Trusted agent mechanism in virtual power plant optimization regulation and control and reputation evaluation method thereof

Abstract

The invention discloses a trusted agent mechanism in virtual power plant optimization regulation and control and a reputation evaluation method thereof, and the method comprises the steps: an agent node submits necessary identity information and capability proof through an intelligent contract, and completes the registration of the agent node; according to the capability value and the reputation value of the agent node, the authority value of the agent node is automatically calculated through the smart contract; the virtual power plant issues task requirements and distributes tasks to the agent nodes according to the authority values of the agent nodes; the smart contract monitors the behavior of the agent node in real time, and records a monitoring result on the block chain; updating the reputation value of the agent node, and storing the updated reputation value on the block chain; and dynamically adjusting the cooperation strategy of the agent node, and implementing a reward and punishment mechanism according to the updated reputation value. Based on a DAO proxy mechanism, autonomous operation and mutual supervision of proxy nodes are realized, and transparency and credibility of a virtual power plant system are improved.

Description

Technical Field

[0001] The present invention relates to the field of energy management technology, and in particular to a trusted agent mechanism and a reputation evaluation method thereof in virtual power plant optimization and regulation. Background Art

[0002] As an important form of distributed energy management, the optimization and regulation of virtual power plants are of great significance to the efficient utilization of energy and system stability. By integrating multiple resources such as distributed power sources, energy storage equipment, and controllable loads, virtual power plants can achieve flexible scheduling and optimal configuration of energy, thereby improving energy utilization efficiency and promoting the stable operation of the power system. However, the diversity and complexity of proxy nodes in the virtual power plant environment pose challenges to the trust management of the system. These proxy nodes may include distributed power sources (such as photovoltaic power generation, wind power generation, etc.), energy storage equipment, controllable loads, electric vehicles, and microgrids, each of which has different capabilities, states, and interests.

[0003] At present, research on virtual power plants at home and abroad has made some progress, but there are still deficiencies in the trust evaluation, behavior monitoring and cooperation strategies of agent nodes. Traditional trust management methods often rely on static evaluation indicators and fixed cooperation strategies, which are difficult to adapt to the dynamic characteristics of agent nodes in virtual power plants and the complex and changing energy market environment. This leads to the virtual power plant system being unable to make the best scheduling decisions when facing various uncertainties and risks, thus affecting the overall performance and efficiency of the system. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies of the above-mentioned prior art and to provide a trusted agent mechanism and a reputation evaluation method for the optimization and regulation of a virtual power plant, so as to solve the problems of virtual power plant system performance and efficiency caused by insufficient trust management of agent nodes and poor cooperation strategies in virtual power plants.

[0005] To achieve the above object, the technical solution of the present invention is:

[0006] A trusted agent mechanism and reputation evaluation method in virtual power plant optimization and control, including:

[0007] The agent node submits the necessary identity information and capability proof through the smart contract to complete the registration of the agent node;

[0008] According to the ability and reputation value of the proxy node, its authority value is automatically calculated through the smart contract;

[0009] The virtual power plant publishes task requirements and assigns tasks to proxy nodes according to the authority values ​​of the proxy nodes;

[0010] Smart contracts monitor the behavior of proxy nodes in real time and record the monitoring results on the blockchain;

[0011] Update the reputation value of the proxy node and store the updated reputation value on the blockchain;

[0012] Dynamically adjust the cooperation strategy of proxy nodes and implement reward and punishment mechanisms according to the updated reputation values.

[0013] Optionally, the authority value of the proxy node is automatically calculated through a smart contract based on the capability value and reputation value of the proxy node, including:

[0014] Through the smart contract, the root calculates the capacity value C of each proxy node i i ;

[0015] Obtain the reputation R of each proxy node i based on the reputation system i ;

[0016] C-based i With R i , calculate the authority value P of each proxy node i i :

[0017] P i =αC i +βR i (1)

[0018] Where α and β are weight coefficients, satisfying α+β=1; the capacity of proxy node i is C i Calculate as follows:

[0019]

[0020] where c im Represents the ability of proxy node i on the mth resource, where M is the number of resource types

[0021] Optionally, the virtual power plant publishes task requirements and allocates tasks to proxy nodes according to authority values ​​of the proxy nodes; including:

[0022] Virtual power plants publish task requirements through smart contracts;

[0023] The proxy node submits task applications based on its own capabilities and current load;

[0024] Automatically evaluate the matching degree M between task j and agent node i through smart contracts ij

[0025] Calculate the task assignment probability P through smart contracts ij .

