Source-load bilateral virtual power plant real-time regulation operation method, device and equipment

Abstract

The application provides a kind of source and load bilateral virtual power plant real-time regulation operation method, device and equipment, relating to virtual power plant scheduling technical field.The method comprises: collecting the distribution network parameter of target distribution network, real-time operation parameter and the source and load data of the distributed equipment connected with target distribution network;Under the constraint of distribution network parameter, according to real-time operation parameter and source and load data, generate uncertainty scenario data;According to uncertainty scenario data, calculate conditional value at risk;Conditional value at risk is input into upper model, with the maximum sharp ratio as the goal, to obtain electricity price incentive strategy;Electricity price incentive strategy is input into lower model, with the minimum user energy cost as the goal, to obtain the optimal regulation operation strategy of virtual power plant.The application quantifies the influence of uncertain factors on incentive policy through conditional value at risk, avoids the scheduling maladjustment caused by ignoring extreme risk in incentive strategy, and makes the operation of virtual power plant more robust under complex scenarios.

Description

Technical Field

[0001] This invention relates to the field of virtual power plant dispatching technology, and in particular to a method, apparatus and equipment for real-time control and operation of a virtual power plant on both the source and load sides. Background Technology

[0002] With the acceleration of energy transition, distributed energy resources are being integrated into distribution networks on a large scale. Virtual power plants (VPS), as core platforms aggregating heterogeneous resources such as photovoltaics, energy storage, electric vehicles, and flexible loads, participate in peak shaving and valley filling of the power grid and smooth fluctuations in renewable energy output through coordinated regulation. They have become a key technology for improving power system flexibility and optimizing the utilization efficiency of distribution network assets, and are of great significance for promoting renewable energy consumption and the safe and stable operation of the power grid. Incentive strategies are the core link for VPS to guide user-side resources to actively respond to dispatch. Their rationality directly determines the willingness of distributed resources to respond and the effectiveness of regulation. That is, only through precise incentives can dispersed resources be leveraged to form aggregated regulation capabilities, thereby supporting VPS in achieving refined coordinated dispatch of distributed resources and realizing efficient collaboration between VPS and the distribution network.

[0003] However, the inventors discovered that significant challenges remain in the formulation of incentive strategies for virtual power plants and their subsequent dispatch implementation. Specifically, existing virtual power plant optimization models often prioritize maximizing expected returns, failing to adequately consider operational risks arising from uncertainties such as fluctuations in renewable energy output and sudden changes in load demand. Furthermore, they lack risk quantification mechanisms adapted to dispatch requirements. This results in poor robustness of incentive strategies in real-world, complex scenarios. They cannot provide precise guidance for the coordinated dispatch of distributed resources, nor can they balance the economic benefits of virtual power plants with the safe operation requirements of the distribution network. Inappropriate incentives may even lead to insufficient resource response, failure to implement dispatch commands, or grid overload, hindering the full realization of the potential for coordinated dispatch of distributed resources. Therefore, it is urgent to optimize the logic of incentive strategy formulation to provide reliable support for the efficient and safe coordinated dispatch of distributed resources in virtual power plants. Summary of the Invention

[0004] This invention provides a method, apparatus, and equipment for real-time control and operation of a virtual power plant on both the source and load sides. This addresses the problem that current incentive strategies fail to quantify the risks of uncertainties, resulting in scheduling strategies determined based on incentive strategies being unable to balance the economic benefits of virtual power plants with the safe operation requirements of the distribution network.

[0005] In a first aspect, embodiments of the present invention provide a method for real-time control and operation of a source-load dual-side virtual power plant, comprising: Collect distribution network parameters, real-time operating parameters, and source-load data of distributed devices connected to the target distribution network. Under the constraints of distribution network parameters, uncertain scenario data are generated based on real-time operating parameters and source load data; Based on uncertain scenario data, the conditional value of risk is calculated; and the conditional value of risk is input into the upper-level model to obtain the electricity price incentive strategy with the goal of maximizing the Sharpe ratio; wherein, the conditional value of risk is used to characterize the loss under extreme scenarios; By inputting the electricity price incentive strategy into the lower-level model, and aiming to minimize the user's energy costs, the optimal control and operation strategy of the virtual power plant is obtained.

[0006] In one possible implementation, the uncertainty scenario data includes the probability of the uncertain scenario occurring and the revenue of the virtual power plant under the uncertain scenario. Based on the uncertainty scenario data, the conditional value of risk is calculated, including: Based on the probability of occurrence of uncertain scenarios in each uncertain scenario data and the revenue of virtual power plants under uncertain scenarios, uncertain scenarios are filtered to obtain target uncertain scenarios; among them, the probability of occurrence of uncertain scenarios in target uncertain scenarios is greater than a preset probability threshold, or the revenue of virtual power plants under uncertain scenarios is less than a preset revenue threshold. The conditional value of risk is obtained by averaging the returns in scenarios where the objective is uncertain.

[0007] In one possible implementation, the uncertainty scenario data includes the probability of the uncertain scenario occurring and the revenue of the virtual power plant under the uncertain scenario. Based on the uncertainty scenario data, the conditional value of risk is calculated, including: Based on the probability of occurrence of uncertain scenarios in each uncertain scenario data and the revenue of virtual power plants under uncertain scenarios, uncertain scenarios are filtered to obtain target uncertain scenarios; among them, the probability of occurrence of uncertain scenarios in target uncertain scenarios is greater than a preset probability threshold, or the revenue of virtual power plants under uncertain scenarios is less than a preset revenue threshold. The economic risk value is obtained by averaging the returns of scenarios with uncertain objectives. For each uncertain scenario, the real-time operating parameters and the source-load data of the distributed devices connected to the target distribution network are used to calculate the real-time power flowing through the converter and the line transmission power. Uncertain scenarios where the real-time power exceeds the rated capacity of the converter or the line transmission power is less than the expected minimum value of the load rate are considered high-risk scenarios. Calculate the overload loss for each high-risk scenario based on real-time power and rated capacity; Calculate the low load loss for each high-risk scenario based on the expected minimum value of line transmission power and load rate; The average of the sum of overload loss and low load loss under each high-risk scenario is taken as the asset risk value under the high-risk scenario. The sum of the average return in scenarios with uncertain objectives and the asset risk value in high-risk scenarios is used as the conditional risk value.

