Virtual power plant scheduling optimization method responding to demand bidding

Abstract

The invention relates to a virtual power plant scheduling optimization method responding to demand bidding, and relates to the technical field of virtual power plants. And determining a scheduling strategy of the virtual power plant according to the demand bidding of the power grid, the predicted income determined based on the real-time operation data of the plurality of energy units and the historical market transaction data, the scheduling strategy being an operation optimization strategy for the energy units. On one hand, the scheduling strategy is determined by combining the demand bidding and the predicted income of the power grid, so that the scheduling strategy can meet the demand bidding and the predicted income; and on the other hand, the historical market transaction data is utilized to formulate the scheduling strategy of the virtual power plant, so that the strategy formulating difficulty is reduced, and the efficiency of the scheduling strategy can be improved. Therefore, according to the technical scheme, the scheduling strategy of the virtual power plant can be simply and efficiently optimized in response to demand bidding, and then the reliability and stability of operation and scheduling of the power grid are improved.

Description

Technical Field

[0001] This disclosure relates to the field of virtual power plant technology, and more specifically, to a virtual power plant scheduling optimization method that responds to demand bidding. Background Technology

[0002] A virtual power plant is a smart grid technology that aggregates user-adjustable resources from multiple points, covering a wide area, and with small individual capacities through advanced Internet of Things (IoT), 5G communication, and big data technologies. It is not a physical power plant, but rather participates in the operation and dispatch of the power grid through a distributed power management system.

[0003] In related technologies, the effects of different scheduling strategies are predicted and evaluated by simulating actual operation, and optimization algorithms are used to optimize the operation strategy of virtual power plants in order to meet the operation and scheduling needs of the power grid. Summary of the Invention

[0004] The purpose of this disclosure is to provide a virtual power plant dispatch optimization method that responds to demand bidding. This method can respond to demand bidding and optimize the dispatch strategy of virtual power plants in a simple and efficient manner, thereby improving the reliability and stability of power grid operation and dispatch.

[0005] To achieve the above objectives, this disclosure provides a virtual power plant dispatch optimization method in response to demand bidding, comprising: acquiring demand bidding data of the power grid and historical market transaction data of the virtual power plant, wherein the historical market transaction data includes historical power grid bidding data, historical virtual power plant revenue data, and historical virtual power plant dispatch strategies with corresponding relationships; acquiring real-time operating data corresponding to multiple energy units of the virtual power plant, and determining the predicted revenue of the virtual power plant based on the real-time operating data; determining a first virtual power plant dispatch strategy based on the demand bidding data, the predicted revenue data, and the historical market transaction data, wherein the first virtual power plant dispatch strategy includes an operation optimization strategy corresponding to a first energy unit, wherein the first energy unit is the energy unit involved in the historical virtual power plant dispatch strategy; and optimizing the dispatch of multiple energy units of the virtual power plant according to the first virtual power plant dispatch strategy.

[0006] Optionally, the real-time operating data includes: energy unit type, energy unit runtime, and energy unit operating power. Determining the predicted revenue of the virtual power plant based on the real-time operating data includes: identifying energy units with a preset energy unit type from among the multiple energy units, where the revenue impact value corresponding to the preset energy unit type is higher than that corresponding to other energy unit types; identifying energy units whose runtime meets a preset runtime range from among the energy units with the preset energy unit type, based on their respective energy unit runtimes; and determining the predicted revenue of the virtual power plant based on the operating power of the energy units whose runtime meets the preset runtime range.

[0007] Optionally, determining the predicted revenue of the virtual power plant based on the operating power of energy units whose operating duration meets a preset operating duration range includes: obtaining the historical load demand of the virtual power plant; determining the predicted load demand based on the historical load demand of the virtual power plant; determining the load impact weight based on the predicted load demand and the operating power of energy units whose operating duration meets a preset operating duration range, wherein if the operating power of the energy unit can meet the predicted load demand, the load impact weight is positive, and if the operating power of the energy unit cannot meet the predicted load demand, the load impact weight is negative; and determining the predicted revenue of the virtual power plant based on the operating power of energy units whose operating duration meets a preset operating duration range, the load impact weight, and a pre-trained revenue prediction model.

[0008] Optionally, determining the first virtual power plant dispatch strategy based on the demand bidding, the predicted revenue, and the historical market transaction data includes: comparing the demand bidding with the historical grid bidding in the historical market transaction data to determine a target historical grid bidding that matches the demand bidding; comparing the predicted revenue with the historical virtual power plant revenue corresponding to the target historical grid bidding to determine a target historical virtual power plant revenue that matches the predicted revenue; and determining the first virtual power plant dispatch strategy based on the historical virtual power plant dispatch strategy corresponding to the target historical virtual power plant revenue.

[0009] Optionally, the historical virtual power plant scheduling strategy includes multiple operation optimization strategies and optimization costs corresponding to the operation optimization strategies for each energy unit. Determining the first virtual power plant scheduling strategy based on the historical virtual power plant scheduling strategy corresponding to the target historical virtual power plant revenue includes: if there is only one historical virtual power plant scheduling strategy corresponding to the target historical virtual power plant revenue, determining that historical virtual power plant scheduling strategy as the first virtual power plant scheduling strategy; if there are multiple historical virtual power plant scheduling strategies corresponding to the target historical virtual power plant revenue, determining the number of energy units involved in each of the multiple historical virtual power plant scheduling strategies; determining the historical virtual power plant scheduling strategy with the fewest involved energy units from the multiple historical virtual power plant scheduling strategies; and determining the historical virtual power plant scheduling strategy with the lowest optimization cost corresponding to the operation optimization strategy from the historical virtual power plant scheduling strategy with the fewest involved energy units as the first virtual power plant scheduling strategy.

[0010] Optionally, the step of optimizing the scheduling of each energy unit of the virtual power plant according to the first virtual power plant scheduling strategy includes: determining the predicted operating data corresponding to the first energy unit based on the first virtual power plant scheduling strategy and the real-time operating data corresponding to the first energy unit; predicting the external revenue and transaction volume deviation of the virtual power plant based on the predicted operating data corresponding to the first energy unit and the real-time operating data corresponding to energy units other than the first energy unit; determining the evaluation result of the first virtual power plant scheduling strategy based on the external revenue and the transaction volume deviation, wherein the evaluation result is used to characterize whether the first virtual power plant scheduling strategy is the optimal scheduling strategy; and optimizing the scheduling of each energy unit of the virtual power plant based on the evaluation result of the first virtual power plant scheduling strategy.

