction weight calculated based on transaction size and risk. It then uses an asymmetric positive and negative feedback increment calculation method to calculate the positive and negative feedback increments separately. Finally, it calculates the global positive and negative feedback increments for each microgrid based on these positive and negative feedback increments.

[0035] The aforementioned improved reputation management model for multi-microgrid P2P energy trading calculates a comprehensive transaction weight based on transaction size and risk, including:

[0036] Based on the historical P2P energy transactions of Duowei.com, the normalized weight of the corresponding transaction size is calculated to measure the transaction size; at the current moment... microgrid With micro-network Normalized weights of transaction size The calculation formula is:

[0037] ,

[0038] ,

[0039] In the formula, Indicates the current time microgrid peer-to-peer microgrid The promised delivery volume; Indicates to microgrid A collection of micro-networks for transactions; Represents any time from the initial time to time t; microgrid Largest transaction commitment in history;

[0040] Based on the peer microgrid's reputation score, calculate the corresponding microgrid reputation risk weight for the transaction to measure transaction risk; at the current moment... microgrid With the peer microgrid Micro-network reputation risk weighting of transactions The calculation formula is:

[0041] ,

[0042] In the formula, This represents the preset risk amplification factor; Indicates to microgrid Micronet on the other end of the transaction Current moment The overall risk level of default; Indicates the peer microgrid Current moment Reputation score;

[0043] The overall transaction weight is calculated based on the normalized weight of the transaction commitment amount and the microgrid reputation risk weight based on the peer microgrid reputation score; at the current moment... microgrid With the peer microgrid Overall weight of transactions The calculation formula is:

[0044] .

[0045] The aforementioned improved reputation management model for multi-microgrid P2P energy trading, wherein the asymmetric positive and negative feedback incremental calculation method includes:

[0046] Current moment Positive feedback increment of transactions between microgrid i and its counterpart microgrid j The calculation formula is:

[0047] ,

[0048] In the formula, Indicates the preset positive feedback index; Indicates the current time micronet peer-to-peer microgrid The fulfillment rate; the current time micronet peer-to-peer microgrid fulfillment rate The calculation formula is:

[0049] ,

[0050] In the formula, Indicates the current time microgrid peer-to-peer microgrid Committed delivery volume Indicates the current time microgrid peer-to-peer microgrid The actual delivery volume;

[0051] Current moment Negative feedback increment of transactions between microgrid i and its counterpart microgrid j The calculation formula is:

[0052] ,

[0053] In the formula, This indicates the preset negative feedback index.

[0054] The aforementioned improved reputation management model for multi-microgrid P2P energy trading is characterized by the calculation of the global positive and negative feedback increments of each microgrid, including:

[0055] The positive and negative feedback increments of each microgrid at the current moment in the global transaction are summarized respectively, and the global positive and negative feedback increments of each microgrid at the current moment are obtained accordingly.

[0056] Micronet i Current Moment Global positive feedback increment The calculation formula is: ,

[0057] Micronet i Current Moment Global negative feedback increment The calculation formula is: .

[0058] The aforementioned improved reputation management model for multi-microgrid P2P energy trading, wherein calculating the priority score of each microgrid based on the reputation score, and calculating the coupling weight of the corresponding transaction through the priority scores of the microgrids of both trading parties, includes:

[0059] Calculate the priority score for each micronet at the corresponding time based on its reputation score. For micronet i at the current time... Priority score The calculation formula is:

[0060] ,

[0061] In the formula, This represents the preset reputation amplification factor;

[0062] The coupling weight of the corresponding transaction is obtained by multiplying the priority scores of the two micronetworks involved in the transaction. Micronetwork i and the peer micronetwork j are the two parties in the transaction, and the coupling weight of the transaction between micronetwork i and the peer micronetwork j is calculated accordingly. The calculation formula is:

[0063] ,

[0064] In the formula, This indicates that the peer micronetwork j is at the current time. Priority score.

[0065] The aforementioned multi-microgrid P2P energy trading method is characterized in that the step of adjusting the priority of each microgrid transaction by weighting the distributed optimization algorithm through the coupling weights to generate a multi-microgrid P2P energy trading strategy includes:

[0066] The distributed optimization algorithm is the C-ADMM algorithm; the consensus penalty term of the C-ADMM algorithm is weighted by the coupling weight of the corresponding transaction, wherein the coupling weight is positively correlated with the priority of the corresponding microgrid transaction; by adjusting the priority of each microgrid transaction, a multi-microgrid P2P energy trading strategy is generated.

[0067] Local variables of microgrid i in C-ADMM algorithm In the iterative formula, the consensus penalty term after weighting the coupling weights is:

[0068] ,

[0069] In the formula, This represents the original consensus penalty term of the C-ADMM algorithm; Indicates the penalty factor; Let represent the common variables in the transaction between microgrid i and its counterpart microgrid j during the k-th iteration; Let represent the dual variable of the transaction between microgrid i and its counterpart microgrid j in the k-th iteration.

[0070] Secondly, the present invention provides a multi-microgrid P2P energy trading system with an improved reputation management model, comprising: a reputation calculation module, a weight generation module, and a strategy generation module;

[0071] The reputation calculation module is used to: calculate the reputation score of each microgrid in real time based on the P2P energy transaction history of the multi-microgrid using an improved reputation management model; the improvement of the reputation management model includes: introducing a dynamic forgetting factor calculated based on the change in reputation score, and adjusting the weight of historical information and real-time information in the reputation score calculation to improve the timeliness of reputation management.

[0072] The weight generation module is used to: calculate the priority score of each micro-network based on the reputation score, and calculate the coupling weight of the corresponding transaction based on the priority scores of the micro-networks of both parties to the transaction;

[0073] The strategy generation module is used to: weight the distributed optimization algorithm through the coupling weights, adjust the priority of each microgrid transaction, and generate a multi-microgrid P2P energy trading strategy.

[0074] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0075] This invention improves the reputation management model of a multi-microgrid P2P energy trading method by introducing a dynamic forgetting factor calculated based on changes in reputation scores. This adjusts the weights of historical and real-time information in the reputation score calculation, enabling the improved reputation management model to retain the memory of each microgrid's historical energy trading behavior and to evaluate the current energy trading behavior of microgrids in real time. This improves the timeliness of reputation management and, consequently, the reliability of evaluating the energy trading behavior of each microgrid, thus solving the problem of low reliability and energy waste inherent in multi-microgrid P2P energy trading methods.

[0076] This invention retains the underlying mathematical framework of reputation mean in the Beta reputation model, introduces a dynamic forgetting mechanism based on P2P energy transaction history to improve the timeliness of reputation management, introduces a comprehensive transaction weight based on energy transaction scale and transaction risk to measure the importance of each energy transaction, and introduces asymmetric positive and negative feedback incremental calculation based on the comprehensive transaction weight to improve the ability to identify energy transaction risks.