[0008] In one possible implementation, the uncertainty scenario data includes the probability of the uncertain scenario occurring and the revenue of the virtual power plant under the uncertain scenario. Based on the uncertainty scenario data, the conditional value of risk is calculated, including: For each uncertain scenario, the fluctuation contribution of each distributed device is calculated based on its corresponding historical and real-time operating parameters. The sum of the fluctuation contribution of each distributed device and the product of the source load data of each distributed device is used as the revenue adjustment value. The difference between the revenue of the virtual power plant under uncertain scenarios and the revenue adjustment value is used as the target revenue of the virtual power plant under each uncertain scenario. Based on the probability of occurrence of uncertain scenarios in each uncertain scenario data and the target revenue of the virtual power plant under uncertain scenarios, uncertain scenarios are filtered to obtain target uncertain scenarios; among them, the probability of occurrence of uncertain scenarios in target uncertain scenarios is greater than a preset probability threshold, or the target revenue of the virtual power plant under uncertain scenarios is less than a preset revenue threshold. The conditional value of risk is obtained by averaging the target returns in scenarios where the target is uncertain.

[0009] In one possible implementation, the Sharpe ratio is determined based on the ratio of the expected total profit of the virtual power plant to the conditional risk value. Before inputting the conditional risk value into the upper-level model to obtain the electricity price incentive strategy with the goal of maximizing the Sharpe ratio, the following steps are also included: Determine the net load curve and divide the net load curve into peak and valley periods; Time-of-use pricing is determined based on peak and off-peak periods; Accordingly, the conditional value at risk is input into the upper-level model, with the goal of maximizing the Sharpe ratio, to obtain the electricity price incentive strategy, including: By inputting conditional value at risk and time-of-use pricing into the upper-level model, and under the market information constraints of the virtual power plant, the upper-level model is solved with the goal of maximizing the Sharpe ratio, thus obtaining the electricity price incentive strategy.

[0010] In one possible implementation, a net load curve is determined and divided into peak and trough periods, including: Based on real-time operating parameters and source load data, determine the net load curve for each time period, and determine the maximum and minimum values ​​of the net load curve; Membership is determined based on the maximum and minimum values ​​of the net load curve to divide the net load curve into peak and valley periods.

[0011] In one possible implementation, membership is assigned based on the maximum and minimum values ​​of the net load curve to divide the net load curve into peak and trough periods, including: For the maximum and minimum values ​​of the net load curve, the peak and valley membership degrees corresponding to each time period are calculated according to the preset membership function. For any given time period, if the membership degree of the peak segment corresponding to that time period is greater than that of the valley segment, then that time period belongs to the peak segment. If the membership degree of the peak segment corresponding to this time period is equal to the membership degree of the valley segment, then this time period belongs to the flat segment; If the membership degree of the peak segment corresponding to a given time period is less than that of the valley segment, then that time period belongs to the valley segment.

[0012] In one possible implementation, the electricity price incentive strategy is input into the lower-level model, aiming to minimize the user's energy costs, to obtain the optimal control and operation strategy for the virtual power plant, including: The electricity price incentive strategy is input into the lower-level model. With the goal of minimizing the user's energy cost, the lower-level model is solved under constraints to obtain the optimal control and operation strategy of the virtual power plant. The constraints include energy storage system constraints, electric vehicle constraints, and transferable load constraints.

[0013] Secondly, embodiments of the present invention provide a real-time control and operation device for a dual-side virtual power plant, comprising: The data acquisition module is used to collect the distribution network parameters, real-time operating parameters, and source-load data of the distributed devices connected to the target distribution network. The determination module is used to generate uncertain scenario data based on real-time operating parameters and source load data under the constraints of distribution network parameters; The solution module is used to calculate the conditional value of risk based on uncertain scenario data; and input the conditional value of risk into the upper-level model to obtain the electricity price incentive strategy with the goal of maximizing the Sharpe ratio; wherein, the conditional value of risk is used to characterize the loss under extreme scenarios; The solution module is also used to input the electricity price incentive strategy into the lower-level model, with the goal of minimizing the user's energy cost, to obtain the optimal control and operation strategy of the virtual power plant.

[0014] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation thereof.

[0015] In this embodiment of the invention, real-time operating parameters and source-load data are used to generate uncertain scenario data. Based on this data, conditional value of risk is calculated to quantify the impact of uncertain factors on incentive policies. This avoids operational mishaps caused by neglecting extreme risks in incentive strategies, making the virtual power plant more robust in complex scenarios. Furthermore, this embodiment combines upper and lower layer models to determine the virtual power plant's control and operation strategy. Specifically, in the upper layer model, the conditional value of risk determined based on the uncertain scenario data is used as input, and the output price incentive strategy aims to maximize the Sharpe ratio, ensuring both the economic benefits of the virtual power plant and the operational safety of the power grid. In the lower layer model, the control and operation scheme is generated by combining the price incentive signal with the goal of minimizing user energy costs. This leverages the proactive response of distributed resources and adapts to the safety requirements of the distribution network, ultimately achieving efficient synergy between the virtual power plant's control capabilities and the safe operation of the distribution network. Attached Figure Description

