A Multi-Scenario Collaborative Control Method and System for Virtual Power Plants for Power Supply Security

Abstract

This invention relates to the field of virtual power plant control technology, and provides a method and system for multi-scenario collaborative control of virtual power plants for power supply security. The method includes: periodically acquiring virtual power plant supply and demand time-series data; analyzing the supply and demand fluctuation characteristics of the virtual power plant supply and demand time-series data to generate supply and demand fluctuation time-series data; based on the supply and demand fluctuation time-series data, performing scenario matching analysis based on a pre-built multi-scenario control strategy library to obtain an initial scenario control strategy; performing scenario adaptability simulation analysis on the initial scenario control strategy, and optimizing the initial scenario control strategy based on the simulation analysis results to obtain a target scenario control strategy; decomposing the control objectives in the target scenario control strategy into hierarchical and partitioned control objectives to obtain hierarchical and partitioned control objectives, and executing virtual power plant control according to the hierarchical and partitioned control objectives. This invention can significantly improve the collaborative optimization capability and operational resilience of virtual power plants in complex and uncertain environments, providing reliable technical support for power supply security.

Description

Technical Field

[0001] This invention relates to the field of virtual power plant control technology, and in particular to a method and system for multi-scenario collaborative control of virtual power plants for power supply security. Background Technology

[0002] Virtual power plants, as an important means for power systems to cope with energy transition and ensure power supply, play a crucial role in achieving efficient utilization of new energy sources and ensuring stable grid operation. They provide flexibility and stability to the power system by aggregating distributed energy resources, energy storage devices, and flexible loads to form a collaboratively regulated energy network.

[0003] However, most existing virtual power plant control methods rely on centralized dispatch of a single flexible resource, which is difficult to adapt to the random fluctuations in renewable energy supply and the diverse changes in load demand. The control efficiency is low and the support for the dynamic response of the power grid is insufficient. Some multi-flexible resource collaborative dispatch designs lack efficient interaction of multi-flexible resources in dynamic environments and quantitative assessment of scenario adaptation, resulting in insufficient resource utilization and difficulty in balancing power grid stability and resource collaborative optimization. Consequently, the control effect of virtual power plants in complex scenarios is limited, making it difficult to fully realize their potential to effectively meet the power supply needs. Summary of the Invention

[0004] The purpose of this invention is to provide a multi-scenario collaborative control method for virtual power plants aimed at ensuring power supply. By combining a smart scenario control strategy acquisition mechanism based on supply and demand fluctuations with a scenario adaptability analysis mechanism and a hierarchical and partitioned decomposition mechanism for control targets, the virtual power plant control design can not only achieve efficient interaction between source, load and storage in complex and uncertain environments, and improve the grid's peak shaving and valley filling capabilities, but also effectively smooth out supply and demand fluctuations, ensure stable grid operation, and thus significantly improve the collaborative optimization capability and operational resilience of virtual power plants, providing reliable technical support for ensuring power supply.

[0005] To achieve the above objectives, it is necessary to provide a method and system for multi-scenario collaborative control of virtual power plants for power supply security, addressing the aforementioned technical issues.

[0006] In a first aspect, embodiments of the present invention provide a multi-scenario collaborative control method for virtual power plants oriented towards power supply security. The virtual power plant includes multiple resource aggregation regions, and the resource aggregation regions include various flexible resources. The method includes:

[0007] The system periodically acquires virtual power plant supply and demand time-series data, analyzes the supply and demand fluctuation characteristics of the virtual power plant supply and demand time-series data, and generates corresponding supply and demand fluctuation time-series data; the virtual power plant supply and demand time-series data includes new energy supply time-series data and load demand time-series data;

[0008] Based on the time-series data of supply and demand fluctuations, scenario matching analysis is performed on a pre-built multi-scenario control strategy library to obtain the corresponding initial scenario control strategy. The multi-scenario control strategy library includes benchmark feature vectors of supply and demand fluctuations under various fluctuation scenarios and benchmark control strategies for each scenario. The benchmark control strategy for each scenario includes control parameters, control constraints, and control objectives.

[0009] The initial scene control strategy is subjected to scene adaptability simulation analysis, and the initial scene control strategy is optimized based on the corresponding simulation analysis results to obtain the target scene control strategy.

[0010] The control objectives in the target scenario control strategy are decomposed into hierarchical and partitioned control objectives to obtain hierarchical and partitioned control objectives. Based on the hierarchical and partitioned control objectives, virtual power plant control is performed for the corresponding period.

[0011] Furthermore, the construction steps of the multi-scenario control strategy library include:

[0012] Source, load and storage information is extracted from the historical operation data of the virtual power plant to obtain the power plant's historical energy production sequence, power plant's historical load demand sequence and power plant's energy storage charging and discharging power sequence.

[0013] According to the preset time period, the power plant's historical energy production sequence, the power plant's historical load demand sequence, and the power plant's energy storage charging and discharging power sequence are divided into time periods to obtain the corresponding sets of energy production subsequences, load demand subsequences, and energy storage charging and discharging power subsequences.

[0014] Fluctuation features are extracted from each subsequence in the energy output subsequence set, the load demand subsequence set, and the energy storage charging and discharging power subsequence set to obtain the corresponding energy fluctuation feature dataset, load fluctuation feature dataset, and energy storage charging and discharging power fluctuation feature dataset.

[0015] Cluster analysis was performed on the energy fluctuation feature dataset, the load fluctuation feature dataset, and the energy storage charging and discharging power fluctuation feature dataset to obtain multiple source-load-storage interaction scenarios.

[0016] The energy fluctuation characteristic data and load fluctuation characteristic data of each of the source-load-storage interaction scenarios are spliced ​​together to obtain the corresponding supply and demand fluctuation benchmark feature vector. Based on the historical control strategies of each of the source-load-storage interaction scenarios, the corresponding scenario benchmark control strategy is obtained.

[0017] The supply and demand fluctuation benchmark feature vectors and scenario benchmark control strategies of each source-load-storage interaction scenario are summarized to obtain the multi-scenario control strategy library.

[0018] Furthermore, the historical operating data of the virtual power plant includes historical sequences of regional energy production, regional load demand, and regional energy storage charging and discharging power for each resource aggregation area;

[0019] The steps of extracting source-load-storage information from the historical operation data of the virtual power plant to obtain the historical energy production sequence of the power plant, the historical load demand sequence of the power plant, and the energy storage charging and discharging power sequence of the power plant include:

[0020] The historical sequences of regional energy production, regional load demand, and regional energy storage charging and discharging power in each resource aggregation region are sequentially cleaned and time-aligned to obtain the corresponding regional energy production sequence, regional load demand sequence, and regional energy storage charging and discharging power sequence to be analyzed.

[0021] The energy production sequences of all resource aggregation regions to be analyzed are summarized according to the sampling time to obtain the corresponding historical energy production sequences of power plants. The load demand sequences of all resource aggregation regions to be analyzed are summarized according to the sampling time to obtain the corresponding historical load demand sequences of power plants.

[0022] The energy storage charging and discharging power sequences of each resource aggregation region to be analyzed are weighted and summarized with the corresponding regional energy storage capacity ratio as the weight to obtain the corresponding power plant energy storage charging and discharging power sequence.

