Virtual power plant multi-agent collaborative power dispatching method, device, medium and product

Abstract

This application relates to the field of power system automation control technology, and discloses a method, equipment, medium, and product for multi-entity collaborative power dispatching in a virtual power plant. The method includes: determining the response deviation characteristics of each participating entity based on their decision-making autonomy information and real-time operating status; conducting a dispatching risk assessment based on the response deviation characteristics and the current system operating status to obtain the assessment result; generating a draft coordinated dispatching scheme based on the assessment result and the privacy protection rules of each participating entity; the privacy protection rules characterize the mapping relationship between the data dimensions that each participating entity is allowed to disclose and the anonymization accuracy; and determining the target multi-entity coordinated dispatching instruction based on the feedback responses of each participating entity to the draft coordinated dispatching scheme. This method can at least solve the technical problems in related technologies, such as the difficulty in quantifying the differences in interests and information asymmetry among multiple entities, and the lack of a dynamic balance mechanism between privacy protection and dispatching accuracy.

Description

Technical Field

[0001] This application relates to the field of power system automation control technology, and in particular to a method, equipment, medium and product for multi-entity collaborative power dispatching of a virtual power plant. Background Technology

[0002] Virtual power plants, as a new form of power system organization, integrate distributed renewable energy sources, energy storage devices, and flexible loads to achieve optimized allocation and efficient utilization of power resources. They play a crucial role in promoting energy transition, ensuring grid stability, and improving economic efficiency. With the continued rapid increase in the proportion of renewable energy, virtual power plants have become an indispensable component of modern power systems.

[0003] However, the inventors have found at least the following technical problems in the related technologies: the related technologies are difficult to quantify the differences in interests and the degree of information asymmetry among multiple subjects, which leads to the failure of scheduling prediction and inaccurate risk assessment of adjustment capacity; the system lacks a dynamic balance mechanism between privacy protection and scheduling accuracy, and cannot achieve real-time and accurate correction of instructions while respecting the autonomy of the subjects. Summary of the Invention

[0004] One objective of this application is to provide a method, device, medium, and product for multi-entity collaborative power dispatching in a virtual power plant, at least to address the technical problems in related technologies, such as the difficulty in quantifying the differences in interests and the degree of information asymmetry among multiple entities, and the lack of a dynamic balance mechanism between privacy protection and dispatching accuracy.

[0005] To achieve the above objectives, some embodiments of this application provide the following aspects:

[0006] In a first aspect, some embodiments of this application provide a multi-entity collaborative power dispatching method for a virtual power plant. The method includes: determining the response deviation characteristics of each participating entity based on its decision-making autonomy information and real-time operating status; conducting a dispatching risk assessment based on the response deviation characteristics and the current system operating status to obtain an assessment result; generating a draft coordinated dispatching scheme based on the assessment result and the privacy protection rules of each participating entity; the privacy protection rules are used to characterize the mapping relationship between the data dimensions that each participating entity is allowed to disclose and the anonymization accuracy; and determining a target multi-entity coordinated dispatching instruction based on the feedback responses of each participating entity to the draft coordinated dispatching scheme.

[0007] Secondly, some embodiments of this application also provide an electronic device, the electronic device comprising: one or more processors; and a memory storing computer program instructions, which, when executed, cause the processor to perform the steps of the method described above.

[0008] Thirdly, some embodiments of this application also provide a computer-readable medium having computer program instructions stored thereon, which can be executed by a processor to implement the method described above.

[0009] Fourthly, some embodiments of this application also provide a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the method described above.

[0010] Compared with related technologies, the solution provided in this application, by determining the response deviation characteristics based on the decision-making autonomy information and real-time operating status of each participating entity, can quantify the behavioral differences of multiple entities in the game process into calculable features, thereby providing a predictive basis for subsequent scheduling and solving the problem of scheduling prediction failure caused by information asymmetry. By generating a draft coordinated scheduling scheme based on scheduling risk assessment results and privacy protection rules, a dynamic correlation can be established between ensuring the safe operation of the power grid and respecting the data privacy of each participating entity, solving the problem of imbalance between privacy protection and scheduling accuracy. Furthermore, by determining the target multi-entity coordinated scheduling instruction based on the feedback response of each participating entity and using the feedback mechanism to revise the draft, it can ensure that the final issued scheduling instruction conforms to the actual response capabilities of each participating entity, thereby improving the regulation capability and stability of the virtual power plant on the power grid. Attached Figure Description

[0011] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0012] Figure 1 An exemplary flowchart of a multi-entity collaborative power dispatching method for a virtual power plant is provided for some embodiments;

[0013] Figure 2 An exemplary structural diagram of an electronic device is provided for some embodiments. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0015] First Embodiment

[0016] The first embodiment relates to a multi-entity collaborative power dispatching method for virtual power plants. This method can be applied to virtual power plant control systems that include multiple participating entities (such as distributed power sources, energy storage systems, controllable loads, etc.). Figure 1 As shown, the method may include the following steps:

[0017] Step S101: Determine the response deviation characteristics of each participating entity based on the decision-making autonomy information and real-time operating status of each participating entity;

[0018] Step S102: Based on the response deviation characteristics and the current system operating status, a scheduling risk assessment is performed to obtain the assessment result;

[0019] Step S103: Based on the evaluation results and the privacy protection rules of each participating entity, a draft coordinated scheduling scheme is generated; the privacy protection rules are used to characterize the mapping relationship between the data dimensions that each participating entity is allowed to disclose and the anonymization accuracy.