[0016] Figure 1 This is an application scenario diagram of the source-load dual-side virtual power plant real-time control and operation method provided in the embodiments of the present invention; Figure 2 This is a flowchart illustrating the implementation of the real-time control and operation method for a virtual power plant on both the source and load sides provided in this embodiment of the invention. Figure 3 This is a schematic diagram of the dynamic time-of-use electricity pricing time period division and pricing provided in an embodiment of the present invention; Figure 4 This is a flowchart of the two-layer optimized architecture of the real-time control and operation method of the source-load dual-side virtual power plant provided in the embodiments of the present invention; Figure 5 This is a schematic diagram illustrating the results of virtual power plant operation and control provided in this embodiment of the invention. Figure 6 This is a schematic diagram of the structure of the source-load dual-side virtual power plant real-time control and operation device provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0017] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0018] Figure 1 This diagram illustrates an application scenario of the real-time control and operation method for a dual-side virtual power plant provided in this invention. Figure 1As shown, the virtual power plant acts as an aggregation platform, integrating multiple distributed devices, including photovoltaic power generation systems, energy storage devices, and adjustable loads. The virtual power plant interacts with the distribution network dispatch center through a central control platform, receiving real-time electricity price signals or incentive commands from the distribution network and issuing optimized dispatch strategies to lower-level distributed resource units. Through this collaborative mechanism, unified control of distributed resources is achieved, assisting the distribution network in peak shaving and valley filling, improving asset utilization efficiency, and responding to the dynamic incentive needs of the virtual power plant.

[0019] Based on this application scenario, this invention provides a method for real-time control and operation of a virtual power plant on both the source and load sides. The flowchart of this method is as follows: Figure 2 As shown, the method may include: Step 110: Collect the distribution network parameters, real-time operating parameters, and source load data of the distributed devices connected to the target distribution network.

[0020] In this embodiment, the distribution network parameters of the target distribution network may include transformer rated capacity, line capacity limit, and expected minimum load rate; real-time operating parameters may include load data; and the source-load data of distributed devices connected to the target distribution network may include the charging and discharging data of distributed devices, the power generation data of new energy devices, and load data.

[0021] Among them, the charging and discharging data of distributed devices may include energy storage system capacity, charging and discharging efficiency, total charging demand of electric vehicles, and upper and lower limits of charging and discharging power.

[0022] The power generation data of new energy equipment can include the predicted output curves of photovoltaic and wind power.

[0023] Load data may include the base load curve and the upper limit of the transferable load.

[0024] In addition, market data and system operating parameters can be obtained. Market data may include time-of-use pricing, upper and lower limits of incentive pricing, and grid purchase pricing. System operating parameters may include data such as time period length, scheduling cycle, and confidence level.

[0025] The above data can be obtained in real time or offline through the data interface. After obtaining the above data, relevant system state variables can be initialized, such as energy storage state of charge, electric vehicle charging state, load transfer amount, etc.

[0026] Step 120: Under the constraints of distribution network parameters, generate uncertainty scenario data based on real-time operating parameters and source load data.

[0027] In this embodiment, a random scene generation method can be used to characterize the uncertainty of the source load data in order to obtain relevant data corresponding to multiple uncertain scenes.

[0028] For example, random fluctuation scenarios for photovoltaic and wind power output can be generated based on historical data or probability distributions; simultaneously, considering load forecasting errors, random disturbances are introduced to generate load fluctuation scenarios; and Latin hypercube sampling or Monte Carlo methods are used to generate these scenarios. Each scene Assigning probability of occurrence The scene set is represented as:

[0029] in Scenes Next period The system displays the output and load values ​​of photovoltaic and wind power. By constructing multiple scenarios, this virtual power plant can comprehensively consider various uncertainties during optimization, thereby improving the robustness and adaptability of the strategy.

[0030] Alternatively, based on real-time operating parameters and source load data, while ensuring the safe operation of the distribution network, multiple uncertainty scenarios can be generated using clustering methods with distribution network parameters as constraints, and the relevant data within each uncertainty scenario can be used as uncertainty scenario data.

[0031] Step 130: Calculate the conditional value of risk based on the uncertainty scenario data; and input the conditional value of risk into the upper-level model to obtain the electricity price incentive strategy with the goal of maximizing the Sharpe ratio; wherein, the conditional value of risk is used to characterize the loss under extreme scenarios.

[0032] When solving the upper-level model, conditional value at risk and time-of-use electricity price can be used as inputs, market information as constraints, and Sharpe ratio as the objective to solve the upper-level model and obtain the electricity price incentive strategy.

[0033] The upper-level objective function, which aims to maximize the Sharpe ratio, can be expressed as:

[0034] In the formula, The expected total profit of the virtual power plant; This is the conditional risk value.

[0035] Expected total profit Calculated using the following formula:

[0036] In the formula, For the scene s The probability, For virtual power plants in time periods t Electricity sales price, For electricity sales capacity, For electricity purchase price, For the amount of electricity purchased.

[0037] Accordingly, time-of-use electricity pricing can be obtained in the following ways: Determine the net load curve and divide the net load curve into peak and valley periods.

[0038] Time-of-use electricity pricing is determined based on peak and off-peak periods.

[0039] In this embodiment, the net load curve is determined based on real-time load data and renewable energy output data. First, the net load value for each time period is determined, and then the net load curve is determined based on these net load values. Accordingly, the net load value for each time period is calculated using the following formula:

[0040] in, For time period t Base load power, , These are the predicted power outputs of photovoltaic and wind power, respectively.

[0041] The net load curve is divided into peak and valley periods, and time-of-use electricity prices are determined based on these periods. Figure 3 This is a diagram illustrating the dynamic time-of-use pricing system provided in this embodiment of the invention. It shows the different time periods and their corresponding price levels, demonstrating the dynamic matching relationship between price signals and the net load curve.

[0042] Step 140: Input the electricity price incentive strategy into the lower-level model, and obtain the optimal control and operation strategy of the virtual power plant with the goal of minimizing the user's energy cost.

[0043] In this embodiment, the lower-level objective function, which aims to minimize the user's energy cost, can be expressed as:

[0044] In the formula, Total energy cost for the user; for t Electricity price during the specified time period; , These are the charging and discharging power of the energy storage, respectively. , These are the charging and discharging power of electric vehicles, respectively. for t Incentive electricity prices for specific time periods; The amount of transferable load; The unit cycle loss cost of energy storage; This is the penalty coefficient; This represents the minimum load rate of the transformer. This refers to the rated capacity of the transformer. This represents the power flowing through the transformer.