[0023] Furthermore, the supply and demand fluctuation time series data includes energy supply fluctuation time series data and load demand fluctuation time series data;

[0024] The step of obtaining the corresponding initial scenario control strategy by performing scenario matching analysis based on the supply and demand fluctuation time series data and a pre-built multi-scenario control strategy library includes:

[0025] The corresponding fluctuation features are extracted from the time series data of energy supply fluctuation and the time series data of load demand fluctuation, respectively, to generate corresponding energy supply fluctuation features and load demand fluctuation features.

[0026] Based on the energy supply fluctuation characteristics and the load demand fluctuation characteristics, corresponding energy supply fluctuation feature vectors and load demand fluctuation feature vectors are generated respectively.

[0027] The energy supply fluctuation feature vector and the load demand fluctuation feature vector are concatenated to obtain the supply and demand fluctuation feature vector.

[0028] The similarity analysis is performed between the supply and demand fluctuation feature vector and the supply and demand fluctuation benchmark feature vector corresponding to each scenario control strategy in the multi-scenario regulation strategy library to obtain the corresponding scenario similarity value.

[0029] Based on all the scene similarity values, several relevant scene control strategies are obtained based on a preset similarity threshold;

[0030] The scene similarity values ​​corresponding to each of the relevant scene control strategies are normalized to obtain the corresponding similarity weights.

[0031] Based on the similarity weights of the various related scenario control strategies, all the related scenario control strategies are weighted and fused to obtain the initial scenario control strategy.

[0032] Furthermore, the control parameters include new energy control parameters, energy storage control parameters, and load control parameters;

[0033] The steps of performing scene adaptability simulation analysis on the initial scene control strategy and optimizing the initial scene control strategy based on the corresponding simulation analysis results to obtain the target scene control strategy include:

[0034] Based on the initial scenario control strategy and the virtual power plant supply and demand time series data, a virtual power plant strategy scheduling simulation is performed to obtain the corresponding simulation operation data.

[0035] Based on the simulation data, corresponding scenario adaptability evaluation indicators are obtained; the scenario adaptability evaluation indicators include supply and demand balance error rate, response delay time, regulation resource utilization rate, and over-limit frequency; the regulation resource utilization rate includes new energy utilization rate, load utilization rate, and energy storage utilization rate.

[0036] The scene adaptability evaluation indexes are compared and analyzed with the corresponding scene adaptability evaluation index thresholds to obtain the corresponding scene adaptability evaluation results.

[0037] Obtain the scenario adaptability assessment index items that do not meet the standards in the scenario adaptability assessment results, and obtain the control parameters to be optimized from the control parameters based on the preset correlation analysis algorithm;

[0038] Based on preset adjustment rules, each of the control parameters to be optimized in the control parameters is adjusted to obtain the target scene control strategy.

[0039] Furthermore, the control targets include control values ​​for new energy production, load demand, and energy storage charging and discharging power.

[0040] The hierarchical and regional control objectives include the regional control objectives of each resource aggregation area and the resource control objectives of each flexible resource within each resource aggregation area.

[0041] The step of decomposing the control targets in the target scene control strategy into hierarchical and partitioned control targets includes:

[0042] The control targets in the target scenario control strategy are hierarchically split based on the preset regional target allocation principle to obtain the regional control targets of each resource aggregation region.

[0043] Based on the regional regulation objectives of each resource aggregation area, the regional objectives are decomposed according to the resource scheduling optimization model within the preset area to obtain the resource regulation objectives of each flexible resource within the corresponding resource aggregation area; the resource scheduling optimization model within the preset area is constructed with the optimization objective of balancing new energy consumption, regulation costs and grid operation risks.

[0044] Furthermore, the regional regulation targets include regional new energy production regulation targets, regional load demand regulation targets, and regional energy storage charging and discharging power regulation targets;

[0045] The step of splitting the control targets in the target scenario control strategy into hierarchical targets based on a preset regional target allocation principle to obtain the regional control targets of each resource aggregation region includes:

[0046] Based on the proportion of new energy installed capacity and current output capacity of each resource aggregation region, the new energy production regulation and allocation ratio of the corresponding resource aggregation region is obtained, and based on the new energy production regulation and allocation ratio and the new energy production regulation value, the regional new energy production regulation target of the corresponding resource aggregation region is obtained.

[0047] Based on the load capacity and preset control potential coefficient of different control levels in each resource aggregation area, the regional load control potential is summarized and analyzed to obtain the corresponding regional load control potential ratio. Based on the regional load control potential ratio and the load demand control value, the corresponding regional load demand control target is obtained.

[0048] Based on the energy storage power regulation type of the energy storage charging and discharging power regulation value, the energy storage charging and discharging power allocation ratio of each resource aggregation area is obtained, and based on the energy storage charging and discharging power allocation ratio and the energy storage charging and discharging power regulation value, the corresponding regional energy storage charging and discharging power regulation target is obtained.

[0049] Further, the step of obtaining the energy storage charging and discharging power allocation ratio of each resource aggregation area based on the energy storage power regulation type of the energy storage charging and discharging power regulation value includes:

[0050] When the energy storage power regulation type is energy storage charging power regulation, the energy storage charging efficiency, unit charging cost and allowable charging power ratio of each resource aggregation area are obtained respectively.

[0051] Based on the principle of setting the charging allocation ratio, the energy storage charging efficiency, unit charging cost, and allowable charging power ratio of each resource aggregation area are weighted and fused to obtain the corresponding regional energy storage charging power allocation ratio. The principle of setting the charging allocation ratio includes that the regional energy storage charging power allocation ratio is directly proportional to the energy storage charging efficiency and the allowable charging power ratio, and the regional energy storage charging power allocation ratio is inversely proportional to the unit charging cost.

[0052] Furthermore, the step of obtaining the energy storage charging and discharging power allocation ratio of each resource aggregation area based on the energy storage power regulation type of the energy storage charging and discharging power regulation value further includes:

[0053] When the energy storage charging and discharging power regulation value is the energy storage discharging power regulation value, the energy storage discharging efficiency, unit discharging cost and allowable discharging power ratio of each resource aggregation area are obtained respectively.

[0054] Based on the principle of setting the discharge allocation ratio, the energy storage discharge efficiency, unit discharge cost, and allowable discharge power ratio of each resource aggregation area are weighted and fused to obtain the corresponding regional energy storage discharge power allocation ratio. The principle of setting the discharge allocation ratio includes that the regional energy storage discharge power allocation ratio is directly proportional to the energy storage discharge efficiency and the allowable discharge power ratio, and the regional energy storage discharge power allocation ratio is inversely proportional to the unit discharge cost.

[0055] Secondly, embodiments of the present invention provide a multi-scenario collaborative control system for virtual power plants oriented towards power supply security. The virtual power plant includes multiple resource aggregation areas, and the resource aggregation areas include various flexible resources. The system includes:

[0056] The fluctuation analysis module is used to periodically acquire virtual power plant supply and demand time-series data, and to analyze the supply and demand fluctuation characteristics of the virtual power plant supply and demand time-series data to generate corresponding supply and demand fluctuation time-series data; the virtual power plant supply and demand time-series data includes new energy supply time-series data and load demand time-series data;

[0057] The scenario matching module is used to perform scenario matching analysis based on the supply and demand fluctuation time series data and a pre-built multi-scenario control strategy library to obtain the corresponding initial scenario control strategy. The multi-scenario control strategy library includes supply and demand fluctuation benchmark feature vectors and scenario benchmark control strategies under various fluctuation scenarios. The scenario benchmark control strategy includes control parameters, control constraints, and control targets.