[0026] Optionally, the smart contract calculates the task allocation probability P by the following formula:ij for:

[0027]

[0028] Where E j is the urgency of task j, L i is the current load of proxy node i, and N is the total number of proxy nodes.

[0029] Optionally, the dynamically adjusting the cooperation strategy of the proxy node and implementing a reward and punishment mechanism according to the updated reputation value include:

[0030] The reputation value R of each node i Compared with the threshold θ, if R i ≥θ, then other nodes will continue to cooperate normally with node i; otherwise, reduce the task priority P of node i i , and limit the number of tasks it participates in T i ;

[0031] Calculate the reward distribution value r obtained by node i i .

[0032] Optionally, the reward distribution value r obtained by node i is calculated as follows: i :

[0033] Assume that the contribution of proxy node i is D i It is expressed as:

[0034] D i =γT i +δQ i +∈S i +ζP i (4)

[0035] Among them, T i is the number of tasks completed by agent node i, Q i is the quality of the task completed by agent node i, S i is the resource saving of proxy node i, γ, δ, ∈, ζ are weight coefficients, and γ+δ+∈+ζ=1; if the total reward is recorded as r, then the reward r allocated to proxy node i i It is expressed as:

[0036]

[0037] According to the contribution of the proxy node D i Based on the evaluation results, the smart contract automatically calculates the reward it deserves according to formula (5), and the reward distribution result is recorded on the blockchain.

[0038] Optionally, the reward and punishment mechanism is: for reputation R iImproved node i, incentive r i It has also been improved; and for reputation R i Reduce the node i, incentive r i A reduction occurs as a penalty.

[0039] Optionally, the reputation value of the updating proxy node is stored in the blockchain;

[0040] Each proxy node l calculates and provides feedback on the comprehensive reputation evaluation score of all other nodes i according to the set evaluation index system.

[0041] Calculate the final reputation value R of proxy node i i .

[0042] Optionally, the final reputation value R of proxy node i is calculated as follows: i :

[0043] Calculate the average score of agent node i And determine the maximum and minimum values ​​of the average scores of all proxy nodes, denoted as s max and min . The average score is then mapped to a range of reputation values ​​using linear interpolation:

[0044]

[0045] Optionally, the evaluation index system includes task completion status, response time, resource contribution, and violation records.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] The present invention is based on the proxy mechanism of DAO (Decentralized Autonomous Organization), realizes the autonomous operation and mutual supervision of proxy nodes, and improves the transparency and credibility of the virtual power plant system; by updating the reputation value of the proxy node, it realizes the accurate evaluation of the trust of the proxy node, and provides a basis for optimizing the cooperation strategy; through dynamic cooperation strategy adjustment and reputation reward and punishment mechanism, it improves the stability and efficiency of the virtual power plant system, and reduces the potential risks caused by low reputation nodes. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A DAO-based virtual power plant optimization and control system architecture diagram provided in an embodiment of the present application;

[0049] Figure 2 Functional module diagram for optimizing the control platform of virtual power plant;

[0050] Figure 3 Schematic diagram of agent nodes in virtual power plant optimization and control;

[0051] Figure 4 Flowchart of the implementation steps for the trusted agent mechanism and its reputation evaluation method in the optimization and control of virtual power plants. DETAILED DESCRIPTION

[0052] Example:

[0053] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0054] DAO is an autonomous organization built on the blockchain that exists and runs in the form of code. It uses smart contracts to execute organizational rules and decision-making processes without relying on traditional hierarchical management agencies or intermediaries, thus achieving the characteristics of decentralization and autonomy.