[0011] Optionally, the step of optimizing the scheduling of each energy unit of the virtual power plant based on the evaluation result of the first virtual power plant scheduling strategy includes: if the evaluation result of the first virtual power plant scheduling strategy indicates that the first virtual power plant scheduling strategy is the optimal scheduling strategy, executing the first virtual power plant scheduling strategy and updating the historical market transaction data based on the first virtual power plant scheduling strategy, the demand bidding, and the predicted revenue; if the evaluation result of the first virtual power plant scheduling strategy indicates that the first virtual power plant scheduling strategy is not the optimal scheduling strategy, determining the historical virtual power plant revenue that meets the preset virtual power plant revenue conditions from the historical market transaction data, determining the second energy unit based on the energy unit involved in the historical virtual power plant scheduling strategy corresponding to the historical virtual power plant revenue that meets the preset virtual power plant revenue conditions and the first energy unit; determining the second virtual power plant scheduling strategy based on the external revenue, the transaction volume deviation, and the real-time operation data corresponding to the second energy unit, wherein the second virtual power plant scheduling strategy includes the operation optimization strategy corresponding to the second energy unit; and executing the first virtual power plant scheduling strategy and the second virtual power plant scheduling strategy to optimize the scheduling of each energy unit of the virtual power plant.

[0012] Optionally, determining the second virtual power plant scheduling strategy based on the external revenue, the transaction volume deviation, and the real-time operating data corresponding to the second energy unit includes: obtaining historical external revenue and a transaction volume deviation threshold; if the external revenue is greater than or equal to the average of the historical external revenue, and the transaction volume deviation is greater than or equal to the transaction volume deviation threshold, determining the second virtual power plant scheduling strategy for a first operating parameter in the real-time operating data corresponding to the second energy unit, wherein the first operating parameter does not affect the output power of the second energy unit; if the external revenue is less than the average of the historical external revenue, and / or the transaction volume deviation is less than the transaction volume deviation threshold, determining the second virtual power plant scheduling strategy for a second operating parameter in the real-time operating data corresponding to the second energy unit, wherein the second operating parameter affects the output power of the second energy unit.

[0013] Optionally, the virtual power plant scheduling optimization method further includes: acquiring a scheduling optimization request triggered by a virtual power plant management user, the scheduling optimization request including demand cost and demand revenue; determining an evaluation result of the scheduling optimization request based on the demand bidding, the predicted revenue, the demand cost, and the demand revenue, the evaluation result of the scheduling optimization request being used to characterize whether the scheduling optimization request is a normal scheduling optimization request; if the scheduling optimization request is a normal scheduling optimization request, optimizing the scheduling of each energy unit of the virtual power plant again based on the demand cost and demand revenue; if the scheduling optimization request is not a normal scheduling optimization request, outputting a prompt message, the prompt message being used to indicate that the scheduling optimization request is invalid.

[0014] Optionally, the virtual power plant scheduling optimization method further includes: after determining that the optimized scheduling of multiple energy units of the virtual power plant has been completed, acquiring the optimized scheduling operation data corresponding to each of the multiple energy units; determining the operation status evaluation result and performance evaluation result corresponding to each of the multiple energy units based on the optimized scheduling operation data; determining the energy unit optimization strategy based on the operation status evaluation result and the performance evaluation result; and optimizing the virtual power plant according to the energy unit optimization strategy.

[0015] By combining the aforementioned technical solution with grid demand bidding, predicted revenue determined from real-time operational data of multiple energy units, and historical market transaction data, a dispatch strategy for the virtual power plant is determined. This dispatch strategy is an operational optimization strategy for the energy unit. On the one hand, determining the dispatch strategy by combining grid demand bidding and predicted revenue ensures that the dispatch strategy meets both demand bidding and predicted revenue requirements. On the other hand, utilizing historical market transaction data to formulate the virtual power plant dispatch strategy reduces the difficulty of strategy formulation and improves the efficiency of the dispatch strategy. Therefore, this technical solution can respond to demand bidding, perform simple and efficient optimization of the virtual power plant dispatch strategy, and thus improve the reliability and stability of grid operation and dispatch.

[0016] Other features and advantages of this disclosure will be described in detail in the following detailed description section. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the present disclosure and form part of the specification. They are used together with the following detailed description to explain the present disclosure, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram illustrating an application scenario of a virtual power plant according to an exemplary embodiment.

[0018] Figure 2This is a structural block diagram of a virtual power plant according to an exemplary embodiment.

[0019] Figure 3 This is a flowchart illustrating a virtual power plant scheduling optimization method based on an exemplary embodiment of demand bidding.

[0020] Figure 4 This is a block diagram illustrating a virtual power plant scheduling optimization device that responds to demand bidding, according to an exemplary embodiment.

[0021] Figure 5 This is a block diagram illustrating an electronic device according to an exemplary embodiment. Detailed Implementation

[0022] The specific embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit this disclosure.

[0023] A virtual power plant is a smart grid technology that aggregates user-adjustable resources from multiple points, covering a wide area, and with small individual capacities through advanced Internet of Things (IoT), 5G communication, and big data technologies. It is not a physical power plant, but rather participates in the operation and dispatch of the power grid through a distributed power management system.

[0024] A virtual power plant mainly consists of three parts: a power generation system, energy storage devices, and a communication system. Distributed power management aggregates different types of distributed energy sources through technologies such as control metering and communication. The energy storage system stores and releases energy to balance grid load. The communication system ensures real-time data transmission and accurate execution of control commands.

[0025] Virtual power plants, relying on the internet and modern information and communication technologies, aggregate various resources scattered across the power grid, such as distributed power sources, energy storage, and loads, to achieve coordinated and optimized operation control and market transactions. This enables multi-energy complementarity on the power supply side and flexible interaction on the load side. They generate electricity and participate in the energy market; they also participate in ancillary services such as peak shaving and frequency regulation by adjusting power output.