[0058] The strategy acquisition module is used to perform scene adaptability simulation analysis on the initial scene control strategy, and optimize the initial scene control strategy based on the corresponding simulation analysis results to obtain the target scene control strategy.

[0059] The target decomposition module is used to decompose the control targets in the target scenario control strategy into hierarchical and partitioned control targets to obtain hierarchical and partitioned control targets, and to execute virtual power plant control for the corresponding period according to the hierarchical and partitioned control targets.

[0060] This invention provides a method and system for multi-scenario collaborative control of virtual power plants for power supply security. The method periodically acquires virtual power plant supply and demand time-series data, including renewable energy supply time-series data and load demand time-series data. It analyzes the supply and demand fluctuation characteristics of the virtual power plant supply and demand time-series data to generate corresponding supply and demand fluctuation time-series data. Based on this data, it performs scenario matching analysis using a pre-built multi-scenario control strategy library that includes supply and demand fluctuation benchmark feature vectors and scenario benchmark control strategies under various fluctuation scenarios to obtain the corresponding initial scenario control strategy. It then performs scenario adaptability simulation analysis on the initial scenario control strategy, optimizes it based on the simulation results to obtain the target scenario control strategy, and decomposes the control objectives in the target scenario control strategy into hierarchical and partitioned control objectives. Finally, it executes virtual power plant control for the corresponding period based on the hierarchical and partitioned control objectives. Compared with existing technologies, this virtual power plant multi-scenario collaborative control method for power supply security, based on a virtual power plant control design that combines a smart scenario control strategy acquisition mechanism for supply and demand fluctuations with a scenario adaptability analysis mechanism and a hierarchical and partitioned decomposition mechanism for control targets, can not only achieve efficient source-load-storage interaction of virtual power plants in complex and uncertain environments, improving the grid's peak shaving and valley filling capabilities, but also effectively smooth out supply and demand fluctuations, ensuring stable grid operation. This significantly enhances the collaborative optimization capability and operational resilience of virtual power plants, providing reliable technical support for power supply security and showing promising engineering application prospects. Attached Figure Description

[0061] Figure 1 This is a flowchart illustrating the multi-scenario collaborative control method for virtual power plants oriented towards power supply security in this embodiment of the invention.

[0062] Figure 2 This is a schematic diagram of the structure of a virtual power plant multi-scenario collaborative control system for power supply assurance in an embodiment of the present invention;

[0063] The attached figures are labeled as follows:

[0064] 1. Fluctuation Analysis Module; 2. Scene Matching Module; 3. Strategy Acquisition Module; 4. Target Decomposition Module. Detailed Implementation

[0065] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. Obviously, the embodiments described below are only part of the embodiments of this invention and are used to illustrate the invention, but are not intended to limit the scope of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0066] In one embodiment, such as Figure 1 As shown, a multi-scenario collaborative control method for virtual power plants aimed at ensuring power supply is provided. The virtual power plant includes multiple resource aggregation areas, which in turn include various flexible resources, including distributed new energy sources (solar power, hydropower, and wind power, etc.) and adjustable loads (industrial loads, building loads, and transportation loads, etc.). The architecture of the virtual power plant can be understood to be divided into a virtual power plant management layer, a resource aggregation layer (including multiple resource aggregation areas), and a resource layer (including various flexible resources) from top to bottom. The corresponding virtual power plant architecture description can be found in existing layered and partitioned architectures, which will not be detailed here. The multi-scenario collaborative control method for virtual power plants provided by this invention is applicable to the control of virtual power plants with layered and partitioned architectures. The specific method includes:

[0067] S11. Periodically acquire virtual power plant supply and demand time-series data, and analyze the supply and demand fluctuation characteristics of the virtual power plant supply and demand time-series data to generate corresponding supply and demand fluctuation time-series data. The virtual power plant supply and demand time-series data can be understood as multi-dimensional time-series data acquired according to a preset virtual power plant control cycle based on a relevant virtual power plant operation information monitoring system, such as a SCADA (Supervisory Control And Data Acquisition) system. This data is used to analyze the fluctuations in new energy production supply and load demand of the virtual power plant, including new energy supply time-series data and load demand time-series data. It should be noted that the preset virtual power plant control cycle can be set based on actual application control needs to determine the sequence length of the virtual power plant supply and demand time-series data. The actual collection interval of the virtual power plant supply and demand time-series data will be shorter than the preset virtual power plant control cycle. For example, the preset virtual power plant control cycle may be one day or one hour, while the actual collection interval may be in the minute or second range; no specific limitation is made here.

[0068] In this embodiment, the new energy supply time series data can be understood as the total new energy output supply time series data for all resource aggregation areas within the entire virtual power plant. This can be obtained by summing the new energy output values ​​at the same sampling time in the new energy output sequences of all resource aggregation areas, and the new energy output value at each sampling time in the new energy output sequence of each resource aggregation area is the cumulative value of all new energy output in the corresponding resource aggregation area. Correspondingly, the load demand time series data can be understood as the total load demand time series data for all resource aggregation areas within the entire virtual power plant. This can be obtained by summing the load demand values ​​at the same sampling time in the load demand sequences of all resource aggregation areas, and the load demand value at each sampling time in the load demand sequence of each resource aggregation area is the cumulative value of all load demand in the corresponding resource aggregation area.

[0069] In practical applications, after cleaning the time-series data of renewable energy supply and load demand (removing outliers, filling in missing values, etc.), the two data are then time-aligned according to a preset time window (set according to demand) to obtain time-aligned renewable energy supply and load demand time-series data. Then, first-order difference processing is performed on the time-aligned renewable energy supply and load demand time-series data respectively to obtain the corresponding energy supply fluctuation time-series data and load demand fluctuation time-series data, which are the supply and demand fluctuation time-series data used for subsequent intelligent scenario matching. It should be noted that the data cleaning, time alignment, and first-order difference processing in this embodiment can all be implemented with reference to relevant existing technologies, and will not be detailed here.

[0070] S12. Based on the supply and demand fluctuation time series data, scenario matching analysis is performed based on a pre-built multi-scenario control strategy library to obtain the corresponding initial scenario control strategy; the multi-scenario control strategy library includes supply and demand fluctuation benchmark feature vectors and scenario benchmark control strategies under various fluctuation scenarios; wherein, the supply and demand fluctuation benchmark feature vector can be understood as a feature vector obtained by splicing the new energy production supply fluctuation characteristics and load demand fluctuation characteristics; the scenario benchmark control strategy includes control parameters, control constraints and control targets.