[0077] The reputation assessment of this invention has better timeliness and accuracy, and its assessment perspective is more comprehensive. It has a better resistance to malicious behaviors such as reputation laundering. The entire system is more efficient, stable and reliable than other P2P energy trading systems. Attached Figure Description

[0078] Figure 1 This is a schematic diagram of a multi-microgrid P2P energy trading method based on an improved reputation management model according to Embodiment 1 of the present invention;

[0079] Figure 2 This is a schematic diagram of the reputation scores and fulfillment rates of five microgrids under different fault scenarios in Embodiment 1 of the present invention;

[0080] Figure 3 This is based on the traditional method in Embodiment 1 of the present invention. A schematic diagram of 24-hour energy trading in response to microgrid 5 whitewashing behavior using a reputation model-based energy trading method;

[0081] Figure 4This is a schematic diagram of 24-hour energy trading in response to microgrid P2P energy trading methods using an improved reputation management model according to Embodiment 1 of the present invention, which addresses the whitewashing behavior of microgrid 5. Detailed Implementation

[0082] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments and specific features in the embodiments are detailed descriptions of the technical solution of the present application, rather than limitations thereof. In the absence of conflict, the embodiments and technical features in the embodiments can be combined with each other.

[0083] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0084] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0085] Example 1:

[0086] Figure 1 This is a flowchart of the multi-microgrid P2P energy trading method of the improved reputation management model in Embodiment 1 of the present invention. This flowchart only illustrates the logical sequence of the method described in this embodiment. Provided there are no conflicts, different methods may be used in other possible embodiments of the present invention. Figure 1 Complete the steps shown or described in the order indicated.

[0087] The methods include:

[0088] S1: Based on the P2P energy trading history of the multi-microgrid, an improved reputation management model is used to calculate the reputation score of each microgrid in real time. The calculation of the reputation score of each microgrid includes: based on a pre-set dynamic forgetting factor calculated according to the change in reputation score and the P2P energy trading history of the multi-microgrid, the historical positive and negative feedback accumulation and the real-time global positive and negative feedback increment are weighted and fused respectively to calculate the real-time positive and negative feedback accumulation of each microgrid, and the reputation score of each microgrid is updated according to the real-time positive and negative feedback accumulation.

[0089] S2: Calculate the priority score of each micro-network based on the reputation score, and calculate the coupling weight of the corresponding transaction through the priority scores of the micro-networks of both parties to the transaction;

[0090] S3: By weighting the distributed optimization algorithm with the coupling weights, the priority of each microgrid transaction is adjusted to generate a multi-microgrid P2P energy trading strategy.

[0091] The specific implementation of each step is described below:

[0092] The reputation model, also known as the Beta reputation model, uses the number of positive feedback signals from the Beta distribution. and the number of negative feedback This study examines the interactions between related entities and quantifies their reputation using statistics from the Beta distribution. The Beta distribution statistics include the reputation mean, the most commonly used reputation metric, which represents the average trustworthiness of an entity. The formula for calculating the reputation mean R is: ;

[0093] In the Beta reputation model, the incremental calculation of positive and negative feedback is symmetrical. For example, if a new positive feedback record is added, the increment is 1. Correspondingly, for each new negative feedback record, the increment is 1. Asymmetric positive and negative feedback increment calculation refers to using different calculation methods for positive and negative feedback increments.

[0094] This embodiment improves upon... Reputation Model:

[0095] This embodiment retains the underlying mathematical framework of reputation mean in the Beta reputation model, introduces a dynamic forgetting mechanism based on P2P energy trading history (Formulas 1 and 2, Formulas 5-8) to improve the timeliness of reputation management; introduces a comprehensive transaction weight based on transaction size and transaction risk to measure the importance of the transaction (Formulas 9-12); and introduces asymmetric positive and negative feedback incremental calculation based on the comprehensive transaction weight to improve the ability of reputation management to identify transaction risks (Formulas 13 and 15).

[0096] Step S1 includes:

[0097] Based on the P2P energy trading history of multi-microgrid, a dynamic forgetting factor calculated based on the change in reputation score is introduced in the Beta reputation model reputation score calculation. The cumulative positive and negative feedback at the previous moment and the global positive and negative feedback increment at the current moment are weighted and fused to calculate the cumulative positive and negative feedback of each microgrid at the current moment.

[0098] Micronet i Current Moment Positive feedback accumulation The calculation formula is:

[0099] ,(1)

[0100] In the formula, Indicates the current moment of microgrid i The dynamic forgetting factor; Indicates the previous moment of the microgrid i The cumulative amount of positive feedback; Indicates the current moment of microgrid i The global positive feedback increment;

[0101] Micronet i Current Moment negative feedback accumulation The calculation formula is:

[0102] (2),

[0103] In the formula, Indicates the previous moment of the microgrid i The cumulative amount of negative feedback; Indicates the current moment of microgrid i The global negative feedback increment;

[0104] Positive feedback accumulation microgrid The cumulative amount of positive feedback at the current time t, the cumulative amount of positive feedback The larger the microgrid, the more likely it is to be a microgrid. There are more reliable and well-performing interaction records; the cumulative amount of negative feedback. microgrid The cumulative amount of negative feedback at the current time t, the cumulative amount of negative feedback The larger the microgrid, the more likely it is to be a microgrid. There are more high-risk or dishonest interaction records.

[0105] To achieve a smooth transition between preserving history and incorporating new evidence, the cumulative positive feedback is updated in each iteration. With negative feedback accumulation Afterwards, micronet Positive feedback accumulation With negative feedback accumulation All will be used to update the next moment of the micro-network. reputation score .

[0106] Update the reputation score of each micronet for the next time step based on the current cumulative positive and negative feedback. (Micronet i next time step) reputation score The calculation formula is:

[0107] (3)

[0108] In the formula, Indicates the current moment of microgrid i The cumulative amount of positive feedback; Indicates the current moment of microgrid i The cumulative amount of negative feedback; and This represents the preset smoothing term coefficient. In this embodiment, The value is 1. The value is 2, and the next time step of microgrid i is... reputation score The calculation formula is:

[0109] (4)

[0110] In the formula, the +1 smoothing term in the numerator and the +2 smoothing term in the denominator are used for the positive feedback accumulation in the initial stage. With negative feedback accumulation When the score is low, prevent reputation score from being low. Excessive fluctuations in values;

[0111] The +1 smoothing term in the numerator and the +2 smoothing term in the denominator are used for the positive feedback accumulation in the initial stage. With negative feedback accumulation When the score is low, prevent reputation score from being low. Excessive fluctuations in numerical values.