[0055] DAO has the characteristics of decentralization, transparency, and autonomy. DAO achieves decentralization of the organization through blockchain and smart contracts, allowing each member to directly participate in decision-making and management without going through a centralized management agency. All transactions and decision-making processes of DAO are recorded on the blockchain for anyone to review, thus ensuring the transparency of the organization. At the same time, DAO automatically executes decisions and rules through smart contracts without human intervention, thus achieving organizational autonomy.

[0056] The architecture of virtual power plant optimization control system based on DAO is as follows Figure 1 As shown in Figure 1, smart contracts are the core component of DAO, which stipulates the organization's rules, decision-making process, fund allocation, etc. Once deployed on the blockchain, smart contracts cannot be tampered with, thus ensuring the stability and reliability of the organization. The decision-making process of DAO is usually completed through voting. Members can submit proposals and decide whether to pass the proposal by voting. The voting results will be recorded on the blockchain and the corresponding decisions will be automatically executed.

[0057] As the core of the entire optimization and control system, the virtual power plant optimization and control platform is responsible for integrating and managing the functions of each module to achieve the overall optimization and control of the virtual power plant. It receives information from the agent node, processes it through each module, and finally outputs the optimal energy scheduling strategy. Figure 2 As shown in the figure, the virtual power plant optimization and control platform mainly includes an agent node registration module, an authority allocation module, a task allocation module, a behavior monitoring module, a reputation evaluation module and a dynamic cooperation strategy adjustment module.

[0058] The agent node registration module is responsible for the registration and identity authentication of new agent nodes. The agent node needs to submit necessary information to join the system. The information submitted by the agent node includes but is not limited to public keys, digital signatures, historical task completion records, resource management capability certificates, etc. This information is stored on the blockchain, and the authenticity of this information is verified through a consensus mechanism. Once verified, this information will be recorded on the blockchain to form an unalterable registration record, and the verified agent nodes will be included in the system management. In the optimization and control of virtual power plants, agent nodes refer to various distributed energy management entities or intelligent devices that participate in the operation of virtual power plant systems. In the case of Figure 3 In the virtual power plant system shown, the agent nodes involved in optimization and regulation include the following types:

[0059] Distributed power generation agent: Renewable energy generation units such as photovoltaic power generation and wind power generation, as energy suppliers, participate in the energy optimization and scheduling of the system.

[0060] Energy storage equipment agent: including battery energy storage, pumped storage, etc., as an energy storage unit, it can release or absorb energy when needed to balance the supply and demand and improve energy utilization efficiency.

[0061] Controllable load agents: such as industrial electricity, commercial electricity, etc., as energy consumers, accept the dispatch and control of the system to achieve flexible load adjustment and optimal allocation of energy.

[0062] Electric vehicle agent: As a mobile energy storage device, electric vehicles can also participate in energy optimization scheduling and charging management.

[0063] Virtual power plant agent: Receives electricity price information from the power market, combines it with the virtual power plant’s own power generation and load forecast information, summarizes the information and formulates an overall operating strategy.

[0064] The authority allocation module is used to assign corresponding permissions to the agent node according to its registration information and historical performance. These permissions determine the operation scope and participation level of the agent node in the virtual power plant system, ensuring the security and stability of the virtual power plant system. i , reputation is R i , then the authority of proxy node i can be expressed as:

[0065] P i =αC i +βR i (1)

[0066] Among them, α and β are weight coefficients, which are used to adjust the proportion of ability and reputation in authority allocation to satisfy α+β=1.

[0067] The capability C of proxy node ii It can be calculated as follows:

[0068]

[0069] where c im It represents the capability of agent node i on the mth resource (which can be calculated based on the efficiency and quality of the tasks completed by the agent node in the past), where M is the number of resource types.

[0070] The task allocation module is used to intelligently allocate tasks to appropriate proxy nodes based on the energy demand of the virtual power plant and the capacity status of the proxy nodes. It considers the priority and urgency of the task as well as the current load and available resources of the proxy nodes to achieve optimal task allocation.