[0071] In this embodiment, the control parameters include new energy control parameters, energy storage control parameters, and load control parameters. New energy control parameters may include the output range of new energy units, ramp rate limits, economic costs and physical time consumption for unit start-up and shutdown, etc. Energy storage control parameters may include time-segmented energy storage charging / discharging power thresholds, safe range of energy storage state of charge, and energy storage charging / discharging response priorities (e.g., prioritizing response to power deficits), etc. Load control parameters may include the reduction amount and time period of interruptible loads, the transfer amount and time shift window of shiftable loads, and the control range of temperature-controlled loads, etc. Control constraints can be understood as strategy operating boundaries, and may include allowable power deviation thresholds, reserve capacity requirements, grid security constraints (allowable voltage fluctuation range), and unit control parameters. Controlling cost ceilings, etc.; the corresponding control objectives can be understood as the comprehensive resource control requirements determined by solving the comprehensive optimization objectives of the virtual power plant (e.g., minimizing comprehensive control costs) under the control parameters and control constraints. These include the control values ​​for new energy output, load demand, and energy storage charging and discharging power. The control value for new energy output can be positive or negative (positive means increasing new energy output, negative means decreasing new energy output), the control value for load demand can be positive or negative (positive means decreasing load electricity demand, negative means stimulating load electricity demand), and the control value for energy storage charging and discharging power can be positive, negative, or zero (positive means energy storage releases energy into the virtual power plant, negative means energy storage absorbs energy from the virtual power plant, and zero means energy storage is idle).

[0072] To ensure the efficiency and reliability of acquiring control strategies for supply and demand fluctuation scenarios in virtual power plants during practical applications, this embodiment preferably analyzes and organizes control strategies for typical supply and demand fluctuation scenarios based on historical operating data of virtual power plants to construct a multi-scenario control strategy library. The number and types of fluctuation scenarios in this strategy library vary depending on the actual acquired historical operating data of virtual power plants, and are not specifically limited here. Specifically, the construction steps of the multi-scenario control strategy library include:

[0073] Source-load-storage information is extracted from the historical operation data of the virtual power plant to obtain the historical energy production sequence, historical load demand sequence, and energy storage charging and discharging power sequence of the power plant. The acquisition period of the historical operation data of the virtual power plant can be as long as possible while ensuring the efficiency of calculation and analysis, so as to ensure the accuracy of subsequent source-load-storage interaction scenario extraction. Considering that the fluctuation of new energy production and load demand may have certain seasonal effects, the total data duration of the historical operation data of the virtual power plant should be greater than one year in principle.

[0074] Specifically, the historical operation data of the virtual power plant includes historical sequences of regional energy production, regional load demand, and regional energy storage charging and discharging power for each resource aggregation area; the step of extracting source-load-storage information from the historical operation data of the virtual power plant to obtain the historical energy production sequence, the historical load demand sequence, and the energy storage charging and discharging power sequence of the power plant includes:

[0075] The historical sequences of regional energy production, regional load demand, and regional energy storage charging and discharging power in each resource aggregation region are sequentially cleaned and time-aligned to obtain the corresponding regional energy production sequence, regional load demand sequence, and regional energy storage charging and discharging power sequence to be analyzed. The data cleaning and time alignment processes are as described above and will not be repeated here.

[0076] The energy production sequences of all resource aggregation regions to be analyzed are summarized according to the sampling time to obtain the corresponding historical energy production sequences of power plants. Similarly, the load demand sequences of all resource aggregation regions to be analyzed are summarized according to the sampling time to obtain the corresponding historical load demand sequences of power plants. That is, the historical energy production of power plants at each sampling time in the historical energy production sequence is the sum of the energy production at the corresponding sampling time in the energy production sequences of all resource aggregation regions to be analyzed; and the historical load demand of power plants at each sampling time in the historical load demand sequence is the sum of the load demand at the corresponding sampling time in the load demand sequences of all resource aggregation regions to be analyzed.

[0077] The energy storage charging and discharging power sequences of each resource aggregation region to be analyzed are weighted and summarized with the corresponding regional energy storage capacity ratio as the weight to obtain the corresponding power plant energy storage charging and discharging power sequence. The regional energy storage capacity ratio can be understood as the ratio of the regional energy storage capacity of the resource aggregation region to the total energy storage capacity of the entire virtual power plant. That is, the power plant energy storage charging and discharging power at each sampling time in the power plant energy storage charging and discharging power sequence is the cumulative value of the product of the energy storage charging and discharging power at the corresponding sampling time in the energy storage charging and discharging power sequences of all resource aggregation regions to be analyzed and the regional energy storage capacity ratio.

[0078] This embodiment obtains the comprehensive energy output, comprehensive load demand, and comprehensive energy storage charging and discharging power of the entire virtual power plant by summarizing the energy output, load demand, and regional energy storage charging and discharging power of each resource aggregation area. This effectively ensures the reliability of subsequent supply and demand fluctuation scenario analysis using the entire virtual power plant as the research object.

[0079] According to the preset time period segmentation duration, the power plant's historical energy output sequence, the power plant's historical load demand sequence, and the power plant's energy storage charging and discharging power sequence are respectively segmented into time periods to obtain corresponding sets of energy output subsequences, load demand subsequences, and energy storage charging and discharging power subsequences. The preset time period segmentation duration can be consistent with the virtual power plant's control cycle to ensure the rationality of intelligent scenario matching based on a multi-scenario control strategy library. For example, if the virtual power plant's control cycle is one day, the preset time period segmentation duration can also be set to one day, thus segmenting the power plant's historical energy output sequence, power plant's historical load demand sequence, and power plant's energy storage charging and discharging power sequence into multiple subsequences with a corresponding sequence duration of one day, obtaining corresponding sets of subsequences.

[0080] Fluctuation features are extracted from each subsequence in the energy output subsequence set, the load demand subsequence set, and the energy storage charging and discharging power subsequence set to obtain corresponding energy fluctuation feature datasets, load fluctuation feature datasets, and energy storage charging and discharging power fluctuation feature datasets. The energy fluctuation feature dataset includes the average output, output fluctuation variance, and peak-to-valley output difference (the difference between maximum and minimum energy output) for each time period. The load fluctuation feature dataset includes the average load, load fluctuation variance, and peak-to-valley load difference (the difference between maximum and minimum load demand) for each time period. The energy storage charging and discharging power fluctuation feature dataset includes the average energy storage charging and discharging power and the average rate of change of energy storage charging and discharging power. It should be noted that the acquisition process of the energy fluctuation feature dataset, load fluctuation feature dataset, and energy storage charging and discharging power fluctuation feature dataset can refer to existing methods for calculating related fluctuation features, and will not be detailed here.

[0081] Cluster analysis is performed on the energy fluctuation feature dataset, the load fluctuation feature dataset, and the energy storage charging and discharging power fluctuation feature dataset to obtain multiple source-load-storage interaction scenarios. The cluster analysis process may include concatenating feature data from the same time period in the energy fluctuation feature dataset, load fluctuation feature dataset, and energy storage charging and discharging power fluctuation feature dataset to obtain corresponding feature vectors, and then performing cluster analysis based on the cosine similarity of the feature vectors as a distance metric to obtain multiple source-load-storage interaction scenarios. This embodiment uses energy fluctuation features, load fluctuation features, and energy storage charging and discharging power fluctuation features simultaneously to construct source-load-storage interaction scenarios, which can effectively capture the interaction relationship between source, load, and storage under supply and demand fluctuations in virtual power plants, providing a reliable guarantee for the effectiveness of extracting typical operating scenarios of virtual power plants.