[0045] To obtain the optimal control and operation strategy for the virtual power plant, the electricity price incentive strategy output by the upper-level model is input into the lower-level model. With the goal of minimizing the user's energy cost, the lower-level model is solved under constraints to obtain the optimal control and operation strategy for the virtual power plant.

[0046] In this embodiment, the constraints include energy storage system constraints, electric vehicle constraints, and transferable load constraints. Furthermore, safety verification conditions must be considered during the solution process. Specifically, the energy storage system must satisfy the mutual exclusion constraint of charging and discharging states, meaning that charging and discharging cannot occur simultaneously within the same time period. Its charging and discharging power must be limited to the rated range, and its state of charge must follow a dynamic update equation and always remain within a safe range. The constraints for electric vehicles are similar to those for energy storage, but additionally, they must satisfy the total charging demand constraint, meaning that the total charging amount within the scheduling cycle must reach the user's preset value. Transferable loads must satisfy time-period characteristic constraints, meaning that loads can only be transferred within allowed time periods, and the total transfer amount must remain constant, while the transfer amount in a single time period must not exceed the upper limit. The system must satisfy power balance constraints, meaning that all resource outputs and loads must be matched in real time. Transformer safety constraints include that the absolute value of the load must not exceed its rated capacity, and a penalty term guides the actual load rate to not be lower than the expected minimum value to ensure the utilization rate of distribution network assets.

[0047] During the solution process, the objective function and constraints are integrated into a standard mathematical optimization problem, and a commercial or open-source optimization solver is invoked for efficient solution. The solver uses algorithms such as branch and bound and cutting planes to search for the optimal combination of decision variables that minimizes energy consumption costs, including the charging and discharging power of energy storage and electric vehicles per time period, and the transfer amount of transferable loads, while satisfying all operational and safety constraints. The final output is a refined scheduling scheme for user-side resources throughout the entire scheduling cycle.

[0048] In summary, this invention generates uncertain scenario data from real-time operating parameters and source-load data. Based on this data, it calculates the conditional value of risk, quantifies the impact of uncertain factors on incentive policies, and avoids scheduling mishaps caused by neglecting extreme risks in incentive strategies, making the virtual power plant more robust in complex scenarios. Furthermore, this invention combines upper and lower layer models to determine the virtual power plant's scheduling strategy. Specifically, in the upper layer model, the conditional value of risk determined based on the uncertain scenario data is used as input, and the output price incentive strategy aims to maximize the Sharpe ratio, ensuring both the economic benefits of the virtual power plant and the operational safety of the power grid. In the lower layer model, the goal is to minimize user energy costs, and a control and operation scheme is generated by combining the price incentive signal. This leverages the proactive response of distributed resources and adapts to the safety requirements of the distribution network, ultimately achieving efficient synergy between the virtual power plant's control capabilities and the safe operation of the distribution network.

[0049] Figure 4 This is a flowchart of the two-layer optimized architecture of the source-load dual-side virtual power plant real-time control and operation method provided in the embodiments of the present invention. The following is based on... Figure 4 The method provided in the embodiments of the present invention will be described.

[0050] The method provided in the embodiments of the present invention is based on Figure 4 The proposed two-layer optimization architecture focuses on balancing the risks and benefits of the virtual power plant at the upper layer, outputting precise electricity price incentive signals. The lower layer focuses on optimizing user energy costs, generating dispatch schemes adapted to distribution network security, and uploading these schemes to the upper-layer virtual power plant for closed-loop feedback optimization. Ultimately, this achieves the technical benefits of robust virtual power plant operation, efficient utilization of distribution network assets, and reduced user costs.

[0051] Specifically, the process begins by collecting data such as distribution network parameters, real-time operating parameters, source-load data, and market information. These data are then used to generate uncertainty scenarios, quantify the risks of these scenarios, and determine the conditional risk value.

[0052] Then, conditional value at risk and dynamic time-of-use electricity price are input into the upper-level model. Under market information constraints, the optimal electricity price incentive strategy is solved through optimization algorithms. The electricity price incentive strategy includes time-of-use electricity price for electricity sales, incentive price, and purchase price for electricity.

[0053] The lower-level model receives the electricity price incentive strategy output by the three-level model, aiming to minimize user energy costs. It clarifies the operating rules for energy storage systems, electric vehicles, and transferable loads, and incorporates distribution network security constraints for solution, resulting in a refined scheduling scheme for each distributed resource. Among these, the comprehensive electricity cost, incentive revenue, equipment losses, and transformer low-load penalties are determined.

[0054] The resulting optimal control and operation strategy for the virtual power plant not only meets the user's need for minimum cost but also aligns with the goals of power grid safety and efficient asset utilization.

[0055] Finally, the control and operation strategy output by the lower-level model is fed back to the upper-level model. If the risk quantification of a certain scenario is insufficient during the control and operation, such as the risk of overload still existing in extreme scenarios, the scenario generation and conditional risk value calculation can be iterated again, and the electricity price incentive strategy can be adjusted to form a closed-loop mechanism of "generation-optimization-scheduling-feedback".

[0056] Figure 5 This is a schematic diagram of the results of the virtual power plant implementation and control provided in the embodiment of the present invention; it shows the synergistic effect of the charging and discharging behavior of resources such as energy storage and electric vehicles with the net load of the system under the dynamic electricity price incentive strategy, and verifies the effectiveness of the proposed method in improving asset utilization and reducing system costs.

[0057] In an optional embodiment, determining the net load curve and dividing the net load curve into peak and trough periods may include: Based on real-time operating parameters and source load data, the net load curve is generated, and the maximum and minimum values ​​of the net load curve are determined.

[0058] Membership is determined based on the maximum and minimum values ​​of the net load curve to divide the net load curve into peak and valley periods.

[0059] In this embodiment, the maximum and minimum values ​​of the net load curve are extracted using the following formula: .