[0071] Let the urgency of task j be E j , the current load of proxy node i is L i (including the number of assigned tasks and resource usage), the matching degree between task j and proxy node i is M ij (Based on task requirements and the authority P of proxy node i i The probability that task j is assigned to agent node i is P ij It can be expressed as:

[0072]

[0073] Where N is the total number of proxy nodes. This formula ensures that the task is assigned to the most suitable proxy node with the lowest current load, while taking into account the urgency of the task.

[0074] The proxy node in DAO calculates the matching degree between each proxy node and the task according to formula (3) and selects an optimal proxy node to perform the task. The task allocation results are recorded on the blockchain to ensure the transparency and traceability of task allocation.

[0075] The behavior monitoring module is used to monitor the behavior of the agent node in real time, including its task execution (such as completion time, completion quality), resource usage (such as resource consumption, resource efficiency) and interaction with other agent nodes. These behavior data are recorded on the blockchain to form an unalterable behavior record. Other nodes can evaluate the performance of the agent node by querying the behavior record on the blockchain. This module collects this information to provide data support for the reputation evaluation module.

[0076] The reputation evaluation module is used to evaluate the reputation of the proxy node based on the data provided by the behavior monitoring module. The reputation evaluation module uses a comprehensive reputation evaluation indicator system to calculate the reputation value of each proxy node and update it to the reputation database. The reputation value reflects the reliability and willingness to cooperate of the proxy node.

[0077] Specifically, the reputation evaluation index system covers multiple key dimensions of the proxy node, including task completion, response time, resource contribution, and violation records. The specific index definitions are as follows:

[0078] Table 1 Proxy node reputation evaluation index system

[0079]

[0080]

[0081]

[0082]

[0083] The above indicators can comprehensively reflect the behavioral characteristics and reputation status of the proxy nodes, and provide basic data for subsequent reputation calculations.

[0084] In order to achieve continuous monitoring and evaluation of the behavior of agent nodes, the behavior data of agent nodes can be collected in real time through sensors or loggers deployed in the virtual power plant system, including task execution records, resource usage, etc. Based on the collected data, a reputation calculation method based on recommendation and feedback is adopted here. This method relies on the recommendation and feedback of other nodes in the system to evaluate the reputation of a specific agent node. Its core idea is to use group wisdom to evaluate individual performance, thereby forming a relatively objective and comprehensive reputation evaluation system.

[0085] The virtual power plant system encourages other nodes to evaluate the performance of the proxy node. According to the evaluation index system in Table 1, a comprehensive reputation score of the node is obtained. Assume that there is a proxy node i in the system, which receives recommendations and feedback from N other nodes. The comprehensive reputation score of each feedback is divided into (l=1, 2, ..., N). First, calculate the average score of proxy node i. And determine the maximum and minimum values ​​of the average scores of all proxy nodes, denoted as s max and min . Then use linear interpolation to map the average score to a reputation value range (assuming the reputation value range is 0-100):

[0086]

[0087] If the average score is already in numerical form and within a known range, you can directly use the above formula to convert it. If the average score is in graded form, you need to convert it to numerical form before calculating it.

[0088] It is worth noting that it is necessary to ensure that the collected feedback information is authentic and reliable. Identity verification and other means can be used to prevent false feedback. In addition, the reputation value of the proxy node is updated regularly to reflect its latest performance. The frequency and conditions for reputation value updates can be set, such as updating once a certain number of new feedbacks are received.

[0089] The dynamic cooperation strategy adjustment module is used to dynamically adjust the cooperation strategy between proxy nodes according to the results of the reputation evaluation module. It prioritizes cooperation with high-reputation nodes and limits cooperation with low-reputation nodes to improve the overall performance and efficiency of the system. At the same time, the module can also adjust the cooperation strategy in real time to adapt to the new environment according to the changes in the system status.

[0090] Specifically, a threshold θ can be set. When assigning tasks, only the credit R i ≥θ, the nodes i will cooperate. i <θ, we can set restrictions or isolation measures, such as reducing its task priority P i , or limit the number of tasks it participates in T i .