[0082] The energy fluctuation characteristic data and load fluctuation characteristic data of each of the source-load-storage interaction scenarios are concatenated to obtain the corresponding supply and demand fluctuation benchmark feature vector. Based on the historical control strategies of each of the source-load-storage interaction scenarios, the corresponding scenario benchmark control strategy is obtained. It should be noted that the supply and demand fluctuation benchmark feature vector is obtained only by concatenating the energy fluctuation characteristic data and load fluctuation characteristic data to facilitate efficient matching with the supply and demand fluctuation data of the virtual power plant.

[0083] The supply and demand fluctuation benchmark feature vectors and scenario benchmark control strategies of each source-load-storage interaction scenario are summarized to obtain the multi-scenario control strategy library.

[0084] This embodiment constructs a multi-scenario control strategy library based on the source-load-storage interaction relationship and supply-demand fluctuation analysis during virtual power plant operation. This not only ensures the scientific validity and practical relevance of the extracted scenarios within the library but also guarantees efficient and reliable intelligent scenario matching based on supply-demand fluctuation data from actual virtual power plant operation. This allows for accurate acquisition of the required scenario control strategies and timely response to scheduling needs. It should be noted that to ensure the continued reliability of the multi-scenario control strategy library in practical applications, it can be updated and optimized based on the online operation data of the virtual power plant according to the aforementioned construction steps. This ensures the adaptability of the scenarios in the multi-scenario control strategy library to the actual supply-demand fluctuation scenarios of the virtual power plant.

[0085] Specifically, the step of obtaining the corresponding initial scenario control strategy by performing scenario matching analysis based on the supply and demand fluctuation time series data and a pre-built multi-scenario control strategy library includes:

[0086] The time series data of energy supply fluctuations and the time series data of load demand fluctuations are respectively extracted to generate corresponding energy supply fluctuation features and load demand fluctuation features. Among them, the energy supply fluctuation features include the average output, the variance of output fluctuations, and the peak-to-valley difference of output, and the load demand fluctuation features include the average load, the variance of load fluctuations, and the peak-to-valley difference of load. The specific extraction process will not be detailed here.

[0087] Based on the energy supply fluctuation characteristics and the load demand fluctuation characteristics, corresponding energy supply fluctuation feature vectors and load demand fluctuation feature vectors are generated respectively. The energy supply fluctuation feature vector can be understood as a feature vector obtained by sequentially concatenating the data such as the average output, the variance of output fluctuation, and the peak-to-valley difference of output in the energy supply fluctuation characteristics. The corresponding load demand fluctuation feature vector can be understood as a feature vector obtained by sequentially concatenating the data such as the average load, the variance of load fluctuation, and the peak-to-valley difference of load in the load demand fluctuation characteristics.

[0088] The energy supply fluctuation feature vector and the load demand fluctuation feature vector are concatenated to obtain the supply and demand fluctuation feature vector; that is, the supply and demand fluctuation feature vector is a higher-dimensional feature vector obtained by concatenating the energy supply fluctuation feature vector and the load demand fluctuation feature vector.

[0089] The similarity analysis is performed between the supply and demand fluctuation feature vector and the supply and demand fluctuation benchmark feature vector corresponding to each scenario control strategy in the multi-scenario regulation strategy library to obtain the corresponding scenario similarity value; wherein, the similarity analysis can be implemented using existing related similarity analysis techniques.

[0090] Based on all the scene similarity values, several relevant scene control strategies are obtained based on a preset similarity threshold. The preset similarity threshold can be set according to actual application needs and is not specifically limited here. The corresponding relevant scene control strategies can be understood as one or more scene control strategies in the multi-scene control strategy library whose scene similarity value with the current supply and demand fluctuation feature vector is higher than the preset similarity threshold.

[0091] The scene similarity values ​​corresponding to each of the relevant scene control strategies are normalized to obtain the corresponding similarity weights; wherein, the normalization process can be implemented using any existing normalization technique, such as max-min normalization.

[0092] Based on the similarity weights of the various related scenario control strategies, all the related scenario control strategies are weighted and fused to obtain the initial scenario control strategy; wherein, the initial scenario control strategy can be understood as a control strategy obtained by weighting all parameters of the related scenario control strategies according to their corresponding similarity weights.

[0093] This embodiment effectively ensures the reliability and real-time performance of obtaining actual scenario control strategies by matching the supply and demand fluctuation data of the actual virtual power plant with known classic fluctuation scenarios in the multi-scenario control strategy library, thereby ensuring the high efficiency of virtual power plant scheduling optimization response.

[0094] S13. Perform scene adaptability simulation analysis on the initial scene control strategy, and optimize the initial scene control strategy based on the corresponding simulation analysis results to obtain the target scene control strategy.

[0095] Scenario-adaptive simulation analysis can be understood as constructing a virtual simulation environment based on existing virtual power plant simulation tools (such as GridLAB-D), incorporating new energy sources, loads, energy storage, and grid operation constraints. This environment is used to simulate and verify the effectiveness of the initial scenario control strategy in fluctuating scenarios, ensuring a target scenario control strategy that more closely reflects actual supply and demand fluctuations. Specifically, the steps of performing scenario-adaptive simulation analysis on the initial scenario control strategy and optimizing it based on the simulation results to obtain the target scenario control strategy include:

[0096] Based on the initial scenario control strategy and the virtual power plant supply and demand time series data, a virtual power plant strategy scheduling simulation is performed to obtain the corresponding simulation operation data. The simulation operation data may include the total energy supply value, the total load demand value, the strategy effective delay (the time interval between the issuance of the strategy and its effective date), the actual power of renewable energy, the actual power of load, the actual power of energy storage, the allowed power of renewable energy, the allowed power of load, the allowed power of energy storage, and the number of grid over-limits (such as the number of voltage over-limits, current over-limits, and frequency over-limits).

[0097] Based on the simulation data, corresponding scenario adaptability evaluation indicators are obtained. These indicators include supply-demand balance error rate, response delay time, regulation resource utilization rate, and limit violation frequency. Regulation resource utilization rate includes renewable energy utilization rate, load utilization rate, and energy storage utilization rate. The supply-demand balance error rate is the percentage of the absolute difference between total energy supply and total load demand, with a lower rate indicating stronger policy mitigation capabilities. Response delay time reflects the policy's effective delay and fluctuation response efficiency. Renewable energy utilization rate is the ratio of actual renewable energy usage power to permitted renewable energy usage power; load utilization rate is the ratio of actual load usage power to permitted load usage power; and energy storage utilization rate is the ratio of actual energy storage usage power to permitted energy storage usage power, effectively reflecting the activation efficiency of various resources. Limit violation frequency is the ratio of the number of grid limit violations to the number of regulation operations, reflecting grid security during policy regulation. It should be noted that in practical applications, indicators such as unit regulation cost, virtual power plant revenue fluctuation, and equipment loss rate can be added based on application requirements.