[0026] In related technologies, the effects of different scheduling strategies are predicted and evaluated by simulating actual operating conditions. Optimization algorithms are then used to optimize the operating strategies of virtual power plants to meet the operation and scheduling needs of the power grid. Examples of such optimization algorithms include genetic algorithms, particle swarm optimization, simulated annealing, or hybrid optimization algorithms.

[0027] The complex optimization algorithm employed makes optimization difficult and prevents the achievement of simple and efficient optimization. Consequently, when demand bidding occurs in the power grid, it cannot quickly respond to and optimize demand.

[0028] Based on this, the present disclosure provides a technical solution that combines grid demand bidding and projected revenue to determine a dispatch strategy, enabling the dispatch strategy to meet both demand bidding and projected revenue. Furthermore, by utilizing historical market transaction data to formulate a dispatch strategy for virtual power plants, the difficulty of strategy formulation is reduced, and the efficiency of the dispatch strategy can be improved. Therefore, this technical solution can respond to demand bidding, perform simple and efficient optimization of the dispatch strategy for virtual power plants, and thereby improve the reliability and stability of grid operation and dispatch.

[0029] Figure 1 This is a schematic diagram illustrating an application scenario of a virtual power plant according to an exemplary embodiment, such as... Figure 1 As shown, this application scenario involves power grids, virtual power plants, and external objects.

[0030] Regarding the power grid, it can be understood as the electricity market side, which can issue operation or dispatch instructions for virtual power plants.

[0031] Virtual power plants can be deployed with multiple distributed energy units, which can supply electricity to the market.

[0032] External objects can be understood as other participants in the virtual power plant, such as electricity sales companies, upper-level management platforms, and external electricity users.

[0033] Between the power grid, virtual power plants, and external objects, both information flow and electrical energy flow are involved. Information flow primarily facilitates information exchange. Electrical energy flow primarily facilitates electrical energy transmission.

[0034] Figure 2 This is a structural block diagram of a virtual power plant according to an exemplary embodiment, such as... Figure 2 As shown, the virtual power plant includes multiple distributed energy units, a management system, and a communication system.

[0035] The distributed multiple energy units can be viewed as an energy storage system within the virtual power plant. These energy units include, for example, energy storage power stations and gas turbine units. The management system enables the scheduling and management of the entire virtual power plant. The communication system facilitates both external and internal communication.

[0036] It is understood that the technical solutions provided in the embodiments of this disclosure can be applied to... Figure 2 The management system in the system is used to optimize the scheduling of virtual power plants.

[0037] Figure 3 This is a flowchart illustrating a virtual power plant scheduling optimization method based on an exemplary embodiment of demand bidding, such as... Figure 3 As shown, the method includes: Step 301: Obtain historical market transaction data for grid demand bidding and virtual power plants. Historical market transaction data includes historical grid bidding, historical virtual power plant revenue, and historical virtual power plant dispatching strategies with corresponding relationships.

[0038] Step 302: Obtain real-time operating data corresponding to multiple energy units of the virtual power plant, and determine the predicted revenue of the virtual power plant based on the real-time operating data.

[0039] Step 303: Determine the dispatch strategy for the first virtual power plant based on demand bidding, projected revenue, and historical market transaction data. The dispatch strategy for the first virtual power plant includes the operation optimization strategy corresponding to the first energy unit, which is the energy unit involved in the historical virtual power plant dispatch strategies.

[0040] Step 304: Optimize the scheduling of multiple energy units of the virtual power plant according to the scheduling strategy of the first virtual power plant.

[0041] In step 301, the demand bidding of the power grid can be understood as the electricity trading price, which can determine the actual revenue of the virtual power plant.

[0042] Within a virtual power plant, historical market transaction data is maintained and continuously updated. This data includes historical grid bidding, historical virtual power plant revenue, and historical virtual power plant dispatching strategies with corresponding relationships.

[0043] Historical grid bidding can be understood as the bidding processes previously involved in virtual power plant operations. Historical virtual power plant revenue can be understood as the revenue ultimately achieved by the virtual power plant under the corresponding bidding conditions. Historical virtual power plant dispatch strategies can be understood as the virtual power plant dispatch strategies ultimately adopted to satisfy historical grid bidding requirements.

[0044] In addition, it can be understood that the revenue of the virtual power plant mentioned above can be understood as the revenue of the power grid managed by the virtual power plant.

[0045] In step 302, the real-time operating data corresponding to each of the multiple energy units can characterize the current operating status of each energy unit. Therefore, the operating data can include: operating power, operating duration, and load, etc. Based on this real-time operating data, the predicted revenue of the virtual power plant can be determined.

[0046] As an optional implementation, the real-time operating data includes: energy unit type, energy unit runtime, and energy unit operating power. Step 302 includes: determining energy units with a preset energy unit type from among the multiple energy units based on their respective energy unit types, wherein the revenue impact value corresponding to the preset energy unit type is higher than the revenue impact value corresponding to other energy unit types; determining energy units whose runtime meets a preset runtime range from among the energy units with the preset energy unit type based on their energy unit runtime; and determining the predicted revenue of the virtual power plant based on the energy unit operating power corresponding to the energy unit whose runtime meets the preset runtime range.

[0047] In this implementation, revenue forecasting is not based on all energy units, but on a subset of energy units that meet specific conditions. In this way, the revenue of the virtual power plant can be maximized as much as possible while meeting demand bidding requirements.

[0048] You can first use preset energy unit types to filter out energy units with higher return impact values. For example, preset energy units could be energy storage power stations or renewable energy units, which are energy units with high returns and low costs.

[0049] Furthermore, by using a preset runtime range, energy units with longer runtimes or those that have just started operating can be selected. For example, the preset runtime range could be either a shorter runtime range or a longer runtime range.

[0050] Furthermore, based on the above screening, the selected energy units can be obtained. The revenue of the virtual power plant is then predicted based on the operating power of these selected energy units.

[0051] In some embodiments, the predicted revenue can be determined using a pre-trained revenue prediction model. The training data for this revenue prediction model may include: energy unit operating power samples and revenue labels. The model is trained using this training data so that it can predict revenue based on operating power.

[0052] In other embodiments, big data can be used to analyze the relationship between operating power and revenue, and revenue can be predicted through quantitative analysis.