[0091] Theoretically, a node with a high reputation usually means that its past behavior was more reliable and efficient, so cooperation with it is more likely to ensure the successful completion of the task. However, a node with a low reputation may be marked by the system due to multiple dishonest behaviors or insufficient capabilities, and cooperation with it may bring potential risks and losses. Therefore, this dynamic adjustment strategy based on reputation value can improve the quality and efficiency of task completion, avoid task failures caused by node dishonesty or insufficient capabilities, reduce the overall performance degradation of the system caused by low-reputation nodes, and improve the stability and security of the system.

[0092] In order to maintain the trust ecology of the virtual power plant system, this application also proposes a reputation reward and punishment mechanism, including a reward mechanism and a punishment mechanism, to motivate and constrain the behavior of proxy nodes and improve the overall efficiency of the system. These rewards can be digital currency, points or other forms of resources, and the specific form can be flexibly selected and adjusted according to actual conditions.

[0093] Assume that the contribution of proxy node i is D i It can be expressed as:

[0094] D i =γT i +δQ i +∈S i +ζP i (4)

[0095] Among them, T i is the number of tasks completed by agent node i, Q iis the quality of the task completed by agent node i, S i is the resource saving of proxy node i, γ, δ, ∈, ζ are weight coefficients, and γ+δ+∈+ζ=1. If the total reward is recorded as r, then the reward r that proxy node i can be allocated i It can be expressed as:

[0096]

[0097] According to the contribution of the proxy node D i Based on the evaluation results, the smart contract automatically calculates the reward it deserves according to formula (5), and the reward distribution results are recorded on the blockchain to ensure the transparency and fairness of the reward distribution.

[0098] The reward distribution adjustment is achieved through the incentive mechanism (i.e., formulas (4) and (5)). i Improved node i, incentive r i The reputation R i Reduce the node i, incentive r i A reduction occurs as a penalty.

[0099] Rewarding high-reputation nodes can motivate them to continue to maintain honest behavior and set an example for other nodes. Punishing or isolating low-reputation nodes can curb their malicious behavior and prevent them from continuing to damage the system. This reputation reward and punishment mechanism can enhance the enthusiasm of high-reputation nodes, promote the improvement of the overall performance of the system, maintain the fairness and stability of the system, and reduce potential risks caused by low-reputation nodes.

[0100] See also Figure 4 As shown, the trusted agent mechanism and reputation evaluation method in the virtual power plant optimization control provided by this embodiment mainly include the following steps:

[0101] S1. The agent node submits the necessary identity information and capability proof through the smart contract to complete the registration of the agent node.

[0102] Specifically, the proxy node submits the necessary identity information and capability proof through the smart contract. The identity information and capability proof include but are not limited to public key, digital signature, historical task completion record, resource management capability proof, etc. This information is stored on the blockchain and its authenticity is verified through the consensus mechanism. Once verified, this information will be recorded on the blockchain to form an unalterable registration record.

[0103] S2. According to the ability and reputation value of the proxy node, its authority value is automatically calculated through the smart contract;

[0104] In this step, the capability value and reputation value can be evaluated through historical data on the blockchain and the interaction between nodes. According to the registration information and historical performance of the agent node, it is assigned corresponding permissions. These permissions determine the operating scope and degree of participation of the agent node in the system, ensuring the security and stability of the virtual power plant system.

[0105] S3. The virtual power plant publishes task requirements and allocates tasks to agent nodes according to the authority values ​​of the agent nodes, so that the optimal agent node can be selected to perform the task based on the urgency of the task, the current load of the agent node, and the matching degree between the task and the ability of the agent node.

[0106] S4. Smart contracts monitor the behavior of proxy nodes in real time and record the monitoring results on the blockchain.

[0107] Specifically, smart contracts can be used to monitor the behavior of proxy nodes in real time, including task completion (such as completion time and completion quality), resource usage (such as resource consumption and resource efficiency), etc. These behavior data are recorded on the blockchain to form an immutable behavior record. Other nodes can evaluate the performance of proxy nodes by querying the behavior records on the blockchain, and can also be used in subsequent incentive mechanisms to provide data support for reputation evaluation.