[0098] The scenario adaptability assessment indicators are compared and analyzed with their corresponding scenario adaptability assessment indicator thresholds to obtain the corresponding scenario adaptability assessment results. The scenario adaptability assessment indicator thresholds can be understood as effective assessment thresholds set for each scenario adaptability assessment indicator, corresponding one-to-one with each scenario adaptability assessment indicator. The value of each scenario adaptability assessment indicator threshold can be selected based on actual application needs, and no specific limitation is made here. Correspondingly, the scenario adaptability assessment results include the assessment results of whether each of the above-mentioned scenario adaptability assessment indicators meets the scenario adaptability assessment indicator threshold requirements. If it does not meet the requirements, the indicator is considered to have failed to meet the standards; if it does, the indicator is considered to have met the standards.

[0099] Obtain the scenario adaptability assessment indicators that do not meet the standards from the scenario adaptability assessment results, and obtain the control parameters to be optimized from the control parameters based on the preset association analysis algorithm. The preset association analysis algorithm can be the Apriori algorithm, which performs association reasoning analysis based on the pre-set association rules between each scenario adaptability assessment indicator and each control parameter to find the control parameters corresponding to each non-compliant indicator as the control parameters to be optimized. The specific implementation process of association reasoning analysis based on the Apriori algorithm can be referred to the relevant existing technology, and will not be described in detail here.

[0100] Based on preset adjustment rules, each of the control parameters to be optimized in the control parameters is adjusted to obtain the target scenario control strategy. The preset adjustment rules can be set according to actual application needs. For example, the adjustment ratio of the control parameters can be set based on the deviation between the unmet scenario adaptability evaluation index and the corresponding scenario adaptability evaluation index threshold. This is used to fine-tune the control parameters to be optimized. Then, based on the obtained adjusted control parameters and the corresponding control constraints, the corresponding control objectives are fine-tuned based on the comprehensive optimization objectives of the virtual power plant. No specific limitations are made here.

[0101] This embodiment obtains the target scenario control strategy by simulating and verifying the control effect of the initial scenario control strategy obtained based on intelligent scenario matching and fine-tuning the control strategy. This can effectively ensure the adaptability of the target scenario control strategy to the actual virtual power plant supply and demand fluctuation scenario, thereby improving the application effect of the control strategy.

[0102] S14. The control targets in the target scenario control strategy are decomposed into hierarchical and partitioned control targets to obtain hierarchical and partitioned control targets. Based on the hierarchical and partitioned control targets, the virtual power plant control is executed for the corresponding period. The control targets can be understood as the control requirements for the entire virtual power plant, including the control values ​​for new energy production, load demand, and energy storage charging and discharging power. The hierarchical and partitioned decomposition can be understood as first decomposing the control targets to each resource aggregation region, and then each resource aggregation region further decomposes the corresponding regional control targets to the control targets of each flexible resource within the region. The resulting hierarchical and partitioned control targets include the regional control targets of each resource aggregation region and the resource control targets of each flexible resource within each resource aggregation region.

[0103] Specifically, the step of decomposing the control targets in the target scenario control strategy into hierarchical and partitioned control targets to obtain hierarchical and partitioned control targets includes:

[0104] The control targets in the target scenario control strategy are hierarchically decomposed based on a preset regional target allocation principle to obtain regional control targets for each resource aggregation region. The preset regional target allocation principle can be understood as a rule for decomposing the control targets into regional control targets for each resource aggregation region. Preferably, this embodiment includes decomposing the new energy production control value in the control targets based on the proportion of new energy installed capacity in each resource aggregation region, decomposing the load demand control value in the control targets based on the load control potential of each resource aggregation region, and decomposing the energy storage charging and discharging power control value in the control targets based on the charging and discharging efficiency and charging and discharging cost of each resource aggregation region. The resulting regional control targets include regional new energy production control targets, regional load demand control targets, and regional energy storage charging and discharging power control targets. Specifically, the step of hierarchically decomposing the control targets in the target scenario control strategy based on the preset regional target allocation principle to obtain regional control targets for each resource aggregation region includes:

[0105] Based on the proportion of new energy installed capacity and the current available output proportion of each resource aggregation region, the new energy production control allocation ratio for the corresponding resource aggregation region is obtained. Then, based on the new energy production control allocation ratio and the new energy production control value, the regional new energy production control target for the corresponding resource aggregation region is obtained. Here, the proportion of new energy installed capacity can be understood as the ratio of the new energy installed capacity of the resource aggregation region to the total new energy installed capacity of the entire virtual power plant. The current available output proportion can be understood as the ratio of the actual available output limit of the resource aggregation region to the new energy installed capacity of that region. This is a correction coefficient added to account for the potential impact of environmental factors on the actual new energy in the resource aggregation region (for example, if the wind speed in resource aggregation region A is lower than the predicted value, the corresponding wind turbine output limit will be reduced to 50% of the corresponding installed capacity, while the wind turbine output limit in other regions can reach 80%, requiring a reduction in their corresponding allocation ratio to ensure resource utilization). In practical applications, the product of the proportion of new energy installed capacity and the current available output proportion of each resource aggregation region is used as the required new energy production control allocation ratio. Then, the required regional new energy production control target is obtained based on the product of the new energy production control allocation ratio and the new energy production control value. This embodiment uses the proportion of new energy installed capacity as the basic allocation ratio, and then calculates the final new energy production control allocation ratio based on the current output ratio affected by the regional environment as a correction coefficient. This approach can simultaneously ensure the fairness and rationality of the allocation of control targets in each region.

[0106] Based on the load capacity and preset control potential coefficient of different control levels in each resource aggregation area, a summary analysis of regional load control potential is performed to obtain the corresponding regional load control potential ratio. Then, based on the regional load control potential ratio and the load demand control value, the corresponding regional load demand control target is obtained. The regional load control potential summary analysis process may include: classifying the adjustable loads in each resource aggregation area according to their corresponding control levels, and setting corresponding preset control potential coefficients for adjustable loads at different control levels (the higher the control level, the larger the corresponding preset control potential coefficient can be selected within the range of 0 to 1); multiplying the sum of the adjustable load capacities at the same control level in the same resource aggregation area by the preset control potential coefficient of the corresponding control level to obtain the load control potential of the corresponding control level; summing the load control potentials of all control levels in the same resource aggregation area to obtain the regional load control potential; and then comparing the regional load control potential of each resource aggregation area with the sum of all regional load control potentials to obtain the corresponding regional load control potential ratio. The corresponding regional load demand control target can be understood as the product of the proportion of regional load control potential and the load demand control value. If the obtained regional load demand control target exceeds the maximum controllable load of the corresponding resource aggregation area, the portion of the regional load demand control target that exceeds the maximum controllable load needs to be allocated to other resource aggregation areas that still have control potential according to the proportion of regional load control potential of other resource aggregation areas, so as to ensure that the regional load demand control targets of each resource aggregation area are within the corresponding allowable control range.

[0107] Based on the energy storage power regulation type of the energy storage charging and discharging power regulation value, the energy storage charging and discharging power allocation ratio of each resource aggregation area is obtained, and based on the energy storage charging and discharging power allocation ratio and the energy storage charging and discharging power regulation value, the corresponding regional energy storage charging and discharging power regulation target is obtained; wherein, the energy storage power regulation type may include energy storage charging power regulation and energy storage discharging power regulation, and the corresponding energy storage charging and discharging power regulation value includes energy storage charging power regulation value and energy storage discharging power regulation value; specifically, the step of obtaining the energy storage charging and discharging power allocation ratio of each resource aggregation area based on the energy storage power regulation type of the energy storage charging and discharging power regulation value includes:

[0108] When the energy storage power regulation type is energy storage charging power regulation, the energy storage charging efficiency, unit charging cost, and allowable charging power ratio of each resource aggregation region are obtained respectively; wherein, the allowable charging power ratio can be understood as the ratio of the allowable charging power of the resource aggregation region to the sum of the allowable charging power of all resource aggregation regions; it should be noted that the energy storage charging efficiency, unit charging cost, and allowable charging power of each resource aggregation region can be obtained based on the relevant virtual power plant operation information monitoring system, which will not be described in detail here.