[0112] Reputation score of Weiwang i Represents a node A comprehensive quantification of performance and risk of default based on historical and current transaction feedback. The numerical range is ;when In other words, if positive feedback far outweighs negative feedback, then Microgrid i is considered highly reliable; when In other words, if there is more negative feedback, then Micronet is considered high-risk or untrustworthy.

[0113] In the improvement of the reputation management model, a dynamic forgetting factor is calculated based on the change in reputation score, including:

[0114] Based on the P2P energy trading history of multiple microgrids, calculate the short-term reputation fluctuations and long-term reputation trends of each microgrid.

[0115] Micronet i Current Moment Short-term reputation fluctuations The calculation formula is:

[0116] (5)

[0117] In the formula, Indicates the current moment of microgrid i Reputation score; Indicates the previous moment of the microgrid i Reputation score; Indicates the basic time period length, used to distinguish between short-term and long-term; short-term reputation fluctuations Indicates a short period of time (1) Reputation fluctuations and long-term reputation trends Indicates long-term (7 items) The average reputation trend;

[0118] Micronet i Current Moment Long-term reputation trend The calculation formula is:

[0119] (6)

[0120] In the formula, Indicates the first seven moments of the microgrid i Reputation score;

[0121] The dynamic forgetting factor is calculated based on the fluctuations in short-term reputation and long-term reputation trends. The greater the fluctuation, the smaller the dynamic forgetting factor calculated by formula (7), which means that the weight of historical data in formula (1) and formula (2) is reduced, and more historical data is forgotten.

[0122] Balancing short-term fluctuations and long-term trends, the current moment of microgrid i Dynamic forgetting factor The calculation formula is:

[0123] (7)

[0124] (8)

[0125] In the formula, By dynamically monitoring the rates of short-term and long-term reputation changes, the dynamic forgetting factor is adjusted in real time. ; This represents the preset balance factor that adjusts the short-term weights; This represents the preset balance factor that adjusts the long-term weights;

[0126] Will Setting it to a larger value can improve short-term response speed when there are significant short-term fluctuations. This indicates a short-term response, when short-term reputation fluctuates. Get bigger The value will decrease; Its role is to determine the magnitude of the impact of a short-term response. Setting a preset value to a large value is equivalent to setting a strategy that prioritizes short-term changes. Any subtle change will affect the final result. This will have a significant impact, thereby achieving the design goal of improving short-term response speed;

[0127] Will Setting it to a larger value can enhance long-term trust when the long-term trend is significant. Indicating long-term trust levels, it reflects the stability of a node's long-term performance, and the long-term reputation trend... Very small The value will be close to 1, indicating that it is trustworthy. Its function is to measure long-term trust in the dynamic forgetting factor. The components in, if Setting a preset value to a large value is equivalent to formulating a strategy that prioritizes long-term trends. Any subtle change will affect the final result. This will have a significant impact, thereby achieving the design goal of enhancing long-term trust.

[0128] This indicates the preset sensitivity to short-term fluctuations. ; This indicates the preset sensitivity to long-term fluctuations. If there are large short-term fluctuations, that is, short-term reputation fluctuations... If it gets bigger, then The dynamic forgetting factor is reduced and calculated. A decrease in size indicates accelerated forgetting; if the performance fluctuates over a long period, it indicates a long-term reputation trend. If it gets bigger, then The dynamic forgetting factor is reduced and calculated. A smaller value indicates accelerated forgetting; if there are no significant short-term fluctuations and the long-term trend remains stable, that is... and To become smaller, accordingly and If it increases, the dynamic forgetting factor... An increase in size indicates a slower rate of forgetting.

[0129] The dynamic forgetting factor is used to balance new and old data: when there are large fluctuations, the new data is used as the standard, and when the data is stable, the old data is retained.

[0130] In this embodiment, the improvement to the reputation management model, besides introducing a dynamic forgetting factor calculated based on changes in reputation score to adjust the weights of historical and real-time information in reputation score calculation to improve the timeliness of reputation management, also includes:

[0131] A comprehensive transaction weight calculated based on transaction size and risk is introduced as a parameter for calculating positive and negative feedback increments, and an asymmetric method for calculating positive and negative feedback increments is designed.

[0132] In the improvement of the reputation management model, a comprehensive weighting of transactions is calculated based on transaction size and risk to measure the importance of a transaction, including:

[0133] Based on the historical P2P energy transactions of Duowei.com, the normalized weight of the corresponding transaction size is calculated to measure the transaction size:

[0134] Measure the current moment microgrid peer-to-peer microgrid Transaction commitment volume This represents the largest transaction commitment in its entire history of foreign transactions. The proportion is the size normalized weight. Scale normalized weights The transaction volume of each micro-network is normalized and the weight of large transaction commitments is increased.

[0135] Current moment microgrid With the peer microgrid Normalized weights of transaction size The calculation formula is:

[0136] (9)

[0137] (10)

[0138] In the formula, Indicates the current time microgrid peer-to-peer microgrid The promised delivery volume; Indicates to microgrid A collection of micro-networks for transactions; Represents any time from the initial time to time t; microgrid Largest transaction commitment in history; The denominator is such that formula (9) calculates ;

[0139] Based on the reputation score of the peer micro-network, the reputation risk weight of the corresponding transaction is calculated to measure the transaction risk; that is, the global default risk of the peer micro-network is adjusted to obtain the micro-network reputation risk weight, wherein the global default risk of the peer micro-network is calculated based on the reputation score of the peer micro-network.

[0140] Current moment microgrid With the peer microgrid Micro-network reputation risk weighting of transactions The calculation formula is:

[0141] (11)

[0142] In the formula, This represents the preset risk amplification factor. This indicates that the counterparty in the transaction with Weinet is Weinet. Current moment The overall level of default risk. The other end microgrid The higher the reputation score, the better the peer micro-network. The lower the risk of default, the better; for end-microgrids Current moment reputation score The lower the value, the higher the reputation risk weight of the micro-network. The larger.

[0143] The overall transaction weight is calculated based on the normalized weight of the transaction commitment amount and the microgrid reputation risk weight based on the peer microgrid reputation score:

[0144] Microgrid reputation risk weight With size normalized weights Perform a product operation to obtain the overall transaction weight. At present microgrid With the peer microgrid Overall weight of transactions The calculation formula is:

[0145] (12)

[0146] Overall trading weight By combining transaction volume with peer microgrid risk, large-scale default transactions and transactions with high-risk peer microgrids are penalized.