[0060] Then, fuzzy mathematics is used to quantify the degree to which the load state of each time period belongs to the peak or valley segment through the membership function. That is, based on the maximum and minimum values ​​of the net load curve and the net load value of each time period, the peak segment membership degree and valley segment membership degree corresponding to each time period are calculated.

[0061] The membership degree of a peak segment is calculated using the following formula:

[0062] Valley segment membership degree is calculated using the following formula:

[0063] By comparing the membership degrees of peak segments and valley segments obtained in each time period, the corresponding peak and valley time periods are determined.

[0064] Specifically, for any given time period, if the membership degree of the peak segment corresponding to that time period is greater than that of the valley segment, then that time period belongs to the peak segment.

[0065] If the membership degree of the peak segment corresponding to this time period is equal to that of the valley segment, then this time period belongs to the flat segment.

[0066] If the membership degree of the peak segment corresponding to a given time period is less than that of the valley segment, then that time period belongs to the valley segment.

[0067] Then, time-of-use electricity prices are determined based on the corresponding time periods.

[0068] Conditional Value at Risk (VaR) can be determined in a variety of ways, and the determination process is illustrated below through some optional examples.

[0069] In an optional embodiment, the uncertainty scenario data includes the probability of the uncertainty scenario occurring and the revenue of the virtual power plant under the uncertainty scenario. Step 130 calculates the conditional value of risk based on the uncertainty scenario data, including: Based on the probability of occurrence of uncertain scenarios in each uncertain scenario data and the revenue of virtual power plants under uncertain scenarios, uncertain scenarios are filtered to obtain target uncertain scenarios; among them, the probability of occurrence of uncertain scenarios in target uncertain scenarios is greater than a preset probability threshold, or the revenue of virtual power plants under uncertain scenarios is less than a preset revenue threshold.

[0070] The conditional value of risk is obtained by averaging the returns in scenarios where the objective is uncertain.

[0071] In this embodiment, the probability of an uncertain scenario occurring is greater than a preset probability threshold, or the revenue of the virtual power plant under an uncertain scenario is less than a preset revenue threshold, which can be used as the target uncertain scenario.

[0072] Alternatively, the probabilities of occurrence for each uncertain scenario can be sorted in descending order; then, the target uncertain scenario can be selected based on the confidence level.

[0073] For scenarios where the objective is uncertain, the conditional value of risk can be obtained using the following formula:

[0074] In the formula, For the value at risk, For the scene The resulting profits.

[0075] In an optional embodiment, the uncertainty scenario data includes the probability of the uncertainty scenario occurring and the revenue of the virtual power plant under the uncertainty scenario. Step 130 calculates the conditional value of risk based on the uncertainty scenario data, including: Based on the probability of occurrence of uncertain scenarios in each uncertain scenario data and the revenue of virtual power plants under uncertain scenarios, uncertain scenarios are filtered to obtain target uncertain scenarios; among them, the probability of occurrence of uncertain scenarios in target uncertain scenarios is greater than a preset probability threshold, or the revenue of virtual power plants under uncertain scenarios is less than a preset revenue threshold.

[0076] The economic risk value is obtained by averaging the returns of scenarios with uncertain objectives.

[0077] For each uncertain scenario, the real-time operating parameters and the source-load data of the distributed devices connected to the target distribution network are used to calculate the real-time power flowing through the converter and the line transmission power.

[0078] Uncertain scenarios where the real-time power exceeds the converter's rated capacity or the line transmission power is less than the expected minimum load rate are considered high-risk scenarios.

[0079] Calculate the overload loss for each high-risk scenario based on real-time power and rated capacity.

[0080] Calculate the low load loss for each high-risk scenario based on the expected minimum value of line transmission power and load rate; The average of the sum of overload loss and low load loss under each high-risk scenario is taken as the asset risk value under the high-risk scenario.

[0081] The sum of the average return in scenarios with uncertain objectives and the asset risk value in high-risk scenarios is used as the conditional risk value.

[0082] In this embodiment, the determination process for the uncertain target scenario and the economic risk value can refer to the above-mentioned related embodiments.

[0083] High-risk scenarios are those that pose safety hazards or are underutilized to distribution network assets, selected from all uncertain scenarios. There are two selection criteria: first, the real-time power flowing through the converter exceeds the converter's rated capacity; second, the line transmission power is lower than the minimum expected load rate set by the distribution network. These scenarios may lead to equipment damage or asset idleness.

[0084] Overload loss is the potential loss caused by the overload operation of the converter in high-risk scenarios. The calculation logic is to first determine the power of the overloaded part, that is, the difference between the real-time power flowing through the converter and the rated capacity of the converter, and then combine the unit overload loss coefficient and the duration of the scenario to calculate the economic losses caused by increased equipment wear and shortened life.

[0085] Low load loss is the loss caused by the line transmission power not reaching the expected minimum load rate in high-risk scenarios. The calculation logic is to first determine the power gap, that is, the difference between the actual power corresponding to the expected minimum load rate and the line transmission power, and then combine the unit idle loss coefficient and the duration of the scenario to calculate the economic loss caused by asset idleness and reduced return on investment.

[0086] Value at Risk (VaR) is a value obtained by averaging the losses of all high-risk scenarios. The calculation first adds the overload loss and low load loss of each high-risk scenario, and then takes the average of the total loss of all high-risk scenarios. It essentially reflects the average risk loss of distribution network assets caused by overload or idleness.

[0087] Conditional Value at Risk (VaR) is the ultimate indicator for comprehensively quantifying the losses of virtual power plants in extreme scenarios. The calculation logic is to add the average return corresponding to the uncertain target scenario, i.e., the economic risk value, to the asset risk value under the high-risk scenario. It covers both economic risk losses and risk losses at the level of asset safety and utilization efficiency, thus comprehensively representing the overall risk under extreme scenarios.