[0109] Based on the principle of setting charging allocation ratios, the energy storage charging efficiency, unit charging cost, and allowable charging power ratio of each resource aggregation region are weighted and fused to obtain the corresponding regional energy storage charging power allocation ratio. The principle of setting the charging allocation ratio includes that the regional energy storage charging power allocation ratio is directly proportional to the energy storage charging efficiency and the allowable charging power ratio, and inversely proportional to the unit charging cost; that is, the regional energy storage charging power allocation ratio can be expressed as:

[0110]

[0111] In the formula, This represents the regional energy storage charging power allocation ratio within the resource aggregation area q; and These represent the unit charging cost and the corresponding reference value for the unit charging cost in the resource aggregation region q, respectively. and These represent the energy storage charging efficiency and the percentage of allowed charging power in the resource aggregation region q, respectively. , and This represents the weighting coefficients. The magnitude of each weighting coefficient can be adjusted according to actual application needs, as long as it meets the following requirements: This ensures the reliability and fairness of the energy storage charging power allocation ratio.

[0112] In practical applications, the corresponding regional energy storage charging power control target can be obtained by multiplying the energy storage charging power allocation ratio and the corresponding energy storage charging and discharging power control value (actually the energy storage charging power control value). However, considering that the obtained regional energy storage charging power control target may exceed the corresponding upper limit of the allowed charging power, in order to ensure the rationality of the regional energy storage charging power control target setting, this embodiment preferably determines whether the regional energy storage charging power control target of each resource aggregation area is greater than the corresponding upper limit of the allowed charging power after obtaining the regional energy storage charging power control target of each resource aggregation area. If so, the charging power of the regional energy storage charging power control target that exceeds the upper limit of the allowed charging power is allocated to other resource aggregation areas with control potential, so as to ensure that the regional energy storage charging power control target of each resource aggregation area is within the corresponding allowed control range.

[0113] Correspondingly, the step of obtaining the energy storage charging and discharging power allocation ratio of each resource aggregation area based on the energy storage power regulation type of the energy storage charging and discharging power regulation value further includes:

[0114] When the energy storage charging and discharging power regulation value is the energy storage discharging power regulation value, the energy storage discharging efficiency, unit discharging cost, and allowable discharging power ratio of each resource aggregation region are obtained respectively; wherein, the allowable discharging power ratio can be understood as the ratio of the allowable discharging power of the resource aggregation region to the sum of the allowable discharging power of all resource aggregation regions; it should be noted that the energy storage discharging efficiency, unit discharging cost, and allowable discharging power of each resource aggregation region can be obtained based on the relevant virtual power plant operation information monitoring system, which will not be described in detail here.

[0115] Based on the principle of setting the discharge allocation ratio, the energy storage discharge efficiency, unit discharge cost, and allowable discharge power ratio of each resource aggregation area are weighted and fused to obtain the corresponding regional energy storage discharge power allocation ratio; the principle of setting the discharge allocation ratio includes that the regional energy storage discharge power allocation ratio is directly proportional to the energy storage discharge efficiency and the allowable discharge power ratio, and inversely proportional to the unit discharge cost.

[0116]

[0117] In the formula, This represents the regional energy storage discharge power allocation ratio within the resource aggregation region q; and These represent the unit discharge cost of the resource aggregation region q and the corresponding reference value for the unit discharge cost, respectively. and These represent the energy storage discharge efficiency and the percentage of allowable discharge power in the resource aggregation region q, respectively. , and This represents the weighting coefficients. The magnitude of each weighting coefficient can be adjusted according to actual application needs, as long as it meets the following requirements: This ensures the reliability and fairness of the energy storage discharge power allocation ratio.

[0118] Similarly, in practical applications, the corresponding regional energy storage discharge power control target can be obtained by multiplying the energy storage discharge power allocation ratio and the corresponding energy storage charge and discharge power control value (actually the energy storage discharge power control value). However, considering that the obtained regional energy storage discharge power control target may exceed the corresponding upper limit of the allowable discharge power, in order to ensure the rationality of the regional energy storage discharge power control target setting, this embodiment preferably determines whether the regional energy storage discharge power control target of each resource aggregation area is greater than the corresponding upper limit of the allowable discharge power after obtaining the regional energy storage discharge power control target of each resource aggregation area. If so, the discharge power of the regional energy storage discharge power control target that exceeds the upper limit of the allowable discharge power is allocated to other resource aggregation areas with control potential, so as to ensure that the regional energy storage discharge power control target of each resource aggregation area is within the corresponding allowable control range.

[0119] Based on the regional control objectives of each resource aggregation region, the regional objectives are decomposed using a pre-defined regional resource scheduling optimization model to obtain the resource control objectives of each flexible resource within the corresponding resource aggregation region. The pre-defined regional resource scheduling optimization model can be understood as an optimization model used to rationally allocate the regional control objectives of each resource aggregation region to each flexible resource within the resource aggregation region according to a certain optimization objective. To ensure low-cost, high-resource-utilization virtual power plant resource collaborative optimization and improve peak-shaving and valley-filling benefits while also ensuring grid operation stability, this embodiment preferably constructs a pre-defined regional resource scheduling optimization model with the optimization objective of balancing control costs and grid operation risks. The specific objective function is expressed as follows:

[0120]

[0121] In the formula,

[0122]

[0123]

[0124] in, and These represent the regional regulation cost and the regional voltage over-limit risk cost, respectively. and Represents the weighting coefficient, and It can be set according to actual application needs; This represents the amount of electricity abandoned by new energy source i; This represents the reduction amount of the adjustable load k; and This indicates the charging and discharging power of the energy storage device l; This represents the unit cost of curtailed electricity for new energy source i; This represents the unit compensation cost for reducing the adjustable load k; and This represents the unit cost of charging and the unit cost of discharging for energy storage device l; and These represent the positive and negative voltage offsets at node m, respectively, both being absolute deviation values; This represents the unit risk cost of the voltage deviation at node m; , , and These represent the number of new energy sources, the number of adjustable loads, the number of energy storage devices, and the number of nodes within the aggregation area, respectively.

[0125] To ensure the efficiency and reliability of obtaining resource control targets for each flexible resource within the resource aggregation area based on the resource scheduling optimization model within the aforementioned preset area, this embodiment preferably includes regional total power balance constraints, new energy output constraints, adjustable load operation constraints, energy storage device operation constraints, and grid voltage risk constraints, specifically expressed as follows:

[0126] 1) Regional total power balance constraint, expressed as:

[0127]

[0128] In the formula, This indicates the actual dispatch output of new energy source i; , and These represent the regional new energy production control target, the regional load demand control target, and the regional energy storage charging and discharging power control target, respectively (positive value is discharging, negative value is charging, and zero value indicates no use).