[0147] In improving the reputation management model, an asymmetric positive and negative feedback incremental calculation method is designed, including:

[0148] Using the overall transaction weight as a parameter for calculating the positive and negative feedback increments, and combining it with nonlinear processing, positive feedback increments based on the performance rate and negative feedback increments based on the global default risk are generated respectively. The nonlinear processing in the calculation of the positive feedback increment refers to the use of a positive feedback index. This involves nonlinear suppression to smooth out performance rewards and prevent excessive reputation gains from a single perfect performance. (Current moment) Positive feedback increment of transactions between microgrid i and its counterpart microgrid j The calculation formula is:

[0149] (13)

[0150] In the formula, This represents the positive feedback exponent, used to suppress extreme increments. ; Indicates the current time micronet peer-to-peer microgrid The contract fulfillment rate;

[0151] Positive feedback incremental calculation emphasizes real-time collaboration and strengthens immediate incentives: real-time fulfillment rate High, then positive feedback increment Large scale can instantly reward good cooperative behavior from both parties in a transaction, incentivize micro-networks to maintain a high level of performance in real-time interactions, and promote the stability of short-term cooperation.

[0152] The current time micronet peer-to-peer microgrid fulfillment rate The calculation formula is:

[0153] (14)

[0154] In the formula, Indicates the current time microgrid peer-to-peer microgrid Committed delivery volume Indicates the current time microgrid peer-to-peer microgrid The actual delivery volume; The fulfillment rate is used to quantify the degree of partial fulfillment in real time. This indicates a serious breach of contract in the transaction. The smaller the value, the more significant the impact of default.

[0155] Nonlinear processing in the negative feedback increment calculation process refers to the use of negative feedback exponents. This is achieved through nonlinear amplification, which amplifies the risk penalty; at the current moment... Negative feedback increment of transactions between microgrid i and its counterpart microgrid j The calculation formula is:

[0156] (15)

[0157] In the formula, Indicates the negative feedback index. ; This indicates the global default risk level of the peer micronetwork j;

[0158] Negative feedback increment calculation is based on global risk assessment, improving the comprehensiveness of risk identification: negative feedback increment The calculation is based on the current time of the peer micronet j. Global default risk The calculation, rather than a symmetrical one, is based on the real-time default rate between microgrid i and its peer microgrid j. The calculation shows that this negative feedback incremental calculation can: avoid the random deviations of local real-time transactions and integrate overall transaction risk information;

[0159] To avoid the accidental deviations in local real-time transactions, the following measures should be taken: a single real-time default by microgrid i and its counterpart microgrid j may be an accidental event caused by a temporary technical failure, but the overall default risk of counterpart microgrid j reflects its historical comprehensive performance in all energy trading cooperation and can better reflect the essential trading risk level of counterpart microgrid j.

[0160] The comprehensive transaction risk information includes: the global default risk level integrates the transaction history of peer microgrid j with other microgrids, rather than just the local data between microgrid i and peer microgrid j. This allows it to capture the potential risks of peer microgrid j within the entire energy trading network. When peer microgrid j frequently defaults on other microgrids, even if peer microgrid i temporarily fulfills its obligations, the actual transaction risk of peer microgrid j remains high. This integrates comprehensive transaction risk information, and the real-time default rate between microgrid i and its counterpart microgrid j. Trading risks that do not reflect the overall picture.

[0161] Current moment microgrid peer-to-peer microgrid fulfillment rate The higher, that is At that time, positive feedback increment The larger the value, the higher the confidence level of the counterpart micronetwork j in trading with micronetwork i; at the current moment... peer microgrid Reputation Score Low, meaning the overall risk of default is relatively low. When it is large, the current time Negative feedback increment of transactions between microgrid i and its counterpart microgrid j The larger the value, the greater the penalty for micro-network i, in order to incentivize micro-network i to actively avoid transacting with high-risk micro-networks.

[0162] The asymmetric positive and negative feedback increment calculation balances the rewards of reputation management for real-time transactions and the penalties for transaction risks. It uses different calculation methods: the positive feedback increment is calculated based on the real-time fulfillment rate of micronet i and peer micronet j, and the negative feedback increment is calculated based on the global default risk of peer micronet j. The global default risk is calculated based on the reputation score of peer micronets, which reflects the overall historical situation, rather than the negative feedback increment calculated based on the real-time default rate corresponding to the real-time fulfillment rate of micronet i and peer micronet j. The real-time default rate = 1 - real-time fulfillment rate.

[0163] Positive feedback incremental computation focuses on real-time interaction, incentivizing good real-time collaboration; negative feedback incremental computation is based on global history, improving the comprehensiveness and stability of risk assessment; asymmetric positive and negative feedback incremental computation avoids the limitations of local real-time data, reduces misjudgments, and can combine real-time and historical data to balance collaboration incentives and risk prevention.

[0164] In formulas (1) and (2), the calculation of the global positive and negative feedback increments of each microgrid includes:

[0165] By summing the positive and negative feedback increments of each microgrid at the current moment in the global transaction, we can obtain the global positive and negative feedback increments of each microgrid at the current moment:

[0166] microgrid Current moment The global positive feedback increment of microgrid i is obtained by summing the positive feedback increments received from all transactions. Micronet i at the current moment Global positive feedback increment The calculation formula is:

[0167] (16)

[0168] microgrid Current moment The global negative feedback increment of microgrid i is obtained by summing the negative feedback increments received from all transactions. Micronet i at the current moment Global negative feedback increment The calculation formula is:

[0169] , (17).

[0170] Step S2 includes:

[0171] For each microgrid Based on its reputation score Computational microgrids At any moment Priority score , Defined as reputation score Power-law magnification, micro-network i at the current moment Priority score The calculation formula is:

[0172] (18)

[0173] In the formula, This represents the preset reputation amplification factor. The larger the micro-network and the higher its reputation score, the higher its priority score;

[0174] The coupling weight of the corresponding transaction is obtained by multiplying the priority scores of the two micronetworks involved in the transaction. Micronetwork i and the peer micronetwork j are the two parties in the transaction, and the coupling weight of the transaction between micronetwork i and the peer micronetwork j is calculated accordingly. The calculation formula is:

[0175] , (19)

[0176] In the formula, This indicates that the peer micronetwork j is at the current time. Priority score.

[0177] The micro-networks of both parties in the transaction constitute the transaction object. The coupling weight of the corresponding transaction is calculated based on the priority scores of the micro-networks of both parties.

[0178] Step S3 includes:

[0179] In one specific embodiment, the distributed optimization algorithm is the C-ADMM algorithm. The consensus penalty term of the C-ADMM algorithm is weighted by the coupling weight of the corresponding transaction to dynamically adjust the influence of each microgrid transaction in the global transaction optimization. The coupling weight is positively correlated with the priority of the corresponding microgrid transaction. By adjusting the priority of each microgrid transaction, a multi-microgrid P2P energy trading strategy is generated.