[0053] In some embodiments, if revenue is predicted directly based on operating power, the accuracy of the revenue obtained may be difficult to guarantee. Therefore, to improve the accuracy of revenue prediction, the predicted revenue of the virtual power plant is determined based on the operating power of energy units whose operating duration meets a preset operating duration range. This can include: obtaining the historical load demand of the virtual power plant; determining the predicted load demand based on the historical load demand of the virtual power plant; determining the load impact weight based on the predicted load demand and the operating power of energy units whose operating duration meets the preset operating duration range, wherein if the operating power of the energy unit can meet the predicted load demand, the load impact weight is positive, and if the operating power of the energy unit cannot meet the predicted load demand, the load impact weight is negative; and determining the predicted revenue of the virtual power plant based on the operating power of energy units whose operating duration meets the preset operating duration range, the load impact weight, and a pre-trained revenue prediction model.

[0054] The historical load demand of the virtual power plant is data that the virtual power plant has been managing, so it can be obtained directly, and this load demand is the real load demand.

[0055] Based on historical load demand, load demand can be predicted. For example, future load demand can be predicted by analyzing changes in historical load demand. For instance, if load demand gradually increases, the predicted load demand will need to be greater than the historical load demand. If load demand remains relatively constant, the predicted load demand will be roughly the same as the historical load demand.

[0056] Based on the determined predicted load demand, it can be determined whether the operating power of the screened energy units meets the predicted load demand. For the conversion of operating power into load capacity, mature technologies in this field can be referenced. Based on the judgment of whether the predicted demand is met, a load influence weight can be determined. This load influence weight can be used as the model weight of the pre-trained prediction model to make the model's prediction results more accurate.

[0057] Therefore, if the operating power of the energy unit can meet the predicted load demand, the load impact weight is positive; if the operating power of the energy unit cannot meet the predicted load demand, the load impact weight is negative.

[0058] Furthermore, the operating power and load impact weights of energy units whose operating time meets the preset operating time range are input into the pre-trained revenue prediction model to obtain the predicted revenue of the virtual power plant output by the revenue prediction model.

[0059] In step 303, the first virtual power plant scheduling strategy includes the operation optimization strategy corresponding to the first energy unit, which is the energy unit involved in the historical virtual power plant scheduling strategy. That is, the first virtual power plant scheduling strategy only targets a portion of the energy units, and these energy units are those that have previously undergone scheduling optimization. In this way, the cost of scheduling optimization can be reduced. For example, optimizing energy units that have not been optimized may bring additional impacts or costs, while optimizing energy units that have been optimized previously is more efficient and less difficult.

[0060] As an optional implementation, step 303 includes: comparing the demand bid with the historical grid bid in the historical market transaction data to determine the target historical grid bid that matches the demand bid; comparing the predicted revenue with the historical virtual power plant revenue corresponding to the target historical grid bid to determine the target historical virtual power plant revenue that matches the predicted revenue; and determining the first virtual power plant scheduling strategy based on the historical virtual power plant scheduling strategy corresponding to the target historical virtual power plant revenue.

[0061] It is understandable that, based on historical market transaction data, including historical grid bidding, historical virtual power plant revenue, and historical virtual power plant dispatching strategies with corresponding relationships, the corresponding historical virtual power plant dispatching strategy can be determined by searching based on demand bidding and predicted revenue.

[0062] Among them, matching bids and matching returns can mean that the values ​​are equal or basically the same, for example, the error can be in the range of 0 to 1.

[0063] In some embodiments, the final number of historical virtual power plant scheduling strategies may be one or more. In this case, to determine a more reasonable scheduling strategy, an optimization cost can be configured for each historical virtual power plant scheduling strategy.

[0064] Therefore, the historical virtual power plant scheduling strategy can include the operation optimization strategy and the optimization cost corresponding to the operation optimization strategy for each energy unit. The optimization cost can be the actual optimization cost incurred when executing the historical virtual power plant scheduling strategy, which is configured by the user.

[0065] Furthermore, based on the historical virtual power plant scheduling strategies corresponding to the target historical virtual power plant revenue, the first virtual power plant scheduling strategy is determined. This can include: if there is only one historical virtual power plant scheduling strategy corresponding to the target historical virtual power plant revenue, that historical virtual power plant scheduling strategy is determined as the first virtual power plant scheduling strategy; if there are multiple historical virtual power plant scheduling strategies corresponding to the target historical virtual power plant revenue, the number of energy units involved in each of the multiple historical virtual power plant scheduling strategies is determined; from the multiple historical virtual power plant scheduling strategies, the historical virtual power plant scheduling strategy involving the fewest energy units is determined; and among the historical virtual power plant scheduling strategies involving the fewest energy units, the historical virtual power plant scheduling strategy with the lowest optimization cost corresponding to the operation optimization strategy is determined as the first virtual power plant scheduling strategy.

[0066] In this implementation, if there is only one historical virtual power plant scheduling strategy, then that historical virtual power plant scheduling strategy can be directly determined as the first virtual power plant scheduling strategy.

[0067] If there are multiple historical virtual power plant scheduling strategies, first determine the number of energy units involved in each strategy, and then select the strategy with the fewest energy units. Next, among the historical virtual power plant scheduling strategies with the fewest energy units, determine the one with the lowest optimization cost corresponding to the optimization strategy as the first virtual power plant scheduling strategy.

[0068] This implementation method can significantly reduce the cost of virtual power plant scheduling optimization.

[0069] In this embodiment of the disclosure, the virtual power plant scheduling strategy includes: an operation optimization strategy corresponding to each energy unit. The operation optimization strategy can be optimization of operating time, optimization of operating power, or optimization of load, etc. For example, optimizing operating time can adjust peak-hour operation to off-peak operation; optimizing operating power can, for example, increase the operating power of energy units with high capacity; optimizing load can, for example, reduce the load of energy units with low capacity.

[0070] In step 304, multiple energy units of the virtual power plant are optimized and scheduled according to the first virtual power plant scheduling strategy.

[0071] In some embodiments, the scheduling strategy of the first virtual power plant can be executed directly. For example, the operation optimization strategy corresponding to the first energy unit can be distributed to the corresponding first energy unit so that the first energy unit can adjust its own operation mode, thereby achieving scheduling optimization.