[0020] Step S104: Based on the feedback responses from each participating entity to the draft coordinated scheduling scheme, determine the target multi-entity coordinated scheduling instruction.

[0021] The following sections will provide a detailed explanation of each of the above steps.

[0022] Specifically, regarding step S101, the participating entities may include the following categories:

[0023] The participating entities may include distributed power sources. For example, distributed power sources may include distributed photovoltaic power generation systems, small wind turbines, micro gas turbines, or biomass power generation equipment. For such participating entities, the decision-making autonomy information may involve output adjustment intentions based on weather forecast deviations or fuel cost fluctuations.

[0024] Furthermore, the participating entities may include distributed energy storage devices. For example, distributed energy storage devices may further include electrochemical battery packs, supercapacitors, flywheel energy storage devices, or pumped hydro storage systems. For energy storage participants, the real-time operating status may involve the current state of charge, battery health, and charge / discharge power limits; the response deviation characteristics can quantify the tendency of the distributed energy storage device to incompletely fulfill its obligations in order to retain electricity for arbitrage during peak electricity prices.

[0025] Additionally, the participating entities may include flexible loads or controllable loads. For example, flexible loads may include industrial production lines equipped with intelligent control terminals, central air conditioning systems in commercial buildings, data center cooling systems, and electric vehicle charging stations. For load-type participating entities, the response deviation characteristics can reflect the probability distribution of actual electricity consumption response values ​​deviating from the scheduling target due to constraints in production processes, indoor environmental comfort requirements, or user travel habits.

[0026] In some embodiments, the participating entities may further include microgrid systems with independent energy management capabilities. The microgrid system can be connected to the virtual power plant as a single node, and the decision-making autonomy information can be determined based on the game relationship between the internal supply and demand balance target of the microgrid system and external dispatch instructions.

[0027] For example, a virtual power plant system can achieve full-dimensional perception of all participating entities. The decision-making autonomy information characterizes the inherent decision-making logic and strategy preferences of participating entities when conducting power transactions or responding to dispatch; the real-time operating status characterizes the current physical regulation capacity boundaries of participating entities. By determining the response deviation characteristics, the potential difference between the actual actions performed by each participating entity and the ideal dispatch instructions can be quantitatively analyzed under conditions of no mandatory constraints or complete information transparency.

[0028] Regarding step S102, for example, the scheduling risk assessment can link the subject's behavioral deviations to the system security level. It is understood that by comprehensively calculating the response deviation characteristics in conjunction with the current system operating state (e.g., grid load peak-valley status, reserve capacity adequacy, tie-line power limitations, etc.), a more accurate risk assessment can be obtained. It should be understood that if the current system operating state is on the verge of a tight balance, even small response deviation characteristics may affect the system. The assessment result can be understood as a quantitative or hierarchical description of such potential systemic crisis.

[0029] For step S103, exemplarily, the process of generating the draft coordinated scheduling scheme is a process of finding a balance between power grid security requirements and commercial privacy requirements. The draft coordinated scheduling scheme, as a pre-scheduling scheme generated based on current assessment results and privacy constraints, can be used to seek consensus among multiple parties.

[0030] For step S104, for example, the virtual power plant system can distribute the draft coordinated scheduling scheme to each participating entity. Upon receiving the draft, each participating entity confirms or adjusts it based on the principle of maximizing its own interests, forming a feedback response. The virtual power plant system can then revise the draft coordinated scheduling scheme based on the feedback response, determining the target multi-entity coordinated scheduling instruction to be issued to each execution terminal, so that the final target multi-entity coordinated scheduling instruction is physically feasible and economically acceptable to all parties.

[0031] It is understandable that in related technologies, existing scheduling schemes typically assume that participating entities will strictly follow instructions in multi-stakeholder interest game dynamics. For example, in actual operation, distributed energy storage entities may choose to retain electricity during the current scheduling period in order to profit from subsequent periods of high electricity prices, thus exhibiting hoarding behavior; or, industrial controllable load entities may be unable to fully respond to peak shaving instructions due to temporary changes in production plans. Because the scheduling systems of related technologies often treat each participating entity as an idealized controlled node, they cannot predict the above behaviors before the instructions are issued, resulting in actual power deficits far exceeding theoretical calculations, thereby distorting the risk assessment of regulation capacity.

[0032] In balancing privacy protection and scheduling accuracy, related technologies typically employ fixed data acquisition strategies. For example, to achieve high-precision power control, a dispatch center might require factories to upload high-frequency internal production line energy consumption data. If a factory refuses to upload data due to concerns about protecting trade secrets, the dispatch system cannot include it in its coordination scope or can only use rough estimates. Conversely, if uploading is mandatory, it may lead to the loss of participating entities. This makes it difficult for related technologies to flexibly switch between the needs of grid security and the privacy demands of stakeholders, making it difficult to obtain sufficient dispatch data while also respecting the autonomy of stakeholders, and hindering the real-time and accurate correction of dispatch instructions.