[0180] Coupling weights are used to adjust the tightness of consensus between trading parties. A larger coupling weight indicates tighter coupling and greater importance of consensus. A large coupling weight between two microgrids with high reputation scores means that there are incentives and opportunities for transactions between them, making it easier for these two microgrids to obtain energy transactions.

[0181] The C-ADMM algorithm gives the local variables of micronet i. Public variables dual variables and penalty parameters For each transaction object Iterate in sequence:

[0182] Local variables of microgrid i The iterative formula is:

[0183] (20)

[0184] In the formula, This represents the original consensus penalty term of the C-ADMM algorithm;

[0185] This represents the consensus penalty term after weighting by coupling weights;

[0186] Indicates the penalty factor; Indicates to microgrid A collection of micro-networks for transactions; Let represent the common variables in the transaction between microgrid i and its counterpart microgrid j during the k-th iteration; Let represent the dual variable of the transaction between microgrid i and its counterpart microgrid j in the k-th iteration.

[0187] Public variables The iterative formula is:

[0188] ,(twenty one),

[0189] Dual variables The iterative formula is:

[0190] ,(twenty two),

[0191] In the formula, k represents the iteration round; microgrid Objective function; local variables This indicates the planned energy trading volume of microgrid i; This represents the planned energy trading volume of microgrid i in the (k+1)th iteration; Represents the planned energy trading volume of the peer microgrid j in the (k+1)th iteration; common variables This represents the consensus transaction volume between micronet i and its peer micronet j; Represents the consensus transaction volume between micronet i and its peer micronet j in the (k+1)th iteration; dual variable Represents local variables and common variables Deviation penalties; Represents the local variables of the (k+1)th iteration. Common variables of the (k+1)th iteration Deviation penalties; Let represent the dual variable of the k-th iteration;

[0192] With its own optimal tendency, combined with the C-ADMM algorithm and through the design of the consensus mechanism, each micronetwork is forced to take into account the synergy with the trading partners when optimizing its own objective function, and finally achieves a dynamic balance between individual interests and group consensus.

[0193] To verify the effectiveness of the improved reputation management model of this invention in multi-microgrid P2P energy trading, the following simulation experiment was conducted:

[0194] Figure 2 The data shows the trends in reputation scores and fulfillment rates for the five microgrids under different failure scenarios. Figure 2 The three sub-plots on the left, from top to bottom, represent the reputation score curves of each microgrid over 28 days under normal operation, occasional failure, and continuous failure scenarios, reflecting the changes in reputation scores of different microgrids over 28 days. Figure 2 The three sub-charts on the right, from top to bottom, represent the 28-day fulfillment rate curves for each micro-network under normal operation, occasional failure, and continuous failure scenarios, respectively. They show the success rate fluctuations of each micro-network during the transaction process. The horizontal axis starts at 1, with 1 to 2 representing the first day. The occasional failure scenario involves setting an occasional failure in a micro-network on day 15, while the continuous failure scenario involves setting a continuous failure in a micro-network from day 15 to day 28. Figure 2 As shown in the subplot on the right, in the scenario of occasional failures on day 15, the fulfillment rate of the orange curve microgrid drops rapidly within a short period on days 15 and 16. Figure 2 As shown in the subplot on the left, the orange curve indicates a brief dip in the micro-network's reputation score on day 15, followed by a rapid recovery. And as... Figure 2 As shown in the lower right subplot, in the continuous failure scenario from day 15 to day 28, the fulfillment rate of the microgrid (orange curve) continued to decline for an extended period, corresponding to... Figure 2 As shown in the lower left subplot, the orange curve indicates that the microgrid's reputation score drops significantly from day 15 and remains at a low level, reflecting the long-term impact of prolonged outages on the microgrid's reputation score. Figure 2 This invention demonstrates that, under fault conditions, the improved reputation management model of this invention enables multi-microgrid P2P energy trading methods to identify and adapt to changes in microgrid behavior.

[0195] In a multi-microgrid P2P energy trading simulation experiment, microgrid 5 is a malicious microgrid with a low reputation score. Figure 3 For traditional A 24-hour energy trading diagram illustrating the reputation model's approach to addressing microgrid 5's whitewashing behavior. Figure 4 This is a schematic diagram of 24-hour energy trading in response to microgrid P2P energy trading methods for improving reputation management models in this invention, addressing the issue of microgrid 5 whitewashing behavior. Figure 3 and Figure 4 The bar chart represents the share of each microgrid participating in energy trading at each time point. Figure 3 and Figure 4 Only the energy trading volume of microgrids 1, 2, 3, 4, and 5 was counted. However, more than these five microgrids participate in energy trading, and energy trading does not only occur between microgrids 1, 2, 3, 4, and 5. Figure 3 and Figure 4 The displayed energy trading volume does not always remain balanced among microgrids 1, 2, 3, 4, and 5. For example, in the first hour, only microgrids 2 and 3 participate in energy trading, but the energy trading volume of microgrid 3 is greater than that of microgrid 2. According to Figure 3 and Figure 4 The comparison of energy trading volumes verifies the effectiveness of the multi-microgrid P2P energy trading method of the improved reputation management model of this invention in resisting malicious reputation whitewashing attacks.

[0196] In multi-microgrid P2P energy trading, the reputation score of each microgrid is positively correlated with its energy trading priority. Reputation whitewashing is the behavior of malicious microgrids with low reputation scores due to historical misconduct, which quickly and improperly improve their reputation scores by executing a large number of small-amount, compliant energy transactions to cover up their past misdeeds.

[0197] conventional Reputation models lack historical decay and effective mechanisms for historical decay and event weighting. Figure 3 As shown, from hour 1 to hour 9, microgrid 5 did not participate in energy trading. In the initial energy trading session in hour 10, microgrids 2, 3, 4, and 5 participated. At this time, the energy trading volumes of the four microgrids were not significantly different. This is because, based on conventional... The reputation model's energy trading approach does not retain a memory of Microgrid 5's past malicious behavior; Figure 4 As shown, in the initial energy transaction of microgrid 5 in the 10th hour, the energy transaction volumes of microgrids 2 and 3 at this time are... Figure 3 The energy trading volumes of microgrids 2 and 3 are the same, while the energy trading volumes of microgrids 4 and 5 are significantly smaller than those of microgrids 2 and 3. This is because the multi-microgrid P2P energy trading method of the improved reputation management model of this invention retains the memory of the historical malicious behavior of microgrid 5. When microgrids 4 and 5 trade, because microgrid 5 has a low reputation score, it is only allowed a smaller energy trading volume. Microgrid 4, which trades with microgrid 5, can also only obtain a smaller energy trading volume.