[0108] S5. Update the reputation value of the proxy node and store the updated reputation value on the blockchain;

[0109] S6. Dynamically adjust the cooperation strategy of the proxy nodes and implement the reward and punishment mechanism according to the updated reputation value.

[0110] It can be seen that this method is based on the proxy mechanism of DAO (Decentralized Autonomous Organization), which realizes the autonomous operation and mutual supervision of proxy nodes, and improves the transparency and credibility of the system; by updating the reputation value of the proxy node, it realizes the accurate evaluation of the trust of the proxy node, and provides a basis for optimizing the cooperation strategy; through dynamic cooperation strategy adjustment and reputation reward and punishment mechanism, the stability and efficiency of the virtual power plant system are improved, and the potential risks caused by low reputation nodes are reduced.

[0111] In a specific embodiment, the step S2 includes:

[0112] S201. Calculate the capacity value C of each proxy node i through the smart contract i ;

[0113] S202: Obtain the reputation R of each proxy node i based on the reputation system i ;

[0114] S203, based on Ci With R i , calculate the authority value P of each proxy node i i :

[0115] P i =αC i +βR i (1)

[0116] Where α and β are weight coefficients, satisfying α+β=1; the capacity of proxy node i is C i Calculate as follows:

[0117]

[0118] where c im Represents the ability of proxy node i on the mth resource, where M is the number of resource types

[0119] In a specific embodiment, step S3 includes:

[0120] S301: The virtual power plant publishes task requirements through smart contracts, including resource type, quantity, urgency, etc.

[0121] S302: The proxy node submits a task application through a smart contract based on its own capabilities and current load;

[0122] S303: Automatically evaluate the matching degree M between task j and proxy node i through smart contract ij ;

[0123] S304: Calculate the task allocation probability P according to formula (3) through the smart contract ij :

[0124] Let the urgency of task j be E j , the current load of proxy node i is L i (including the number of assigned tasks and resource usage), the matching degree between task j and proxy node i is M ij (Based on task requirements and the authority P of proxy node i i Comparison shows that the probability P of task j being assigned to agent node i is ij It can be expressed as:

[0125]

[0126] Where N is the total number of proxy nodes. This formula ensures that the task is assigned to the most suitable proxy node with the lowest current load, while taking into account the urgency of the task.

[0127] The nodes in the DAO calculate the matching degree between each proxy node and the task according to formula (3) and select an optimal proxy node to perform the task. The task allocation results are recorded on the blockchain to ensure the transparency and traceability of task allocation.

[0128] In a specific embodiment, step S5 includes the following steps:

[0129] S501: Each proxy node l calculates and provides feedback on the comprehensive reputation evaluation score of all other nodes i according to the evaluation index system in Table 1

[0130] S502: Calculate the final reputation value R of proxy node i according to formula (6) i .

[0131] Calculate the average score of agent node i And determine the maximum and minimum values ​​of the average scores of all proxy nodes, denoted as s max and min . The average score is then mapped to a range of reputation values ​​using linear interpolation:

[0132]

[0133] In a specific embodiment, step S6 includes the following steps:

[0134] S601: The reputation value R of each node i Compare with the threshold θ and adjust the corresponding cooperation strategy. i ≥θ, then other nodes will continue to cooperate normally with node i; otherwise, reduce the task priority P of node i i , and limit the number of tasks it participates in T i ;

[0135] S602: Calculate the reward distribution value r obtained by node i according to formula (4) (5) i .

[0136] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable ordinary technicians in the field to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made based on the essence of the content of the present invention should be included in the protection scope of the present invention.

Claims

1. A trusted agent mechanism and reputation evaluation method in virtual power plant optimization and control, characterized in that: include: The agent node submits the necessary identity information and capability proof through the smart contract to complete the registration of the agent node; According to the ability and reputation value of the proxy node, its authority value is automatically calculated through the smart contract; The virtual power plant publishes task requirements and assigns tasks to proxy nodes according to the authority values ​​of the proxy nodes; Smart contracts monitor the behavior of proxy nodes in real time and record the monitoring results on the blockchain; Update the reputation value of the proxy node and store the updated reputation value on the blockchain; Dynamically adjust the cooperation strategy of proxy nodes and implement reward and punishment mechanisms according to the updated reputation values.