[0033] It is not difficult to see that, compared with related technologies, the solution provided in this application, by introducing the analysis of the response deviation characteristics into the scheduling process, considers the execution deviation of the participating entities in the game environment, which helps to identify the execution differences that may be caused by multi-entity games; by dynamically combining the risk assessment results with privacy protection rules to generate the scheduling draft, it can reduce the resistance of participating entities caused by excessive data requests, and reduce the power grid security risks caused by information opacity.

[0034] Second Embodiment

[0035] The second embodiment relates to a multi-entity collaborative power dispatching method for a virtual power plant. The second embodiment is an improvement upon the first embodiment, specifically in that it provides a concrete implementation method for determining the response deviation characteristics of each participating entity based on their decision-making autonomy information and real-time operating status, utilizing blockchain data storage and game theory computation techniques.

[0036] Optionally, in some embodiments, the decision autonomy information may include the bidding strategy range and marginal cost curve of each subject; the response deviation characteristics may include the response scheduling deviation probability calculated based on game theory algorithms.

[0037] For example, taking a distributed energy storage device as an example among the participating entities, the decision-making autonomy information may include the range of bidding strategies generated by the distributed energy storage device based on cycle life loss and expected electricity prices. The virtual power plant system can use game theory algorithms to simulate the strategy response of the distributed energy storage device in pursuit of its own profit maximization objective. If the output quota allocated by the dispatch command is above the marginal cost curve of the distributed energy storage device, but the distributed energy storage device finds through game theory deduction that it can obtain higher expected revenue by delaying output, then the response dispatch deviation probability calculated based on the game theory algorithm can objectively reflect the possibility of behavioral deviation of the distributed energy storage device at the execution level, thereby enabling the virtual power plant system to transform abstract profit demands into measurable risk probabilities.

[0038] Specifically, step S101, which involves determining the response deviation characteristics of each participating entity based on their decision-making autonomy information and real-time operational status, may include:

[0039] Step S1011: Extract historical performance data and current physical constraints of each participating entity using blockchain technology;

[0040] Step S1012: Verify the decision-making autonomy information and the real-time operating status based on the historical performance data and the current physical constraints, respectively;

[0041] Step S1013: The verified decision-making autonomy information and the real-time operating status are encapsulated and uploaded to the blockchain to generate an immutable basis for interest objective difference analysis.

[0042] Step S1014: Based on the aforementioned interest objective difference analysis, calculate the behavioral offset of each participating entity under the goal of maximizing its own interests, and determine the response deviation characteristics.

[0043] Regarding step S1011, for example, the blockchain network, as a distributed ledger, can store historical performance data such as past transaction records and response history of each participating entity, as well as current physical constraints such as device parameters and grid connection protocols. In this step, by utilizing the tamper-proof characteristics of the blockchain, the reliability of the data source for calculating the response deviation characteristics can be ensured, preventing participating entities from tampering with the basic data for profit or other reasons.

[0044] Regarding step S1012, for example, the virtual power plant system can compare the bidding strategy currently reported by the participating entity with the participating entity's historical performance data to verify its credit rating; simultaneously, it can also compare the reported real-time operating status with the current physical constraints to eliminate abnormal data. For example, if the adjustable capacity reported by a participating entity exceeds the upper limit of the participating entity's physical constraints, the verification will fail.

[0045] Regarding step S1013, for example, the verified data can be packaged into blocks, broadcast through the consensus mechanism, and stored on the blockchain, forming the basis for analyzing differences in interests. This ensures that all participants conduct subsequent game theory analysis based on the same set of real data, reducing data silos.

[0046] The aforementioned benefit objective difference analysis basis is used to characterize the deviation dimension between the micro-level benefit demands of each participating entity and the overall dispatch objective of the power grid. For example, consider a participating entity including distributed energy storage devices. The overall dispatch objective of the power grid may require the distributed energy storage devices to discharge at full power during the current peak load period to alleviate the power supply pressure on the local power grid and maintain system frequency stability. However, the micro-level benefit demands of the distributed energy storage devices can be calculated based on the marginal cost curve in the decision autonomy information, determining that the battery cycle life loss cost caused by the current discharge behavior is higher than the current dispatch compensation benefit, or predicting a higher electricity price arbitrage space in the future based on the bidding strategy range. In this case, the benefit objective difference analysis basis can integrate the cost strategy data of the distributed energy storage devices with the real-time adjustment demand data of the power grid, characterizing the deviation dimension between the benefit orientation of the participating entities and the overall dispatch objective in terms of power quantity (e.g., the 80kW difference between the 100kW discharge demand of the power grid and the 20kW discharge quantity of the energy storage device's optimal economic output) and time response willingness. Through this quantitative characterization, the virtual power plant system can identify the potential default risks of the distributed energy storage device when executing dispatch instructions.