[0072] In other embodiments, the scheduling strategy of the first virtual power plant can also be evaluated, and the scheduling can be optimized based on the evaluation results.

[0073] Therefore, as an optional implementation, step 304 includes: determining the predicted operating data corresponding to the first energy unit based on the first virtual power plant scheduling strategy and the real-time operating data corresponding to the first energy unit; predicting the external revenue and transaction volume deviation of the virtual power plant based on the predicted operating data corresponding to the first energy unit and the real-time operating data corresponding to energy units other than the first energy unit; determining the evaluation result of the first virtual power plant scheduling strategy based on the external revenue and transaction volume deviation, the evaluation result being used to characterize whether the first virtual power plant scheduling strategy is the optimal scheduling strategy; and optimizing the scheduling of each energy unit of the virtual power plant based on the evaluation result of the first virtual power plant scheduling strategy.

[0074] In this implementation, the predicted operating data corresponding to the first energy unit can be understood as the operating data of the first energy unit after scheduling optimization. Thus, based on the real-time operating data, the predicted operating data can be obtained by adjusting according to the scheduling strategy.

[0075] Regarding external revenue, it can be understood as the welfare benefits generated by the virtual power plant to external parties. This revenue may not be quantified in any specific entity, but rather is merely an evaluative data point. Regarding the deviation in transaction volume, it can be understood as the difference between the standard transaction volume and the actual transaction volume.

[0076] In some embodiments, the predicted operating data corresponding to the first energy unit and the real-time operating data corresponding to the energy units other than the first energy unit are equivalent to the operating data of multiple energy units after scheduling optimization. Based on the operating data of multiple energy units after scheduling optimization, the external revenue and transaction volume deviation of the virtual power plant can be predicted.

[0077] For example, the external revenue of a virtual power plant can be the difference between the projected revenue and a preset revenue threshold. This external revenue can be positive or negative. The preset revenue threshold can be the revenue required by the virtual power plant and can be set by the virtual power plant's manager. Therefore, based on the operational data of multiple energy units after scheduling optimization, the projected revenue can be obtained first (refer to the method for determining the projected revenue in the aforementioned embodiment), and then the external revenue can be determined based on the projected revenue and the preset revenue threshold.

[0078] For example, the deviation in the transaction volume can be the difference between the actual transaction volume of the virtual power plant and the preset transaction volume. The preset transaction volume is the transaction volume set according to the management needs of the virtual power plant, which can constrain the virtual power plant and can be set by higher-level management. Therefore, based on the operating data of multiple energy units after scheduling optimization, the actual transaction volume can be determined first. This actual transaction volume can be determined according to the load capacity that the operating data of multiple energy units after scheduling optimization can support; for example, the actual transaction volume and the load capacity have a 1:1 relationship. Then, the difference between the actual transaction volume and the preset transaction volume is determined to obtain the transaction volume deviation.

[0079] Furthermore, based on the deviation between external revenue and transaction volume, the evaluation result of the first virtual power plant dispatch strategy is determined. This evaluation result can characterize whether the first virtual power plant dispatch strategy is the optimal dispatch strategy.

[0080] In some embodiments, if the external revenue is positive and the transaction volume deviation is positive, then the first virtual power plant dispatching strategy is the optimal dispatching strategy. If the transaction volume deviation is negative, the first virtual power plant dispatching strategy is not the optimal dispatching strategy regardless of whether the external revenue is positive.

[0081] In other embodiments, external revenue can be compared with historical external revenue, and the deviation in transaction volume can be compared with historical transaction volume deviation. If the comparison results indicate that both external revenue and transaction volume deviation are improving (e.g., the deviation changes from negative to positive, external revenue increases, etc.), then the first virtual power plant scheduling strategy is the optimal scheduling strategy. Otherwise, the first virtual power plant scheduling strategy is not the optimal scheduling strategy.

[0082] Furthermore, based on the evaluation results of the first virtual power plant dispatch strategy, the optimization dispatch of each energy unit of the virtual power plant can include: if the evaluation results of the first virtual power plant dispatch strategy indicate that the first virtual power plant dispatch strategy is the optimal dispatch strategy, the first virtual power plant dispatch strategy is executed, and historical market transaction data is updated based on the first virtual power plant dispatch strategy, demand bidding, and predicted revenue; if the evaluation results of the first virtual power plant dispatch strategy indicate that the first virtual power plant dispatch strategy is not the optimal dispatch strategy, historical virtual power plant revenue that meets the preset virtual power plant revenue conditions is determined from the historical market transaction data, and the second energy unit is determined based on the energy unit involved in the historical virtual power plant dispatch strategy corresponding to the historical virtual power plant revenue that meets the preset virtual power plant revenue conditions and the first energy unit; the second virtual power plant dispatch strategy is determined based on external revenue, transaction volume deviation, and real-time operation data corresponding to the second energy unit, the second virtual power plant dispatch strategy including the operation optimization strategy corresponding to the second energy unit; the first virtual power plant dispatch strategy and the second virtual power plant dispatch strategy are executed to optimize the dispatch of each energy unit of the virtual power plant.

[0083] In this implementation, if the evaluation result of the first virtual power plant scheduling strategy indicates that the first virtual power plant scheduling strategy is the optimal scheduling strategy, the first virtual power plant scheduling strategy can be directly executed, and the historical market transaction data can be updated according to the first virtual power plant scheduling strategy, demand bidding, and predicted revenue. For example, the first virtual power plant scheduling strategy, demand bidding, and predicted revenue can be stored in the corresponding historical market transaction data.

[0084] If the evaluation results of the first virtual power plant scheduling strategy indicate that the first virtual power plant scheduling strategy is not the optimal scheduling strategy, then a new scheduling strategy can be determined.

[0085] In some embodiments, the preset virtual power plant revenue condition can be a virtual power plant revenue range, wherein the predicted revenue is within the virtual power plant revenue range.

[0086] In some embodiments, among the energy units involved in the historical virtual power plant scheduling strategy corresponding to the historical virtual power plant revenue that meets the preset virtual power plant revenue conditions, those energy units that are different in type, duration, or power from the first energy unit are identified as the second energy unit. That is, the second energy unit operates in a certain way compared to the first energy unit.