2. The trusted agent mechanism and reputation evaluation method in the virtual power plant optimization and control as claimed in claim 1, characterized in that: According to the capability value and reputation value of the proxy node, the authority value is automatically calculated through the smart contract, including: Through smart contracts, the capacity value C of each proxy node i is calculated i ; Obtain the reputation R of each proxy node i based on the reputation system i ; C-based i With R i , calculate the authority value P of each proxy node i i : P i =αC i +βR i (1) Where α and β are weight coefficients, satisfying α+β=1; the capacity of proxy node i is C i Calculate as follows: where c im It represents the capability of proxy node i on the mth resource, where M is the number of resource types.

3. The trusted agent mechanism and reputation evaluation method in the virtual power plant optimization and control as claimed in claim 1 is characterized in that: The virtual power plant publishes task requirements and allocates tasks to proxy nodes according to the authority values ​​of the proxy nodes; including: Virtual power plants publish task requirements through smart contracts; The proxy node submits task applications based on its own capabilities and current load; Automatically evaluate the matching degree M between task j and agent node i through smart contracts ij Calculate the task assignment probability P through smart contracts ij .

4. The trusted agent mechanism and reputation evaluation method in the virtual power plant optimization and control as claimed in claim 3 is characterized in that: The smart contract calculates the task allocation probability P by the following formula: ij for: Where E j is the urgency of task j, L i is the current load of proxy node i, and N is the total number of proxy nodes.

5. The trusted agent mechanism and reputation evaluation method in the virtual power plant optimization and control as claimed in claim 1, characterized in that: The dynamic adjustment of the cooperation strategy of the proxy node and the implementation of the reward and punishment mechanism according to the updated reputation value include: The reputation value R of each node i Compared with the threshold θ, if R i ≥θ, then other nodes will continue to cooperate normally with node i; otherwise, reduce the task priority P of node i i , and limit the number of tasks it participates in T i ; Calculate the reward distribution value r obtained by node i i .

6. The trusted agent mechanism and reputation evaluation method in the virtual power plant optimization and control as claimed in claim 5, characterized in that: The reward distribution value r obtained by node i is calculated as follows i : Assume that the contribution of proxy node i is D i It is expressed as: D i =γT i +δQ i +∈S i +ζP i (4) Among them, T i is the number of tasks completed by agent node i, Q i is the quality of the task completed by agent node i, S i is the resource saving of proxy node i, γ, δ, ∈, ζ are weight coefficients, and γ+δ+∈+ζ=1; if the total reward is recorded as r, then the reward r allocated to proxy node i i It is expressed as: According to the contribution of the proxy node D i Based on the evaluation results, the smart contract automatically calculates the reward it deserves according to formula (5), and the reward distribution result is recorded on the blockchain.

7. The trusted agent mechanism and reputation evaluation method in the virtual power plant optimization and control as claimed in claim 6 is characterized in that: The reward and punishment mechanism is as follows: For reputation i Improved node i, incentive r i Also improve; for reputation R i Reduce the node i, incentive r i A reduction occurs as a penalty.

8. The trusted agent mechanism and reputation evaluation method in the virtual power plant optimization and control as claimed in claim 1, characterized in that: The reputation value of the updating agent node is stored in the blockchain; Each proxy node l calculates and provides feedback on the comprehensive reputation evaluation score of all other nodes i according to the set evaluation index system. Calculate the final reputation value R of proxy node i i .

9. The trusted agent mechanism and reputation evaluation method in the virtual power plant optimization and control as claimed in claim 8, characterized in that: The final reputation value R of proxy node i is calculated as follows i : Calculate the average score of agent node i And determine the maximum and minimum values ​​of the average scores of all proxy nodes, denoted as s max and min ; The average score is then mapped to a range of reputation values ​​using linear interpolation:

10. The trusted agent mechanism and reputation evaluation method in virtual power plant optimization and control as claimed in claim 8, characterized in that: The evaluation index system includes task completion, response time, resource contribution, and violation records.

Images