[0047] For step S1014, for example, the virtual power plant system can construct a Bayesian game model, setting the real cost type of each participating entity as private information. By solving the Nash equilibrium solution, the optimal strategy output of each participating entity in pursuing its own profit maximization is calculated. The difference between this optimal strategy output and the expected output of the power grid is the behavioral offset, which, after normalization, yields the response scheduling deviation probability.

[0048] Optionally, the response scheduling deviation probability can be quantified using a Bayesian game model and a Sigmoid function, as follows:

[0049] For example, suppose a virtual power plant includes N participating entities, denoted by I = {1, 2, ..., N}. Define the entities. The benefit function (utility function) is:

[0050] ;

[0051] in: as the main body The interest function, as the main body The actual output decision value; λ is the current incentive electricity price or compensation price; For the profit function; The cost function can include fuel costs, battery wear costs, etc. Because of deviation from the expected dispatch instructions of the power grid The resulting penalties; This refers to the power grid's desired dispatch instructions.

[0052] In scenarios with information asymmetry, the true cost / capability types of each entity This is private information. A Bayesian game model can be used to solve for the Nash equilibrium point, yielding the optimal effort when the agent only considers maximizing its own interests. Define the response scheduling deviation probability. for:

[0053] ;

[0054] in: The difference in utility between a subject's self-interest-optimal strategy and its obedience-to-dispatch strategy; The sensitivity coefficient can be set based on historical data or expert experience; It provides power for the dispatching required by the power grid.

[0055] The above formula shows that the greater the private benefit obtained by the subject through concealing information (i.e., the greater the benefit), the more likely the subject is to gain private benefits by concealing information. The larger the value, the higher the probability of a response scheduling deviation (i.e., the closer it is to 1); conversely, if obeying scheduling is more advantageous, the probability of deviation decreases.

[0056] Furthermore, the degree of information state asymmetry can be calculated based on the response scheduling deviation probability and capacity weight of each participating entity.

[0057] For example, the degree of information state asymmetry It can be calculated using the following formula:

[0058] ;

[0059] in, as the main body Capacity weight in the system. A preset threshold for the degree of information asymmetry, if If the degree of information asymmetry exceeds a preset threshold, the system can be determined to be in a state of high information asymmetry, triggering the data update mechanism of the blockchain smart contract.

[0060] It is not difficult to see that, in the embodiments of this application, by using blockchain technology to build a basis for analyzing differences in interests, the problem of data distrust among multiple subjects can be solved; by using game theory algorithms to quantify abstract decision-making autonomy into specific response scheduling deviation probabilities, the default risks hidden behind interest demands can be revealed, enabling the scheduling system to analyze the true intentions of each participating subject.

[0061] Third Embodiment

[0062] The third embodiment relates to a multi-entity collaborative power dispatching method for virtual power plants. The third embodiment is an improvement upon the second embodiment, specifically in that it provides a concrete implementation method for assessing dispatching risks based on the response deviation characteristics and the current system operating state, thereby obtaining the assessment results. In particular, it introduces the quantification of information asymmetry and the mapping of the physical security model.

[0063] Specifically, step S102, which involves assessing scheduling risks based on the response deviation characteristics and the current system operating state to obtain the assessment result, may include:

[0064] Step S1021: Analyze the discrete distribution of the response deviation characteristics of each participating entity in the interest objective difference analysis basis to determine the degree of information state asymmetry of the system;

[0065] Step S1022: If the degree of information asymmetry exceeds a preset threshold, the updated operating status of each participating entity is obtained through the blockchain sharing mechanism.

[0066] Step S1023: Calculate the risk of insufficient adjustment capacity of the system based on the updated operating status, and use the risk of insufficient adjustment capacity as the basis for the evaluation result.

[0067] For step S1021, for example, the virtual power plant system can extract the response scheduling deviation probabilities of multiple participating entities from the interest objective difference analysis basis, and calculate the statistically discrete indicators of these probability values, such as variance, standard deviation, or coefficient of variation.

[0068] In some embodiments, if the response scheduling deviation probabilities of the ten participating entities under the control of the virtual power plant system are all distributed within a very narrow range of 0.05 to 0.08, the discrete distribution of the response deviation characteristics exhibits a high degree of concentration. This concentrated distribution indicates that the deviation between the interests of each participating entity and the scheduling objective has a high degree of consistency and predictability. Based on this, the virtual power plant system can determine that the information asymmetry of the system is low, meaning that the dispatch center's control over the overall behavior is relatively transparent.

[0069] In other embodiments, if the response scheduling deviation probability of five participating entities is 0.02, while the response scheduling deviation probability of another five participating entities rises to 0.45 due to drastic fluctuations in internal production plans, the discrete distribution of the response deviation characteristics exhibits a clear bimodal shape or high variance. This extremely high degree of dispersion characterizes significant differences among different participating entities in concealing their true intentions or execution capabilities, making it impossible for the virtual power plant system to accurately predict all entities using a unified empirical model. In this case, the dispersion of this distribution can be directly quantified as a high level of information asymmetry, indicating that the scheduling system faces severe information barriers and potential risks of unauthorized responses.