[0087] In some embodiments, determining a second virtual power plant scheduling strategy based on external revenue, transaction volume deviation, and real-time operating data corresponding to the second energy unit may include: obtaining historical external revenue and transaction volume deviation thresholds; if external revenue is greater than or equal to the average historical external revenue and the transaction volume deviation is greater than or equal to the transaction volume deviation threshold, determining a second virtual power plant scheduling strategy for a first operating parameter in the real-time operating data corresponding to the second energy unit, wherein the first operating parameter does not affect the output power of the second energy unit; if external revenue is less than the average historical external revenue and / or the transaction volume deviation is less than the transaction volume deviation threshold, determining a second virtual power plant scheduling strategy for a second operating parameter in the real-time operating data corresponding to the second energy unit, wherein the second operating parameter affects the output power of the second energy unit.

[0088] In some embodiments, the transaction volume deviation threshold may be the aforementioned preset transaction volume deviation, or it may be the minimum value among historical transaction volume deviations.

[0089] In some embodiments, the first operating parameter does not affect the output power of the second energy unit, i.e., it does not involve a change in output power. Therefore, the first operating parameter can be the operating cycle. For example, the original operation from 8:00 to 20:00 is adjusted to operation from 0:00 to 12:00. Therefore, in this case, the dispatching strategy of the second virtual power plant can be determined within an acceptable range.

[0090] For example, some permissible adjustment methods for the first operating parameter can be pre-defined. Then, from these adjustment methods, the one that matches the current first operating parameter can be selected to formulate the second virtual power plant strategy. For instance, the second virtual power plant scheduling strategy may include: adjusting the first operating parameter according to the corresponding adjustment method for one of the second energy units.

[0091] It is understandable that the adjustment methods for the first operating parameters of different second energy units may differ.

[0092] In some embodiments, the second operating parameter affects the output power of the second energy unit. Therefore, the second operating parameter may include, but is not limited to, output voltage, operating time, load, and operating power.

[0093] Since the external revenue is less than the historical average external revenue, and / or the transaction volume deviation is less than the transaction volume deviation threshold, for the second energy unit, an adjustment strategy for the second operating parameters can be formulated based on increasing the output power to obtain the second virtual power plant dispatch strategy.

[0094] For example, for some energy units in the second energy unit, increase the output voltage; for some energy units in the second energy unit, reduce the load; for some energy units in the second energy unit, increase the operating power, etc.; for some energy units in the second energy unit, reduce the operating time, etc.

[0095] It is understandable that the adjustment methods for the second operating parameters of different second energy units may differ.

[0096] Therefore, the second virtual power plant scheduling strategy can be an adjustment method for the first or second operating parameters of the second energy unit.

[0097] Furthermore, the first and second virtual power plant scheduling strategies are executed to optimize the scheduling of each energy unit of the virtual power plant. For example, the corresponding adjustment methods are distributed to the corresponding energy units so that each energy unit of the virtual power plant can be optimally scheduled according to the corresponding adjustment methods.

[0098] In some embodiments, the virtual power plant scheduling optimization method may further include: acquiring a scheduling optimization request triggered by a virtual power plant management user, the scheduling optimization request including demand cost and demand revenue; determining an evaluation result of the scheduling optimization request based on demand bidding, predicted revenue, demand cost, and demand revenue, the evaluation result of the scheduling optimization request being used to characterize whether the scheduling optimization request is a normal scheduling optimization request; if the scheduling optimization request is a normal scheduling optimization request, optimizing the scheduling of each energy unit of the virtual power plant again based on demand cost and demand revenue; if the scheduling optimization request is not a normal scheduling optimization request, outputting a prompt message, the prompt message being used to indicate that the scheduling optimization request is invalid.

[0099] In some embodiments, demand costs may include, but are not limited to: market electricity sales costs, generation costs, market electricity purchase costs, load reduction dispatch costs, and deviation penalty costs. Demand benefits may include, but are not limited to: self-interest, external object benefits, and market benefits.

[0100] In some embodiments, the predicted revenue and demand revenue can be summed, and the demand bidding and demand cost can be summed. The two sums are compared. If the sum of revenue is significantly greater than the sum of demand bidding and demand cost, then the scheduling optimization request is not a normal scheduling optimization request. Otherwise, the scheduling optimization request is a normal scheduling optimization request. Here, "significantly greater than" can mean that the sum of revenue is more than five times the sum of demand bidding and demand cost.

[0101] In some embodiments, optimizing the scheduling of each energy unit of the virtual power plant based on demand cost and demand revenue may include: adjusting the predicted revenue based on demand revenue, adjusting the demand bid based on demand cost, and performing optimized scheduling according to the implementation methods described in the foregoing embodiments based on the adjusted revenue and the adjusted bid. Specifically, the predicted revenue may be adjusted to be less than or equal to the demand revenue; and an acceptable bid is determined based on the demand cost, and the demand bid is adjusted based on the acceptable bid.

[0102] In some embodiments, the virtual power plant scheduling optimization method may further include: after determining that the optimized scheduling of multiple energy units of the virtual power plant has been completed, acquiring the optimized scheduling operation data corresponding to each of the multiple energy units; determining the operation status evaluation results and performance evaluation results corresponding to each of the multiple energy units based on the optimized scheduling operation data corresponding to each of the multiple energy units; determining the energy unit optimization strategy based on the operation status evaluation results and performance evaluation results; and optimizing the virtual power plant according to the energy unit optimization strategy.

[0103] The determination that the optimized scheduling of multiple energy units in the virtual power plant is complete can include: upon receiving the optimized scheduling completion information of the corresponding energy units, determining that the optimized scheduling of multiple energy units in the virtual power plant is complete. The optimized scheduling completion information includes the operational data of the corresponding energy units after optimized scheduling.

[0104] In some embodiments, based on the optimized scheduling and operational data corresponding to multiple energy units, the operational status and performance of the multiple energy units can be evaluated. The specific evaluation methods can refer to mature technologies in the field.

[0105] Furthermore, the operational status assessment results and performance assessment results are analyzed. If the analysis results indicate that the operation and performance of multiple energy units are normal, the energy unit optimization strategy is determined to be: not to optimize the number of energy units, but to continuously monitor the operation of the energy units. If the analysis results indicate that there are energy units with abnormal operation and performance, the energy unit optimization strategy is determined to be: to optimize the energy units with abnormalities, and to optimize the number of energy units.