[0088] In an optional embodiment, the uncertainty scenario data includes the probability of the uncertainty scenario occurring and the revenue of the virtual power plant under the uncertainty scenario. Step 130 calculates the conditional value of risk based on the uncertainty scenario data, including: For each uncertain scenario, the fluctuation contribution of each distributed device is calculated based on its corresponding historical and real-time operating parameters.

[0089] The sum of the fluctuation contribution of each distributed device and the product of the source load data of each distributed device is used as the revenue adjustment value.

[0090] The difference between the revenue of the virtual power plant under uncertain scenarios and the revenue adjustment value is used as the target revenue of the virtual power plant under each uncertain scenario.

[0091] Based on the probability of occurrence of uncertain scenarios in each uncertain scenario data and the target revenue of the virtual power plant under uncertain scenarios, uncertain scenarios are filtered to obtain target uncertain scenarios; among them, the probability of occurrence of uncertain scenarios in target uncertain scenarios is greater than a preset probability threshold, or the target revenue of the virtual power plant under uncertain scenarios is less than a preset revenue threshold.

[0092] The conditional value of risk is obtained by averaging the target returns in scenarios where the target is uncertain.

[0093] In this embodiment, the fluctuation contribution of distributed devices refers to the degree of impact of the operational fluctuation of each distributed device on the overall revenue fluctuation of the virtual power plant. The calculation combines the historical and real-time operating parameters of the device to analyze the correlation between the fluctuation of its output or operating status and the system revenue. For example, if the output of a certain type of new energy equipment fluctuates frequently, its fluctuation contribution will be relatively high, and vice versa.

[0094] The revenue adjustment value is a quantitative value used to adjust the revenue of the virtual power plant scenario. The calculation logic is to multiply the fluctuation contribution of each distributed device by the operating data of that device, and then add all the products to get the sum. Its core function is to eliminate or correct the interference of device fluctuations on revenue, so that the revenue is more in line with the actual risk situation.

[0095] The target revenue is the revenue of the virtual power plant scenario after equipment fluctuation correction. It is obtained by subtracting the revenue correction value from the original revenue of the virtual power plant under uncertain scenarios. It can more accurately reflect the actual revenue level that the scenario can bring after deducting the impact of equipment fluctuation, and provide a more reliable basis for subsequent screening of target uncertain scenarios.

[0096] Accordingly, the conditional value at risk is determined as follows: First, for each uncertain scenario, the volatility contribution of each distributed device is calculated by combining historical and real-time operating parameters; then, the return correction value is obtained by summing the products of the volatility contribution value and the device operating data, and the target return is obtained by subtracting the correction value from the original scenario return; next, the target uncertain scenario is selected; finally, the average of the target returns of the target uncertain scenario is taken to obtain the conditional value at risk.

[0097] In summary, this invention generates uncertain scenario data from real-time operating parameters and source-load data. Based on this data, it calculates the conditional value of risk, quantifies the impact of uncertainties on incentive policies, and avoids scheduling mishaps caused by neglecting extreme risks in incentive strategies, making the virtual power plant more robust in complex scenarios. Furthermore, this invention combines upper and lower layer models to determine the virtual power plant's scheduling strategy. Specifically, in the upper layer model, the conditional value of risk determined based on the uncertain scenario data is used as input, and the output price incentive strategy aims to maximize the Sharpe ratio, ensuring both the economic benefits of the virtual power plant and the operational safety of the power grid. In the lower layer model, the scheduling scheme is generated by combining the price incentive signal with the goal of minimizing user energy costs. This leverages the proactive response of distributed resources and adapts to the safety requirements of the distribution network, ultimately achieving efficient synergy between the virtual power plant's control capabilities and the safe operation of the distribution network.

[0098] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0099] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0100] Figure 6 A schematic diagram of the real-time control and operation device for a dual-side virtual power plant provided in an embodiment of the present invention is shown. For ease of explanation, only the parts relevant to the embodiment of the present invention are shown, and are described in detail below: like Figure 6 As shown, the source-load dual-side virtual power plant real-time control and operation device 6 includes: The acquisition module 61 is used to acquire the distribution network parameters, real-time operating parameters, and source load data of the distributed devices connected to the target distribution network. The determination module 62 is used to generate uncertain scenario data based on real-time operating parameters and source load data under the constraints of distribution network parameters; The solution module 63 is used to calculate the conditional value of risk based on uncertain scenario data; and input the conditional value of risk into the upper-level model to obtain the electricity price incentive strategy with the goal of maximizing the Sharpe ratio; wherein, the conditional value of risk is used to characterize the loss under extreme scenarios; The solution module 63 is also used to input the electricity price incentive strategy into the lower-level model, with the goal of minimizing the user's energy cost, to obtain the optimal control and operation strategy of the virtual power plant.

[0101] In one possible implementation, the uncertainty scenario data includes the probability of the uncertainty scenario occurring and the revenue of the virtual power plant under the uncertainty scenario. Solving module 63 is specifically used for: Based on the probability of occurrence of uncertain scenarios in each uncertain scenario data and the revenue of virtual power plants under uncertain scenarios, uncertain scenarios are filtered to obtain target uncertain scenarios; among them, the probability of occurrence of uncertain scenarios in target uncertain scenarios is greater than a preset probability threshold, or the revenue of virtual power plants under uncertain scenarios is less than a preset revenue threshold. The conditional value of risk is obtained by averaging the returns in scenarios where the objective is uncertain.

[0102] In one possible implementation, the uncertainty scenario data includes the probability of the uncertainty scenario occurring and the revenue of the virtual power plant under the uncertainty scenario. Solving module 63 is specifically used for: Based on the probability of occurrence of uncertain scenarios in each uncertain scenario data and the revenue of virtual power plants under uncertain scenarios, uncertain scenarios are filtered to obtain target uncertain scenarios; among them, the probability of occurrence of uncertain scenarios in target uncertain scenarios is greater than a preset probability threshold, or the revenue of virtual power plants under uncertain scenarios is less than a preset revenue threshold. The economic risk value is obtained by averaging the returns of scenarios with uncertain objectives. For each uncertain scenario, the real-time operating parameters and the source-load data of the distributed devices connected to the target distribution network are used to calculate the real-time power flowing through the converter and the line transmission power. Uncertain scenarios where the real-time power exceeds the rated capacity of the converter or the line transmission power is less than the expected minimum value of the load rate are considered high-risk scenarios. Calculate the overload loss for each high-risk scenario based on real-time power and rated capacity; Calculate the low load loss for each high-risk scenario based on the expected minimum value of line transmission power and load rate; The average of the sum of overload loss and low load loss under each high-risk scenario is taken as the asset risk value under the high-risk scenario. The sum of the average return in scenarios with uncertain objectives and the asset risk value in high-risk scenarios is used as the conditional risk value.