[0070] In this way, the virtual power plant system can transform the abstract problem of information asymmetry into an observable distribution of mathematical features, providing a basis for decision-making on whether to trigger the blockchain sharing mechanism to obtain updated operating status.

[0071] Regarding step S1022, for example, when it is determined that the degree of information asymmetry exceeds a preset threshold, the mandatory sharing logic in the smart contract can be automatically triggered, requiring each participating entity to upload the latest, more granular update of the running status through the blockchain, thereby reducing information barriers. Here, smart contracts are one of the core components of blockchain technology, a mature technology, and will not be described in detail in this embodiment.

[0072] Regarding step S1023, the adjustable capacity refers to the maximum power range that participating entities within the virtual power plant system can increase or decrease within a specific time period, limited by physical performance, operational constraints, and current output status. It reflects the system's actual adjustment potential in response to grid commands. In practical applications, the updated operating status can be used to recalculate the adjustable capacity of the entire system to determine whether the system meets current adjustment requirements.

[0073] Optionally, in some embodiments, the scheduling risk assessment may further include the following steps:

[0074] Step S201: Based on the response deviation characteristics, analyze the expected response deficit of multiple entities when receiving scheduling instructions, and determine the total power deficit of the system by combining the real-time operating status of each participating entity.

[0075] Step S202: Input the total power deficit of the system into a preset power grid frequency response model and calculate the expected system frequency offset value;

[0076] Step S203: Determine the power grid frequency stability threat level based on the system frequency offset value, and use the power grid frequency stability threat level as a quantitative indicator for the dispatch risk assessment.

[0077] Regarding step S201, the expected response deficit is used to characterize the power deviation between the actual output of a participant and the dispatch command demand, predicted after considering the game psychology and default probability of the participating entities. For example, the expected response deficit can be determined by multiplying the expected command value by the corresponding response dispatch deviation probability. Specifically, the virtual power plant system can weight the command power allocated to a specific participating entity with the probability of deviation for that participating entity to obtain the power gap that the participating entity may be unable to fulfill as agreed.

[0078] Furthermore, the virtual power plant system can combine the real-time operating status of each participating entity (such as the current state of charge of energy storage devices or production constraints of industrial loads) to correct and sum the aforementioned gaps, thereby determining the total power deficit of the system. This step, by introducing probabilistic quantification, transforms the ideal state of traditional scheduling, which assumes complete compliance of the entities, into a real state based on game risk prediction. This allows the virtual power plant system to anticipate potential system-level power deficits caused by participating entities pursuing individual interests.

[0079] For step S202, for example, the power grid frequency response model may include physical parameters such as system inertia and damping coefficient. Through differential equation simulation, the total power deficit of the system can be mapped to the expected system frequency offset value. The system frequency offset value is used to characterize the degree to which the steady-state frequency of the power grid deviates from the rated frequency under a specific system power deficit impact and its potential impact on the synchronous operation safety of the power grid.

[0080] For example, in step S203, if the system frequency offset value exceeds a safety threshold (e.g., 0.5 Hz), the power grid frequency stability threat level can be determined to be high risk.

[0081] Optionally, the power grid frequency response model can be described using classical system frequency dynamic differential equations, as follows:

[0082] For example, suppose the estimated total power deficit of the system is... The calculation method is as follows:

[0083] ;

[0084] in, It is based on the updated state data and the actual response values ​​of each subject obtained through master-slave game simulation.

[0085] Based on system inertia and damping coefficient Frequency deviation The change over time satisfies the following differential equation:

[0086] ;

[0087] By solving this equation, the expected maximum frequency deviation can be obtained. .

[0088] The power grid frequency stability threat level This can be obtained through maximum frequency deviation mapping. One possible mapping method is:

[0089] ;

[0090] Of course, in some other examples, continuous functions (such as the normalization exponent) can also be used to... The risk coefficient is mapped to a range of 0 to 1, but this application does not specify a particular limit for it.

[0091] It should be noted that this embodiment can also be an improvement based on the first embodiment.

[0092] It is not difficult to see that in this embodiment, a correlation is established between information asymmetry and the power grid frequency stability threat level. Information barriers are perceived by monitoring the distribution of response deviation characteristics, and on-demand updates are achieved using a blockchain sharing mechanism. By inputting the power deficit derived from game theory into the physical frequency model, game risk is transformed into a specific power grid frequency stability threat level, thus achieving the coupling of scheduling decisions and power grid physical security.

[0093] Fourth embodiment

[0094] The fourth embodiment relates to a multi-entity collaborative power dispatching method for virtual power plants. The fourth embodiment is an improvement upon the first embodiment, specifically in that it provides a concrete implementation method for generating a draft coordinated dispatching scheme based on the evaluation results and the privacy protection rules of each participating entity.

[0095] Specifically, step S103, which involves generating a draft coordinated scheduling scheme based on the evaluation results and the privacy protection rules of each participating entity, may include:

[0096] Step S1031: Determine the data openness level of each participating entity based on the evaluation results;

[0097] Step S1032: Match the corresponding target desensitization precision from the privacy protection rules according to the data openness level;

[0098] Step S1033: Based on the target desensitization accuracy, the original operating data of each participating entity is masked or aggregated, and the processed data is used as a constraint to generate the draft of the coordinated scheduling scheme.