[0198] Figure 3 As shown, in the second energy transaction involving microgrid 5 in the 11th hour, microgrids 1, 4, and 5 participated in the energy transaction. The energy transaction volume of microgrids 4 and 5 was significantly greater than that of microgrid 1. This is because, based on conventional... The reputation model's energy trading methodology recognizes the compliant energy trading of Microgrid 5 in the 10th hour, allowing for a larger energy trading volume for Microgrid 5; Figure 4 As shown, the energy trading volume of microgrids 4 and 5 in the 11th hour was similar to that of microgrid 1. This is because the multi-microgrid P2P energy trading method of the improved reputation management model of this invention retains the memory of microgrid 5's past malicious behavior and did not allow microgrid 5 to have a large energy trading volume due to a compliant energy transaction in the 10th hour. However, compared to... Figure 4 The energy trading volumes of Microgrid 5 in the 10th and 11th hours show that after the compliant energy trading in the 10th hour, Microgrid 5 was allowed a slightly larger energy trading volume in the 11th hour. This means that the multi-microgrid P2P energy trading method of the improved reputation management model of this invention retains the memory of the historical malicious behavior of Microgrid 5 while taking into account the positive impact of compliant energy trading of Microgrid 5.

[0199] Figure 3 As shown, in the six compliant energy transactions from hour 12 to hour 17, only Microgrid 4 and Microgrid 5 participated, compared to... Figure 4 , Figure 3 The energy trading volume of both Microgrid 4 and Microgrid 5 was significantly higher than that of the same period last year. Figure 4 This is because it is based on conventional The reputation model's energy trading method continuously recognizes the compliant energy trading of Microgrid 5 in the 10th hour, while the energy trading method of this invention keeps a memory of Microgrid 5's historical malicious behavior and takes into account the positive impact of Microgrid 5's compliant energy trading.

[0200] Figure 3 As shown, in the 9th energy transaction involving microgrid 5 in the 18th hour, microgrids 1 through 5 all participated in energy trading. Microgrid 5 had the largest energy trading volume, which was significantly greater than the other four microgrids. Figure 3 The example shown is based on conventional In the reputation model-based energy trading method, Microgrid 5 successfully cleaned its name, while in contrast... Figure 4 As shown, the energy trading volume of microgrid 5 in the 18th hour was also the largest, but not significantly greater than the other four microgrids. This is because the energy trading method of the present invention retains a memory of the historical malicious behavior of microgrid 5 while also taking into account the positive impact of compliant energy trading of microgrid 5.

[0201] Depend on Figure 3 Analysis shows that Microgrid 5, relying on compliant transactions from hour 10 to hour 17, presented a trustworthy facade in the energy trading market. Other microgrids believed it was safe to trade with this malicious microgrid, and therefore signed large, high-value energy contracts with it in hour 18, indirectly reflecting that its reputation was successfully whitewashed. From Figure 3The energy trading volume clearly shows that malicious microgrid 5 successfully participated in multiple energy transactions between the 10th and 18th hours, and by the 18th hour, microgrid 5's energy trading volume dominated the market. This indicates that based on... Traditional energy trading methods based on reputation models fail to identify such attacks, leading to malicious microgrids being wrongly given excessive trading opportunities and increasing the overall risk in the energy trading market. Figure 4 Analysis shows that the multi-microgrid P2P energy trading method of this invention, which improves reputation management models, significantly reduces the energy trading share of malicious microgrid 5 when faced with the same multiple compliant energy trading operations. This indirectly reflects that microgrid 5's reputation score only slightly increases, almost maintaining the penalty for early negative records. Furthermore, other microgrids recognize the high risk of microgrid 5 and actively choose to abandon transactions with this malicious microgrid, resulting in a decrease in the total market trading volume compared to the previous method. Figure 3 There has been a decline. Meanwhile, even if micronet 4 is not a malicious micronet, its excessive transactions with malicious micronet 5 pose a risk of association. Figure 4 The energy trading data for the 18th hour shows that Microgrid 4 also had its energy trading share reduced. Therefore, according to... Figure 3 and Figure 4 Analysis shows that the multi-microgrid P2P energy trading method of the improved reputation management model of this invention can effectively suppress malicious behavior that uses low-risk transactions to quickly improve reputation scores.

[0202] In this specific application scenario, a standard IEEE 33-node peer-to-peer (P2P) energy trading network is adopted. Each node is connected to a distributed energy provider (DER), and a certain degree of power prediction error exists in actual operation. Transactions between nodes are limited to direct energy purchase and sale contracts, and participants remain anonymous, simulating the complexity and trust challenges of a real decentralized market. The simulation period is 28 days, and the system collects the fulfillment rate and reputation score of each node daily for the previous 24 hours. To differentiate between good and bad behavior, the fulfillment rate is divided into three levels: This indicates a good contract fulfillment rate. This indicates that the compliance rate is tolerable. This indicates that the microgrid's compliance rate has deteriorated to the point where it is prohibited from participating in energy trading; subsequent failure scenarios will only affect the compliance rate. The micro-network unfolds.

[0203] The simulation uses the following five metrics to measure system performance: average reputation score (arithmetic mean of daily reputation scores over 28 days); successful transaction volume over 28 days (total energy transactions successfully completed by all nodes over 28 days); successful transaction rate (proportion of successful transactions to total transactions); and participant retention rate (proportion of active microgrids at the end of the simulation to the initial number of microgrids). Active microgrids refer to the fulfillment rate. The micro-network, when the fulfillment rate At that time, the microgrid was banned from participating in energy trading and became an inactive microgrid.

[0204] The simulation uses the following four scenarios: reputation-free model, Reputation model Reputation model + time forgetting factor, the improved reputation management model of this invention.

[0205] Table 1 Comparison of Indicators under the Standard 33-Node System

[0206] ;

[0207] Table 1 shows that, under the reputation-free model, the successful transaction volume was 482.56 MWh, the successful transaction rate was 84.43%, and the participant retention rate was 67.86%. (Introduction) Following the implementation of the reputation model, successful trading volume increased to 530.20 MWh, the success rate improved to 93.96%, and participant retention rate increased to 93.21%. This demonstrates that the introduction of the reputation management model is crucial for improving the efficiency and security of the energy trading market. The reputation management model incorporates a fixed value as a fixed-time forgetting factor to weight historical and current behaviors, taking into account the time factor and balancing historical and current behavior. The average reputation score increased from 0.81 to 0.86, and the successful transaction rate increased to 94.16%, but the participant retention rate slightly decreased to 92.86%, indicating that reasonable time decay can refine the reputation score allocation. The improved reputation management model of this invention improves the dynamic reputation model and multi-dimensional forgetting factor, not only considering time but also dynamically adjusting the forgetting rate based on multiple dimensions such as reputation fluctuations and event severity. The simulation results are optimal: successful transaction volume of 538.52 MWh, a successful transaction rate of 96.24%, and a participant retention rate of 98.93%. This demonstrates that the multi-microgrid P2P energy trading method of this invention, with its improved reputation management model, can effectively optimize the multi-microgrid P2P energy trading market.