[0106] Optimizing the number of energy units can include reducing the number of similar energy units and increasing the number of renewable energy units.

[0107] In some embodiments, the operating status and performance of multiple energy units can be distributed according to type. If there are energy unit data with significant differences within the same type, they are considered abnormal.

[0108] Furthermore, the energy unit optimization strategy is fed back to the virtual power plant management personnel for execution.

[0109] Figure 4 This is a block diagram illustrating a virtual power plant dispatch optimization device for responding to demand bidding, according to an exemplary embodiment. Figure 4 As shown, the device includes: The acquisition module 401 is used to acquire demand bidding data from the power grid and historical market transaction data from virtual power plants. The historical market transaction data includes historical power grid bidding data, historical virtual power plant revenue data, and historical virtual power plant dispatching strategies with corresponding relationships. It also acquires real-time operating data corresponding to multiple energy units of the virtual power plant and determines the predicted revenue of the virtual power plant based on the real-time operating data.

[0110] The scheduling module 402 is used to determine a first virtual power plant scheduling strategy based on the demand bidding, the predicted revenue, and the historical market transaction data. The first virtual power plant scheduling strategy includes an operation optimization strategy corresponding to a first energy unit, which is the energy unit involved in the historical virtual power plant scheduling strategy. Based on the first virtual power plant scheduling strategy, multiple energy units of the virtual power plant are optimized and scheduled.

[0111] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0112] Figure 5 This is a block diagram illustrating an electronic device 500 according to an exemplary embodiment. For example... Figure 5 As shown, the electronic device 500 may include a processor 501 and a memory 502. The electronic device 500 may also include one or more of a multimedia component 503, an input / output (I / O) interface 504, and a communication component 505.

[0113] The processor 501 controls the overall operation of the electronic device 500 to complete all or part of the steps in the virtual power plant scheduling optimization method for responding to demand bidding. The memory 502 stores various types of data to support the operation of the electronic device 500. This data may include, for example, instructions for any application or method operating on the electronic device 500, and application-related data such as contact data, sent and received messages, pictures, audio, video, etc. The memory 502 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. Multimedia component 503 may include a screen and an audio component. The screen may be, for example, a touchscreen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in memory 502 or transmitted via communication component 505. The audio component also includes at least one speaker for outputting audio signals. I / O interface 504 provides an interface between processor 501 and other interface modules, such as a keyboard, mouse, buttons, etc. These buttons may be virtual or physical buttons. Communication component 505 is used for wired or wireless communication between the electronic device 500 and other devices. Wireless communication may include Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination thereof; therefore, the corresponding communication component 505 may include a Wi-Fi module, a Bluetooth module, or an NFC module.

[0114] In an exemplary embodiment, the electronic device 500 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the XXXX method described above.

[0115] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the virtual power plant scheduling optimization method for responding to demand bidding described above. For example, the computer-readable storage medium may be the memory 502 including program instructions described above, which may be executed by the processor 501 of the electronic device 500 to complete the virtual power plant scheduling optimization method for responding to demand bidding described above.

[0116] In another exemplary embodiment, a computer program product is also provided, which includes a computer program executable by a processor, which, when executed by the processor, implements the virtual power plant scheduling optimization method for responding to demand bidding as described above.

[0117] The preferred embodiments of this disclosure have been described in detail above with reference to the accompanying drawings. However, this disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this disclosure, various simple modifications can be made to the technical solutions of this disclosure, and these simple modifications all fall within the protection scope of this disclosure.

[0118] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, this disclosure will not describe the various possible combinations separately.

[0119] Furthermore, various different embodiments of this disclosure can be combined in any way, as long as they do not violate the spirit of this disclosure, they should also be regarded as the content disclosed in this disclosure.

Claims

1. A virtual power plant scheduling optimization method based on demand bidding, characterized in that, include: Acquire historical market transaction data of grid demand bidding and virtual power plants, including historical grid bidding, historical virtual power plant revenue and historical virtual power plant dispatching strategies with corresponding relationships; The real-time operating data corresponding to the multiple energy units of the virtual power plant is obtained, and the predicted revenue of the virtual power plant is determined based on the real-time operating data. Based on the demand bidding, the predicted revenue, and the historical market transaction data, a first virtual power plant scheduling strategy is determined. The first virtual power plant scheduling strategy includes an operation optimization strategy corresponding to a first energy unit, and the first energy unit is the energy unit involved in the historical virtual power plant scheduling strategy. Based on the first virtual power plant scheduling strategy, multiple energy units of the virtual power plant are optimized and scheduled.

2. The virtual power plant scheduling optimization method according to claim 1, characterized in that, The real-time operating data includes: energy unit type, energy unit operating time, and energy unit operating power. Determining the predicted revenue of the virtual power plant based on the real-time operating data includes: Based on the energy unit types corresponding to the plurality of energy units, energy units with a preset energy unit type are determined from the plurality of energy units, and the revenue impact value corresponding to the preset energy unit type is higher than the revenue impact value corresponding to other energy unit types; Based on the energy unit runtime corresponding to the energy unit with the preset energy unit type, determine the energy units whose energy unit runtime meets the preset runtime range from the energy units with the preset energy unit type. The predicted revenue of the virtual power plant is determined based on the operating power of the energy unit corresponding to the energy unit whose operating time meets the preset operating time range.

3. The virtual power plant scheduling optimization method according to claim 2, characterized in that, The step of determining the predicted revenue of the virtual power plant based on the operating power of energy units whose operating duration meets a preset operating duration range includes: Obtain the historical load demand of the virtual power plant; Based on the historical load demand of the virtual power plant, the predicted load demand is determined; Based on the predicted load demand and the operating power of the energy unit corresponding to the energy unit whose operating time meets the preset operating time range, the load influence weight is determined. If the operating power of the energy unit can meet the predicted load demand, the load influence weight is positive; if the operating power of the energy unit cannot meet the predicted load demand, the load influence weight is negative. The predicted revenue of the virtual power plant is determined based on the operating power of the energy unit corresponding to the energy unit whose operating time meets the preset operating time range, the load influence weight, and the pre-trained revenue prediction model.