[0099] For step S1031, for example, the assessment result (such as the power grid frequency stability threat level) determines the urgency of the system's need for data transparency. If the threat level is high, a higher data openness level is set; if the threat level is low, a lower data openness level is set. The data openness level is used to characterize the depth, frequency, and privacy boundaries that the virtual power plant system requires each participating entity to disclose internal operational data under the current power grid security risk constraints.

[0100] For step S1032, for example, the privacy protection rule presets processing methods at different levels. For instance, a high data openness level corresponds to plaintext of the original data or low-loss desensitization; a low data openness level corresponds to high-strength masking or providing only aggregated data. The target desensitization accuracy is the specific data obfuscation parameter (such as the noise parameter of differential privacy).

[0101] Regarding step S1033, for example, when generating a draft, the virtual power plant system can use anonymized data as input to the optimization model. This ensures that the generated coordinated scheduling scheme draft is calculated within boundaries that satisfy the privacy requirements of all participating parties.

[0102] It should be noted that this embodiment may also be an improvement based on the second embodiment and / or the third embodiment.

[0103] It is easy to see that in this embodiment, by introducing a data openness level linked to the evaluation results, flexible adjustment of privacy protection is achieved. When the power grid faces a security threat, security is prioritized; when the system is secure, the trade secrets of participating entities can be respected to the greatest extent. This mechanism enables the generated scheduling draft to have consistent attributes, reducing the probability that participating entities will refuse to execute it due to privacy concerns.

[0104] Fifth embodiment

[0105] The fifth embodiment relates to a multi-entity collaborative power dispatching method for virtual power plants. The fifth embodiment is an improvement upon the first embodiment, specifically in that it provides a method for determining the specific implementation of the target multi-entity collaborative dispatching instruction based on the feedback responses of each participating entity to the draft coordinated dispatching scheme, thus achieving closed-loop correction.

[0106] Specifically, step S104, which involves determining the target multi-agent coordinated scheduling instruction based on the feedback responses of each participating entity to the draft coordinated scheduling scheme, may include:

[0107] Step S1041: Verify the committed response power of each participating entity to the draft coordinated scheduling scheme through a blockchain smart contract;

[0108] Step S1042: Calculate the deviation between the committed response power and the corresponding output in the draft coordinated scheduling scheme, and determine the response scheduling deviation correction value;

[0109] Step S1043: Use the response scheduling deviation correction value to calculate the compensation for the output of each participating entity in the draft coordinated scheduling scheme, and generate the target multi-entity coordinated scheduling instruction.

[0110] Regarding step S1041, in some examples, after the virtual power plant system issues the draft coordinated dispatch scheme, a certain energy storage participant may provide feedback on a planned response output of 10MW (i.e., the committed response power). This energy storage participant uses its private key to encrypt and sign the 10MW value and timestamp, and uploads the signed data to the blockchain.

[0111] Furthermore, smart contracts deployed on the blockchain can automatically call the public key of the energy storage participant for decryption and verification. If the verification passes, it proves that the 10MW commitment was indeed issued by the energy storage participant, and not forged by a third party, and that the data was not tampered with during transmission. Since the signature record is permanently stored in the distributed ledger, if the energy storage participant fails to achieve the 10MW output in the actual execution phase, the energy storage participant cannot deny the power commitment it made.

[0112] For step S1042, for example, the virtual power plant system can compare the promised response power with the ideal value in the draft to obtain the deviation. Based on the deviations of all participating entities, a global response scheduling deviation correction value is calculated.

[0113] For example, in step S1043, if a shortfall is found in the total committed power, the virtual power plant system can use the response scheduling deviation correction value to adjust the instructions of other participating entities with adjustment capabilities, or call up backup resources to generate the final target multi-entity coordinated scheduling instructions after compensation and balance.

[0114] Optionally, the response scheduling deviation correction value can be calculated using PID control principles or the learning rate method, as detailed below:

[0115] For example, let the subject be... The preset output in the draft coordinated scheduling scheme is: The principal commitment response power obtained after verification through the blockchain smart contract is Define the deviation. Then the response scheduling deviation correction value It can be calculated using the following formula:

[0116]

[0117] in: This is the proportional adjustment coefficient; This is the integral adjustment coefficient; the integral term represents the cumulative correction to historical frequency deviation or historical execution deviation, which can be calculated using past deviation data recorded on the blockchain.

[0118] Furthermore, the ultimate goal is multi-entity coordinated scheduling instructions. for:

[0119]

[0120] Through this correction mechanism, scheduling instructions can be dynamically compensated based on the actual commitment feedback from each participating entity, reducing response deviations at the execution level.

[0121] It should be noted that this embodiment may also be an improvement based on one or more embodiments in the second to fourth embodiments.