[0208] Table 2. Dynamic Comparison of Reputation Scores under Different Failure Scenarios Using the Improved Reputation Management Model of This Invention

[0209] Fault Scenario Average reputation score Minimum reputation score Starting reputation score Reputation Score Recovery decline Normal situation 0.92 0.90 0.92 / / Intermittent failure 0.90 0.85 0.92 ≈3 days 0.07 Persistent failure 0.86 0.78 0.92 / 0.14

[0210] As shown in Table 2, under normal circumstances, the microgrid maintains a high fulfillment rate, with an average reputation score of 0.92 over 28 days, and a minimum of no less than 0.90. The reputation score curve only oscillates slightly. During occasional failures, the fulfillment rate drops rapidly, and the reputation score falls from 0.92 to a minimum of 0.85, before quickly rebounding to its original stable level within approximately 3 days. During persistent failures, the microgrid maintains a low fulfillment rate from the onset, with the reputation score continuously declining to a minimum of 0.78, and an average reputation score of only 0.86 over 28 days, showing no significant recovery. This comparison verifies the ability of the multi-microgrid P2P energy trading method of this invention's improved reputation management model to smoothly punish and quickly recover from short-term fluctuations, as well as its sustained effectiveness in combating long-term misconduct.

[0211] Table 3 Comparison of Reputation Whitewashing Attack Resistance

[0212] ;

[0213] To test whitewashing attacks, 100 successful small-scale energy sales were simulated in a case with an initial reputation score of 0.50: based on Traditional energy trading methods based on reputation models can easily boost a reputation score from an initial 0.50 to 0.92 through small transactions, an increase of 0.42. The improved reputation management model of this invention, using a multi-microgrid P2P energy trading method, only slightly increases the reputation score from 0.50 to 0.53 for the same small transactions, almost maintaining the penalty for early negative records. The comparison results in Table 3 demonstrate that the improved reputation management model of this invention effectively suppresses malicious behavior that uses low-risk transactions to quickly boost reputation scores, effectively resisting reputation whitewashing attacks.

[0214] The above three comparisons demonstrate that the multi-microgrid P2P energy trading method of the present invention, which improves the reputation management model, not only increases the successful transaction volume and success rate of the multi-microgrid P2P energy trading market and improves user retention rate, but also exhibits differentiated response capabilities to occasional and continuous failures, and strong resistance to reputation whitewashing attacks. This fully proves its practical value and robustness in the real decentralized energy market.

[0215] Example 2:

[0216] Based on the same inventive concept as Embodiment 1, this embodiment introduces an improved reputation management model for a multi-microgrid P2P energy trading system, including: a reputation calculation module, a weight generation module, and a strategy generation module;

[0217] The reputation calculation module is used to: calculate the reputation score of each microgrid in real time based on the P2P energy transaction history of the multi-microgrid using an improved reputation management model; the improvement of the reputation management model includes: introducing a dynamic forgetting factor calculated based on the change in reputation score, and adjusting the weight of historical information and real-time information in the reputation score calculation to improve the timeliness of reputation management.

[0218] The weight generation module is used to: calculate the priority score of each micro-network based on the reputation score, and calculate the coupling weight of the corresponding transaction based on the priority scores of the micro-networks of both parties to the transaction;

[0219] The strategy generation module is used to: weight the distributed optimization algorithm through the coupling weights, adjust the priority of each microgrid transaction, and generate a multi-microgrid P2P energy trading strategy.

[0220] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0221] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0222] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0223] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0224] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A multi-microgrid P2P energy trading method with an improved reputation management model, characterized in that, include: Based on the P2P energy trading history of multiple microgrids, an improved reputation management model is used to calculate the reputation score of each microgrid in real time. The calculation of the reputation score of each microgrid includes: based on the pre-set dynamic forgetting factor calculated according to the change in reputation score and the P2P energy trading history of the multi-microgrid, the historical positive and negative feedback accumulation and the real-time global positive and negative feedback increment are weighted and fused respectively to calculate the real-time positive and negative feedback accumulation of each microgrid, and the reputation score of each microgrid is updated according to the real-time positive and negative feedback accumulation. The priority score of each micronetwork is calculated based on the reputation score, and the coupling weight of the corresponding transaction is calculated through the priority scores of the micronetworks of both parties to the transaction. By weighting the distributed optimization algorithm with the coupling weights, the priority of each microgrid transaction is adjusted, and a multi-microgrid P2P energy trading strategy is generated.

2. The multi-microgrid P2P energy trading method with an improved reputation management model according to claim 1, characterized in that, The method involves using a pre-set dynamic forgetting factor calculated based on reputation score changes and the multi-microgrid P2P energy trading history to weight and fuse historical positive and negative feedback accumulations and real-time global positive and negative feedback increments, respectively, to calculate the real-time positive and negative feedback accumulation of each microgrid. The reputation score of each microgrid is then updated based on the real-time positive and negative feedback accumulation, including: Based on the P2P energy trading history of multi-microgrid, a dynamic forgetting factor calculated based on the change in reputation score is introduced in the Beta reputation model reputation score calculation. The cumulative positive and negative feedback at the previous moment and the global positive and negative feedback increment at the current moment are weighted and fused to calculate the cumulative positive and negative feedback of each microgrid at the current moment. Micronet i Current Moment Positive feedback accumulation The calculation formula is: ， In the formula, Indicates the current moment of microgrid i The dynamic forgetting factor; Indicates the previous moment of the microgrid i The cumulative amount of positive feedback; Indicates the current moment of microgrid i The global positive feedback increment; Micronet i Current Moment negative feedback accumulation The calculation formula is: ， In the formula, Indicates the previous moment of the microgrid i The cumulative amount of negative feedback; Indicates the current moment of microgrid i The global negative feedback increment; Update the reputation score of each micronetwork for the next time step based on the current cumulative positive and negative feedback. Micronet iNext Moment reputation score The calculation formula is: ， In the formula, Indicates the current moment of microgrid i The cumulative amount of positive feedback; Indicates the current moment of microgrid i The cumulative amount of negative feedback; and This represents the preset smoothing coefficient.