[0103] In one possible implementation, the uncertainty scenario data includes the probability of the uncertainty scenario occurring and the revenue of the virtual power plant under the uncertainty scenario. Solving module 63 is specifically used for: For each uncertain scenario, the fluctuation contribution of each distributed device is calculated based on its corresponding historical and real-time operating parameters. The sum of the fluctuation contribution of each distributed device and the product of the source load data of each distributed device is used as the revenue adjustment value. The difference between the revenue of the virtual power plant under uncertain scenarios and the revenue adjustment value is used as the target revenue of the virtual power plant under each uncertain scenario. Based on the probability of occurrence of uncertain scenarios in each uncertain scenario data and the target revenue of the virtual power plant under uncertain scenarios, uncertain scenarios are filtered to obtain target uncertain scenarios; among them, the probability of occurrence of uncertain scenarios in target uncertain scenarios is greater than a preset probability threshold, or the target revenue of the virtual power plant under uncertain scenarios is less than a preset revenue threshold. The conditional value of risk is obtained by averaging the target returns in scenarios where the target is uncertain.

[0104] In one possible implementation, the Sharpe ratio is determined based on the ratio of the expected total profit of the virtual power plant to the conditional risk value; the determination module 62 is also used for: Determine the net load curve and divide the net load curve into peak and valley periods; Time-of-use pricing is determined based on peak and off-peak periods; Accordingly, the conditional value at risk is input into the upper-level model, with the goal of maximizing the Sharpe ratio, to obtain the electricity price incentive strategy, including: By inputting conditional value at risk and time-of-use pricing into the upper-level model, and under the market information constraints of the virtual power plant, the upper-level model is solved with the goal of maximizing the Sharpe ratio, thus obtaining the electricity price incentive strategy.

[0105] In one possible implementation, module 62 is specifically used for: Based on real-time operating parameters and source load data, determine the net load curve for each time period, and determine the maximum and minimum values ​​of the net load curve; Membership is determined based on the maximum and minimum values ​​of the net load curve to divide the net load curve into peak and valley periods.

[0106] In one possible implementation, module 62 is specifically used for: For the maximum and minimum values ​​of the net load curve, the peak and valley membership degrees corresponding to each time period are calculated according to the preset membership function. For any given time period, if the membership degree of the peak segment corresponding to that time period is greater than that of the valley segment, then that time period belongs to the peak segment. If the membership degree of the peak segment corresponding to this time period is equal to the membership degree of the valley segment, then this time period belongs to the flat segment; If the membership degree of the peak segment corresponding to a given time period is less than that of the valley segment, then that time period belongs to the valley segment.

[0107] In one possible implementation, the solver module 63 is specifically used for: The electricity price incentive strategy is input into the lower-level model. With the goal of minimizing the user's energy cost, the lower-level model is solved under constraints to obtain the optimal control and operation strategy of the virtual power plant. The constraints include energy storage system constraints, electric vehicle constraints, and transferable load constraints.

[0108] Figure 7 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. For example... Figure 7 As shown, the electronic device 7 of this embodiment includes a processor 70 and a memory 71. The memory 71 stores a computer program 72. When the processor 70 executes the computer program 72, it implements the steps in the various method embodiments described above. Alternatively, when the processor 70 executes the computer program 72, it implements the functions of each module / unit in the various device embodiments described above.

[0109] For example, computer program 72 may be divided into one or more modules / units, which are stored in memory 71 and executed by processor 70 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 72 in electronic device 7.

[0110] Electronic device 7 may include, but is not limited to, processor 70 and memory 71. Those skilled in the art will understand that... Figure 7 This is merely an example of electronic device 7 and does not constitute a limitation on electronic device 7. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 7 may also include input / output devices, network access devices, buses, etc.

[0111] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.

[0112] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0113] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for real-time control and operation of a virtual power plant on both the source and load sides, characterized in that, include: Collect distribution network parameters, real-time operating parameters, and source-load data of distributed devices connected to the target distribution network; Under the constraints of the distribution network parameters, uncertain scenario data is generated based on the real-time operating parameters and the source load data; Calculate the conditional value of risk based on the aforementioned uncertainty scenario data; The conditional risk value is then input into the upper-level model to obtain the electricity price incentive strategy with the goal of maximizing the Sharpe ratio; wherein, the conditional risk value is used to characterize the loss under extreme scenarios; The electricity price incentive strategy is input into the lower-level model, and the optimal control and operation strategy of the virtual power plant is obtained with the goal of minimizing the user's energy cost.

2. The method for real-time control and operation of a source-load dual-side virtual power plant according to claim 1, characterized in that, The uncertainty scenario data includes the probability of the uncertainty scenario occurring and the revenue of the virtual power plant under the uncertainty scenario. The calculation of conditional value of risk based on the uncertainty scenario data includes: Based on the probability of occurrence of uncertain scenarios in each uncertain scenario data and the revenue of virtual power plants under uncertain scenarios, uncertain scenarios are filtered to obtain target uncertain scenarios; wherein, the probability of occurrence of uncertain scenarios in the target uncertain scenarios is greater than a preset probability threshold, or the revenue of virtual power plants under uncertain scenarios is less than a preset revenue threshold. The average return of the target uncertain scenario is used to obtain the conditional value of risk.