[0122] It is easy to see that in this embodiment, a complete closed-loop control circuit is formed through smart contract verification and real-time deviation compensation. The smart contract solves the trust problem of feedback data, while the calculation and application of deviation correction values ​​ensure that the final issued multi-agent coordinated scheduling instructions have fully absorbed the game results and feedback deviations of each participating entity, thereby significantly improving the execution accuracy of virtual power plant scheduling.

[0123] In summary, the virtual power plant multi-entity collaborative power dispatching method provided in this application has significant beneficial effects.

[0124] 1. By using blockchain technology to immutably record decision-making autonomy information and real-time operational status, and then using game theory algorithms to calculate response deviation characteristics based on this, the approach can effectively address the technical challenges of information asymmetry and difficulty in quantification in multi-agent environments. Because the probability of response deviations for each participating entity can be quantified in advance, the scheduling system no longer blindly relies on idealized models but can instead make predictions based on real game risks.

[0125] 2. This application combines dispatch risk assessment with a physical-level power grid frequency response model and dynamically matches privacy protection rules based on this. This mechanism establishes a causal mapping relationship between power grid physical security and subject data privacy. When system frequency stability faces a high threat, security is ensured by increasing transparency; when the threat is low, subject privacy protection rules are respected. This dynamic balancing mechanism eliminates the concerns of participating subjects about data leakage, thereby improving the participation subjects' cooperation and execution rate with dispatch instructions.

[0126] 3. This application verifies the feedback response through a blockchain smart contract and generates the final target multi-entity coordinated dispatch instruction based on the deviation correction value. This closed-loop correction step ensures that the dispatch instruction is not only theoretically optimal, but also fully considers the actual commitment capabilities of each participating entity at the execution level, thereby significantly reducing the response deviation after the dispatch instruction is issued and effectively improving the virtual power plant's support capability for grid frequency stability.

[0127] The steps of the various methods described above are only for clarity. In practice, they can be combined into one step or some steps can be split into multiple steps. As long as they include the same logical relationship, they are all within the scope of protection of this application. Adding insignificant modifications or introducing insignificant designs to the algorithm or process, but without changing the core design of the algorithm and process, are also within the scope of protection of this application.

[0128] Furthermore, some embodiments of this application also provide an electronic device. The electronic device can be various forms of digital computer, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, mainframe computers, cellular phones, smartphones, wearable devices, and other similar computing devices.

[0129] The electronic device includes: one or more processors; and a memory storing computer program instructions, which, when executed, cause the processor to perform the steps of the methods provided in any one or more of the above embodiments.

[0130] Figure 2 An exemplary structural diagram of the electronic device is disclosed. The electronic device includes one or more processors 1101, a memory 1102, an input device 1103, and an output device 1104. The various components are interconnected via a bus or other means (the diagram shows an example of bus connection). The processor 1101 can be used to execute instructions stored in the memory 1102 to control the overall operation of the electronic device. The memory 1102 may include a program storage area and a data storage area, wherein the program storage area stores the operating system and applications required for at least one function; the data storage area stores data created according to the use of the electronic device, etc. The memory 1102 may include high-speed random access memory and may also include non-transitory memory, such as disk storage devices, flash memory devices, or other non-transitory solid-state storage devices. In some embodiments, the memory 1102 may also include storage resources located remotely to the processor and accessible via a network.

[0131] Input device 1103 can be used to receive input numerical or character information or user operation signals, such as a touch screen, keypad, mouse, trackpad, touchpad, indicator, one or more mouse buttons, trackball, joystick, etc. Output device 1104 may include display devices (such as liquid crystal displays, light-emitting diode displays, plasma displays, and optional touch screens), auxiliary lighting devices (such as LEDs), and haptic feedback devices (such as vibration motors), etc.

[0132] To facilitate user interaction, the electronic device may be configured to include a display device (such as an LCD or CRT monitor) and input devices such as a keyboard and pointing devices (e.g., a mouse or touchpad). Feedback can be any form of sensory feedback (e.g., visual feedback, auditory feedback); input may also be received via voice, touch, or other means.

[0133] This application also relates to a computer-readable medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the methods provided in any one or more of the above embodiments. This computer-readable medium may be a memory included in an electronic device, or it may be a standalone storage medium not assembled into the device.

[0134] It should be noted that the computer-readable medium described in this application may be a computer-readable signal medium, a computer-readable storage medium, or a combination of both. Examples include, but are not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. Specific examples of storage media may include, but are not limited to, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory, optical fibers, portable CD-ROMs, optical storage devices, magnetic storage devices, etc., or any suitable combination thereof.

[0135] Computer-readable media may store one or more programs that can be used by or in conjunction with an instruction execution system. The media may be permanent or non-permanent, removable or non-removable, and may store information by any method or technology, including computer-readable instructions, data structures, program modules, or other data.

[0136] The computer program code used to implement the operations of this application can be written in any combination of one or more programming languages, including object-oriented programming languages ​​(such as Java, Smalltalk, and C++) and conventional procedural programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer, partially on a remote computer, or entirely on a remote computer or server. The remote computer can be connected to the user's computer via any network (including a local area network or a wide area network) or can be connected to an external computer.