3. The multi-microgrid P2P energy trading method with an improved reputation management model according to claim 2, characterized in that, The dynamic forgetting factor is calculated based on changes in reputation score, including: Based on the P2P energy trading history of multiple microgrids, calculate the short-term reputation fluctuations and long-term reputation trends of each microgrid. Micronet i Current Moment Short-term reputation fluctuations The calculation formula is: ， In the formula, Indicates the current moment of microgrid i Reputation score; Indicates the previous moment of the microgrid i Reputation score; Micronet i Current Moment Long-term reputation trend The calculation formula is: ， In the formula, Indicates the first seven moments of the microgrid i Reputation score; The dynamic forgetting factor for each micronet at the current moment is calculated based on short-term reputation fluctuations and long-term reputation trends; micronet i at the current moment... Dynamic forgetting factor The calculation formula is: ， ， In the formula, This represents the preset balance factor that adjusts short-term weights; This represents the preset balance factor that adjusts the long-term weights; This indicates the preset sensitivity to short-term fluctuations. ; This indicates the preset sensitivity to long-term fluctuations. .

4. The multi-microgrid P2P energy trading method with an improved reputation management model according to claim 1, characterized in that, The calculation of the reputation score for each micronetwork also includes: The system employs a comprehensive transaction weight calculated based on transaction size and risk. It then uses an asymmetric positive and negative feedback increment calculation method to calculate the positive and negative feedback increments separately. Finally, it calculates the global positive and negative feedback increments for each microgrid based on these positive and negative feedback increments.

5. The multi-microgrid P2P energy trading method with an improved reputation management model according to claim 4, characterized in that, The overall weight of a transaction is calculated based on its size and risk, including: Based on the historical P2P energy transactions of Duowei.com, the normalized weight of the corresponding transaction size is calculated to measure the transaction size; at the current moment... microgrid With the peer microgrid Normalized weights of transaction size The calculation formula is: ， ， In the formula, Indicates the current time microgrid peer-to-peer microgrid The promised delivery volume; Indicates to microgrid A collection of micro-networks for transactions; Represents any time from the initial time to time t; microgrid Largest transaction commitment in history; Based on the peer microgrid's reputation score, calculate the corresponding microgrid reputation risk weight for the transaction to measure transaction risk; at the current moment... microgrid With the peer microgrid Micro-network reputation risk weighting of transactions The calculation formula is: ， In the formula, This represents the preset risk amplification factor; Indicates to microgrid Micronet on the other end of the transaction Current moment The overall risk level of default; Indicates the peer microgrid Current moment Reputation score; The overall transaction weight is calculated based on the normalized weight of the transaction commitment amount and the microgrid reputation risk weight based on the peer microgrid reputation score; at the current moment... microgrid With the peer microgrid Overall weight of transactions The calculation formula is: 。 6. The multi-microgrid P2P energy trading method with an improved reputation management model according to claim 5, characterized in that, The asymmetric positive and negative feedback increment calculation method includes: Current moment Positive feedback increment of transactions between microgrid i and its counterpart microgrid j The calculation formula is: ， In the formula, This indicates the preset positive feedback index; Indicates the current time micronet peer-to-peer microgrid The fulfillment rate; the current time micronet peer-to-peer microgrid fulfillment rate The calculation formula is: ， In the formula, Indicates the current time microgrid peer-to-peer microgrid Committed delivery volume Indicates the current time microgrid peer-to-peer microgrid The actual delivery volume; Current moment Negative feedback increment of transactions between microgrid i and its counterpart microgrid j The calculation formula is: ， In the formula, This indicates the preset negative feedback index.

7. The multi-microgrid P2P energy trading method with an improved reputation management model according to claim 6, characterized in that, The calculation of the global positive and negative feedback increments for each microgrid includes: The positive and negative feedback increments of each microgrid at the current moment in the global transaction are summarized respectively, and the global positive and negative feedback increments of each microgrid at the current moment are obtained accordingly. Micronet i Current Moment Global positive feedback increment The calculation formula is: , Micronet i Current Moment Global negative feedback increment The calculation formula is: .

8. The multi-microgrid P2P energy trading method with an improved reputation management model according to claim 1, characterized in that, The step of calculating the priority score of each micronetwork based on the reputation score, and calculating the coupling weight of the corresponding transaction through the priority scores of the micronetworks of both parties to the transaction, includes: Calculate the priority score for each micronet at the corresponding time based on its reputation score. Micronet i at the current time... Priority score The calculation formula is: ， In the formula, This represents the preset reputation amplification factor; The coupling weight of the corresponding transaction is obtained by multiplying the priority scores of the two micronetworks involved in the transaction. Micronetwork i and the peer micronetwork j are the two parties in the transaction, and the coupling weight of the transaction between micronetwork i and the peer micronetwork j is calculated accordingly. The calculation formula is: ， In the formula, This indicates that the peer micronetwork j is at the current time. Priority score.

9. The multi-microgrid P2P energy trading method according to claim 8, characterized in that, The step of weighting the distributed optimization algorithm using the coupling weights to adjust the priority of transactions in each microgrid and generating a multi-microgrid P2P energy trading strategy includes: The distributed optimization algorithm is the C-ADMM algorithm; the consensus penalty term of the C-ADMM algorithm is weighted by the coupling weight of the corresponding transaction, wherein the coupling weight is positively correlated with the priority of the corresponding microgrid transaction; by adjusting the priority of each microgrid transaction, a multi-microgrid P2P energy trading strategy is generated. Local variables of microgrid i in C-ADMM algorithm In the iterative formula, the consensus penalty term after weighting the coupling weights is: ， In the formula, This represents the original consensus penalty term of the C-ADMM algorithm; Indicates the penalty factor; Let represent the common variables in the transaction between microgrid i and its counterpart microgrid j during the k-th iteration; Let represent the dual variable of the transaction between microgrid i and its counterpart microgrid j in the k-th iteration.

10. A multi-microgrid P2P energy trading system with an improved reputation management model, characterized in that, include: Reputation calculation module, weight generation module, and strategy generation module; The reputation calculation module is used to: calculate the reputation score of each microgrid in real time based on the P2P energy trading history of the multi-microgrid using an improved reputation management model; Improvements to the reputation management model include: introducing a dynamic forgetting factor calculated based on changes in reputation score, and adjusting the weights of historical and real-time information in reputation score calculation to improve the timeliness of reputation management; The weight generation module is used to: calculate the priority score of each micro-network based on the reputation score, and calculate the coupling weight of the corresponding transaction based on the priority scores of the micro-networks of both parties to the transaction; The strategy generation module is used to: weight the distributed optimization algorithm through the coupling weights, adjust the priority of each microgrid transaction, and generate a multi-microgrid P2P energy trading strategy.

Images