3. The method for real-time control and operation of a source-load dual-side virtual power plant according to claim 1, characterized in that, The uncertainty scenario data includes the probability of the uncertainty scenario occurring and the revenue of the virtual power plant under the uncertainty scenario. The calculation of conditional value of risk based on the uncertainty scenario data includes: Based on the probability of occurrence of uncertain scenarios in each uncertain scenario data and the revenue of virtual power plants under uncertain scenarios, uncertain scenarios are filtered to obtain target uncertain scenarios; wherein, the probability of occurrence of uncertain scenarios in the target uncertain scenarios is greater than a preset probability threshold, or the revenue of virtual power plants under uncertain scenarios is less than a preset revenue threshold. The economic risk value is obtained by averaging the returns of the aforementioned uncertain scenario. For each uncertain scenario, the real-time operating parameters and the source-load data of the distributed devices connected to the target distribution network are used to calculate the real-time power flowing through the converter and the line transmission power. Uncertain scenarios where the real-time power is greater than the rated capacity of the converter, or where the line transmission power is less than the expected minimum load rate, are considered high-risk scenarios. Calculate the overload loss for each high-risk scenario based on the real-time power and the rated capacity; Calculate the low load loss for each high-risk scenario based on the line transmission power and the expected minimum load rate. The average of the sum of the overload loss and the low load loss in each high-risk scenario is taken as the asset risk value in the high-risk scenario. The sum of the average return corresponding to the uncertain target scenario and the asset risk value under the high-risk scenario is taken as the conditional risk value.

4. The method for real-time control and operation of a source-load dual-side virtual power plant according to claim 1, characterized in that, The uncertainty scenario data includes the probability of the uncertainty scenario occurring and the revenue of the virtual power plant under the uncertainty scenario. The calculation of conditional value of risk based on the uncertainty scenario data includes: For each uncertain scenario, the fluctuation contribution of each distributed device is calculated based on its corresponding historical and real-time operating parameters. The sum of the fluctuation contribution of each distributed device and the product of the source load data of each distributed device is used as the revenue correction value; The difference between the revenue of the virtual power plant under uncertain scenarios and the revenue correction value is taken as the target revenue of the virtual power plant under each uncertain scenario. Based on the probability of occurrence of uncertain scenarios in each uncertain scenario data and the target revenue of the virtual power plant under uncertain scenarios, uncertain scenarios are filtered to obtain target uncertain scenarios; wherein, the probability of occurrence of uncertain scenarios in the target uncertain scenarios is greater than a preset probability threshold, or the target revenue of the virtual power plant under uncertain scenarios is less than a preset revenue threshold. The conditional value of risk is obtained by averaging the target returns in the uncertain scenario.

5. The method for real-time control and operation of a source-load dual-side virtual power plant according to claim 1, characterized in that, The Sharpe ratio is determined based on the ratio of the expected total profit of the virtual power plant to the conditional value of risk. Before inputting the conditional value at risk into the upper-level model to obtain the electricity price incentive strategy with the goal of maximizing the Sharpe ratio, the following steps are also included: Determine the net load curve and divide the net load curve into peak and valley periods; Time-of-use pricing is determined based on peak and off-peak periods; Accordingly, the step of inputting the conditional value at risk into the upper-level model to obtain the electricity price incentive strategy with the goal of maximizing the Sharpe ratio includes: The conditional value at risk and the time-of-use electricity price are input into the upper-level model. Under the market information constraints of the virtual power plant, the upper-level model is solved with the goal of maximizing the Sharpe ratio to obtain the electricity price incentive strategy.

6. The method for real-time control and operation of a source-load dual-side virtual power plant according to claim 5, characterized in that, The process of determining the net load curve and dividing the net load curve into peak and valley periods includes: Based on the real-time operating parameters and the source load data, determine the net load curve for each time period, and determine the maximum and minimum values ​​of the net load curve; Membership is determined based on the maximum and minimum values ​​of the net load curve to divide the net load curve into peak and valley periods.

7. The method for real-time control and operation of a source-load dual-side virtual power plant according to claim 6, characterized in that, The step of dividing the net load curve into peak and valley periods by membership degree division based on the maximum and minimum values ​​of the net load curve includes: For the maximum and minimum values ​​of the net load curve, the peak and valley membership degrees corresponding to each time period are calculated according to the preset membership function. For any given time period, if the membership degree of the peak segment corresponding to that time period is greater than that of the valley segment, then that time period belongs to the peak segment. If the membership degree of the peak segment corresponding to this time period is equal to the membership degree of the valley segment, then this time period belongs to the flat segment; If the membership degree of the peak segment corresponding to a given time period is less than that of the valley segment, then that time period belongs to the valley segment.

8. The method for real-time control and operation of a source-load dual-side virtual power plant according to claim 1, characterized in that, The step of inputting the electricity price incentive strategy into the lower-level model, with the goal of minimizing the user's energy costs, to obtain the optimal control and operation strategy for the virtual power plant includes: The electricity price incentive strategy is input into the lower-level model. With the goal of minimizing the user's energy cost, the lower-level model is solved under constraints to obtain the optimal control and operation strategy of the virtual power plant. The constraints include energy storage system constraints, electric vehicle constraints, and transferable load constraints.

9. A source-load dual-side virtual power plant real-time control and operation device, characterized in that, include: The acquisition module is used to acquire the distribution network parameters, real-time operating parameters, and source load data of the distributed devices connected to the target distribution network. The determination module is used to generate uncertainty scenario data based on the real-time operating parameters and the source load data, under the constraints of the distribution network parameters. The solution module is used to calculate the conditional value of risk based on the uncertainty scenario data. The conditional risk value is then input into the upper-level model to obtain the electricity price incentive strategy with the goal of maximizing the Sharpe ratio; wherein, the conditional risk value is used to characterize the loss under extreme scenarios; The solution module is also used to input the electricity price incentive strategy into the lower-level model, with the goal of minimizing the user's energy cost, to obtain the optimal control and operation strategy of the virtual power plant.

10. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 8.

Images