[0137] In the above embodiments, the functions can be implemented in whole or in part by software, hardware, firmware, or any combination thereof, for example, by using application-specific integrated circuits, general-purpose computers, or other similar hardware devices. In some embodiments, the software program of this application can be executed by a processor to implement the steps or functions; it can also be implemented by hardware, for example, as a circuit that works in conjunction with the processor to execute the steps or functions.

[0138] This application also provides a computer program product, including one or more computer programs / instructions, which, when executed by a processor, generate all or part of the processes or functions described in this application. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium or transmitted from one storage medium to another via wired (e.g., DSL) or wireless (e.g., wireless, microwave) means. The computer-readable storage medium may be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive).

[0139] The flowcharts or block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of devices, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-specific system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0140] The scope of this application is defined by the appended claims rather than the foregoing description, and is therefore intended to encompass all variations falling within the meaning and scope of equivalents of the claims. No reference numerals in the claims should be construed as limiting the scope of the claims. Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a device claim may also be implemented by a single unit or device in software or hardware. Terms such as "first," "second," etc., are used only for distinguishing descriptions and do not indicate any particular order, nor should they be construed as indicating or implying relative importance.

[0141] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily made by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims, and the above embodiments should be regarded as exemplary and non-limiting.

Claims

1. A method for multi-entity collaborative power dispatching in a virtual power plant, characterized in that, The method includes: Based on the decision-making autonomy information and real-time operational status of each participating entity, the response deviation characteristics of each participating entity are determined; Based on the response deviation characteristics and the current system operating status, a scheduling risk assessment is performed to obtain the assessment results; Based on the evaluation results and the privacy protection rules of each participating entity, a draft coordinated scheduling scheme is generated; the privacy protection rules are used to characterize the mapping relationship between the data dimensions that each participating entity is allowed to disclose and the anonymization accuracy. Based on the feedback responses from each participating entity to the draft coordinated scheduling scheme, the target multi-entity coordinated scheduling instruction is determined.

2. The method according to claim 1, characterized in that, The decision-making autonomy information includes the range of each entity's pricing strategy and marginal cost curve; The response deviation characteristics include the response scheduling deviation probability calculated based on game theory algorithms.

3. The method according to claim 1, characterized in that, The determination of response deviation characteristics for each participating entity based on their decision-making autonomy information and real-time operational status includes: Extract historical performance data and current physical constraints of each participating entity using blockchain technology; The decision-making autonomy information and the real-time operating status are verified based on the historical performance data and the current physical constraints, respectively. The verified decision-making autonomy information and the real-time operating status are encapsulated on the blockchain to generate an immutable basis for analyzing differences in interest objectives. Based on the aforementioned analysis of differences in benefit objectives, the behavioral offset of each participating entity under the goal of maximizing its own benefit is calculated, and the response deviation characteristics are determined.

4. The method according to claim 3, characterized in that, The scheduling risk assessment based on the response deviation characteristics and the current system operating status yields the assessment results, including: By analyzing the discrete distribution of the response deviation characteristics of each participating entity in the aforementioned interest objective difference analysis, the degree of information state asymmetry of the system is determined. If the degree of information asymmetry exceeds a preset threshold, the updated operating status of each participating entity will be obtained through the blockchain sharing mechanism. The risk of insufficient regulation capacity is calculated based on the updated operating status of the system, and this risk is used as the basis for the evaluation results.

5. The method according to claim 4, characterized in that, The scheduling risk assessment also includes: Based on the aforementioned response deviation characteristics, the expected response deficit of multiple entities when receiving scheduling instructions is analyzed, and the total power deficit of the system is determined by combining the real-time operating status of each participating entity. Input the total power deficit of the system into a preset power grid frequency response model to calculate the expected system frequency offset value; The power grid frequency stability threat level is determined based on the system frequency offset value, and the power grid frequency stability threat level is used as a quantitative indicator for the dispatch risk assessment.

6. The method according to claim 1, characterized in that, Based on the evaluation results and the privacy protection rules of each participating entity, a draft coordinated scheduling plan is generated, including: The data openness level of each participating entity will be determined based on the assessment results. Match the corresponding target desensitization precision from the privacy protection rules based on the data openness level; Based on the target desensitization accuracy, the original operating data of each participating entity is masked or aggregated, and the processed data is used as a constraint to generate the draft of the coordinated scheduling scheme.

7. The method according to any one of claims 1 to 6, characterized in that, The step of determining the target multi-agent coordinated scheduling instruction based on the feedback responses of each participating entity to the draft coordinated scheduling scheme includes: The commitment response power of each participating entity to the draft coordinated scheduling scheme is verified through blockchain smart contracts. Calculate the deviation between the promised response power and the corresponding output in the draft coordinated scheduling scheme, and determine the response scheduling deviation correction value; The response scheduling deviation correction value is used to compensate the output of each participating entity in the draft coordinated scheduling scheme, and the target multi-entity coordinated scheduling instruction is generated.

8. An electronic device, characterized in that, The electronic device includes: One or more processors; and A memory storing computer program instructions, which, when executed, cause the processor to perform the steps of the method as described in any one of claims 1 to 7.

9. A computer-readable medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 7.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 7.

Images