4. The virtual power plant scheduling optimization method according to claim 1, characterized in that, The step of determining the first virtual power plant dispatch strategy based on the demand bidding, the predicted revenue, and the historical market transaction data includes: The demand bid is compared with the historical grid bids in the historical market transaction data to determine the target historical grid bid that matches the demand bid; The predicted revenue is compared with the historical virtual power plant revenue corresponding to the target historical grid bidding to determine the target historical virtual power plant revenue that matches the predicted revenue. The first virtual power plant scheduling strategy is determined based on the historical virtual power plant scheduling strategy corresponding to the target historical virtual power plant revenue.

5. The virtual power plant scheduling optimization method according to claim 4, characterized in that, The historical virtual power plant scheduling strategy includes operation optimization strategies and optimization costs corresponding to multiple energy units. Determining the first virtual power plant scheduling strategy based on the historical virtual power plant scheduling strategy corresponding to the target historical virtual power plant revenue includes: If there is only one historical virtual power plant scheduling strategy corresponding to the target historical virtual power plant revenue, then that historical virtual power plant scheduling strategy shall be determined as the first virtual power plant scheduling strategy. If there are multiple historical virtual power plant scheduling strategies corresponding to the target historical virtual power plant revenue, determine the number of energy units involved in each of the multiple historical virtual power plant scheduling strategies; From the multiple historical virtual power plant scheduling strategies, determine the historical virtual power plant scheduling strategy that involves the fewest energy units. Among the historical virtual power plant scheduling strategies with the fewest number of energy units involved, the historical virtual power plant scheduling strategy with the lowest optimization cost corresponding to the operation optimization strategy is determined as the first virtual power plant scheduling strategy.

6. The virtual power plant scheduling optimization method according to claim 1, characterized in that, The step of optimizing the scheduling of each energy unit of the virtual power plant according to the first virtual power plant scheduling strategy includes: Based on the first virtual power plant scheduling strategy and the real-time operation data corresponding to the first energy unit, the predicted operation data corresponding to the first energy unit is determined; Based on the predicted operating data corresponding to the first energy unit and the real-time operating data corresponding to energy units other than the first energy unit, the external revenue and transaction volume deviation of the virtual power plant are predicted. Based on the external revenue and the deviation of the transaction volume, the evaluation result of the first virtual power plant dispatching strategy is determined. This evaluation result is used to characterize whether the first virtual power plant dispatching strategy is the optimal dispatching strategy. Based on the evaluation results of the first virtual power plant scheduling strategy, the scheduling of each energy unit of the virtual power plant is optimized.

7. The virtual power plant scheduling optimization method according to claim 6, characterized in that, The step of optimizing the scheduling of each energy unit of the virtual power plant based on the evaluation results of the first virtual power plant scheduling strategy includes: If the evaluation result of the first virtual power plant scheduling strategy indicates that the first virtual power plant scheduling strategy is the optimal scheduling strategy, the first virtual power plant scheduling strategy is executed, and the historical market transaction data is updated according to the first virtual power plant scheduling strategy, the demand bidding, and the predicted revenue. If the evaluation result of the first virtual power plant scheduling strategy indicates that the first virtual power plant scheduling strategy is not the optimal scheduling strategy, the historical virtual power plant revenue that meets the preset virtual power plant revenue conditions is determined from the historical market transaction data, and the second energy unit is determined based on the energy unit involved in the historical virtual power plant scheduling strategy corresponding to the historical virtual power plant revenue that meets the preset virtual power plant revenue conditions and the first energy unit. Based on the external revenue, the transaction volume deviation, and the real-time operating data corresponding to the second energy unit, a second virtual power plant scheduling strategy is determined, which includes the operation optimization strategy corresponding to the second energy unit. The first virtual power plant scheduling strategy and the second virtual power plant scheduling strategy are executed to optimize the scheduling of each energy unit of the virtual power plant.

8. The virtual power plant scheduling optimization method according to claim 7, characterized in that, The step of determining the second virtual power plant dispatch strategy based on the external revenue, the transaction volume deviation, and the real-time operating data corresponding to the second energy unit includes: Obtain historical deviation thresholds for external revenue and transaction volume; If the external revenue is greater than or equal to the average of the historical external revenue, and the transaction volume deviation is greater than or equal to the transaction volume deviation threshold, the second virtual power plant scheduling strategy is determined based on the first operating parameter in the real-time operating data corresponding to the second energy unit. The first operating parameter does not affect the output power of the second energy unit. If the external revenue is less than the average of the historical external revenue, and / or the transaction volume deviation is less than the transaction volume deviation threshold, the second virtual power plant scheduling strategy is determined based on the second operating parameter in the real-time operating data corresponding to the second energy unit, and the second operating parameter affects the output power of the second energy unit.

9. The virtual power plant scheduling optimization method according to any one of claims 1 to 8, characterized in that, The virtual power plant scheduling optimization method also includes: Obtain scheduling optimization requests triggered by virtual power plant management users, wherein the scheduling optimization requests include demand costs and demand benefits; Based on the demand bidding, the predicted revenue, the demand cost, and the demand revenue, the evaluation result of the scheduling optimization request is determined, and the evaluation result of the scheduling optimization request is used to characterize whether the scheduling optimization request is a normal scheduling optimization request. If the scheduling optimization request is a normal scheduling optimization request, the energy units of the virtual power plant will be optimized and scheduled again based on the demand cost and demand benefit. If the scheduling optimization request is not a normal scheduling optimization request, a prompt message is output, which indicates that the scheduling optimization request is invalid.

10. The virtual power plant scheduling optimization method according to any one of claims 1 to 8, characterized in that, The virtual power plant scheduling optimization method also includes: After determining that the optimized scheduling of multiple energy units of the virtual power plant has been completed, the optimized scheduling operation data corresponding to each of the multiple energy units is obtained; Based on the optimized scheduling operation data corresponding to the multiple energy units, the operation status evaluation results and performance evaluation results corresponding to the multiple energy units are determined respectively. Based on the operational status assessment results and the performance assessment results, an energy unit optimization strategy is determined; The virtual power plant is optimized according to the energy unit optimization strategy.

Images