[0129] 2) Constraints on new energy output, expressed as:

[0130]

[0131] in, The predicted output of new energy i can be obtained based on the existing new energy output prediction mechanism of virtual power plants, which will not be described in detail here. This indicates the maximum amount of electricity that new energy source i can forgo, which can be set according to the actual application scenario.

[0132] 3) Adjustable load operation constraints, expressed as:

[0133]

[0134] in, This represents the maximum reduction of the adjustable load k, which can be set according to the actual application scenario.

[0135] 4) Operating constraints of energy storage devices, expressed as:

[0136]

[0137] In the formula, This indicates the charging and discharging state of the energy storage device l, where 0 represents charging and 1 represents discharging; and These represent the maximum charging power and maximum discharging power of the energy storage device l, respectively, and can be set according to the actual application scenario; , , and These represent the state of charge, initial state of charge, minimum state of charge, and maximum state of charge of energy storage device l, respectively. Indicates the duration of the regulatory cycle; and These represent the charging efficiency and discharging efficiency of energy storage device l, respectively. This indicates the rated capacity of energy storage device l.

[0138] 5) Grid voltage risk constraints can be set based on existing grid operation safety conditions, and are expressed as:

[0139]

[0140]

[0141]

[0142]

[0143]

[0144]

[0145]

[0146] In the formula, This represents the voltage change at node m; This represents the initial voltage at node m; and These represent the minimum and maximum allowable voltages for node m, respectively, and can be set according to the actual application scenario. This represents the sensitivity of node n to changes in injected power at node m. The change in injected power at node n represents the sensitivity and the change in injected power. Both can be calculated using existing technologies, which will not be detailed here. and These represent the upper and lower voltage limits for node m, respectively. A value of 1 indicates that the limit has been exceeded. Represents a large constant, a relatively large and fixed constant used to indicate linearization when limits are exceeded.

[0147] It should be noted that the resource regulation targets for each flexible resource within the resource aggregation area vary depending on the type of flexible resource. The resource regulation target for new energy resources is the new energy output regulation target (power output regulation target), the resource regulation target for adjustable load resources is the load demand regulation target, and the resource regulation target for energy storage resources is the energy storage charging and discharging power regulation target. The specific regulation targets for each flexible resource can be obtained by efficiently solving the resource scheduling optimization model within the aforementioned preset area using existing commercial solvers (such as CPLEX, Gurobi, etc.) based on the actual resource operation data of the resource aggregation area. The specific solution process will not be detailed here. This embodiment allocates regulation targets to each resource aggregation area by designing a resource scheduling optimization model within a preset area with the optimization objective of balancing new energy consumption, regulation costs, and grid operation risks. This not only achieves efficient collaborative optimization of multiple flexible resources with low cost and high resource utilization, improving the peak shaving and valley filling benefits of virtual power plants, but also ensures the safety and stability of grid operation during the regulation process.

[0148] This invention provides virtual power plant supply and demand time-series data, including renewable energy supply time-series data and load demand time-series data, which are acquired periodically. The supply and demand fluctuation characteristics of this virtual power plant time-series data are analyzed to generate corresponding supply and demand fluctuation time-series data. Based on this data, a multi-scenario control strategy library, including pre-constructed baseline feature vectors for various fluctuation scenarios and scenario baseline control strategies, is used for scenario matching analysis to obtain an initial scenario control strategy. The initial scenario control strategy is then subjected to scenario adaptability simulation analysis. Based on the simulation analysis results, the initial scenario control strategy is optimized to obtain a target scenario control strategy, which is then applied to the target scenario. The control objectives in the control strategy are decomposed into hierarchical and partitioned control objectives. Based on these hierarchical and partitioned control objectives, a technical solution for virtual power plant control in corresponding cycles is implemented. The virtual power plant control design, which combines a smart scenario control strategy acquisition mechanism based on supply and demand fluctuations with a scenario adaptability analysis mechanism and a hierarchical and partitioned decomposition mechanism for control objectives, can not only achieve efficient interaction between source, load and storage in virtual power plants under complex and uncertain environments, and improve the grid's peak shaving and valley filling capabilities, but also effectively smooth out supply and demand fluctuations, ensure stable grid operation, and significantly improve the collaborative optimization capability and operational resilience of virtual power plants. This provides reliable technical support for the power supply needs of virtual power plants and has good engineering application prospects.

[0149] It should be noted that although the steps in the flowchart above are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order requirement for the execution of these steps, and they can be executed in other orders.

[0150] In one embodiment, such as Figure 2As shown, a multi-scenario collaborative control system for virtual power plants aimed at ensuring power supply is provided. The virtual power plant includes multiple resource aggregation areas, and each resource aggregation area includes various flexible resources. The system includes:

[0151] The fluctuation analysis module 1 is used to periodically acquire virtual power plant supply and demand time series data, and to perform supply and demand fluctuation characteristic analysis on the virtual power plant supply and demand time series data to generate corresponding supply and demand fluctuation time series data; the virtual power plant supply and demand time series data includes new energy supply time series data and load demand time series data;

[0152] The scenario matching module 2 is used to perform scenario matching analysis based on the supply and demand fluctuation time series data and a pre-built multi-scenario control strategy library to obtain the corresponding initial scenario control strategy. The multi-scenario control strategy library includes supply and demand fluctuation benchmark feature vectors and scenario benchmark control strategies under various fluctuation scenarios. The scenario benchmark control strategy includes control parameters, control constraints, and control targets.

[0153] The strategy acquisition module 3 is used to perform scene adaptability simulation analysis on the initial scene control strategy, and optimize the initial scene control strategy according to the corresponding simulation analysis results to obtain the target scene control strategy.

[0154] The target decomposition module 4 is used to decompose the control targets in the target scenario control strategy into hierarchical and partitioned control targets to obtain hierarchical and partitioned control targets, and to execute virtual power plant control for the corresponding period according to the hierarchical and partitioned control targets.

[0155] Specific limitations regarding the multi-scenario collaborative control system of virtual power plants for power supply security can be found in the limitations of the multi-scenario collaborative control method of virtual power plants for power supply security described above. The corresponding technical effects can be obtained equivalently and will not be repeated here. Each module in the aforementioned multi-scenario collaborative control system of virtual power plants for power supply security can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0156] In summary, the present invention provides a method and system for multi-scenario collaborative control of virtual power plants for power supply security. Based on an intelligent scenario control strategy acquisition mechanism for supply and demand fluctuations, combined with a scenario adaptability analysis mechanism and a hierarchical and partitioned decomposition mechanism for control targets, the virtual power plant control design not only enables efficient interaction between source, load, and storage in complex and uncertain environments, improving the grid's peak-shaving and valley-filling capabilities, but also effectively smooths supply and demand fluctuations, ensuring stable grid operation. This significantly enhances the collaborative optimization capabilities and operational resilience of virtual power plants, providing reliable technical support for the power supply security needs of virtual power plants and demonstrating promising engineering application prospects.

[0157] The various embodiments in this specification are described in a progressive manner. For directly identical or similar parts of the embodiments, refer to each other. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. It should be noted that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0158] The above-described embodiments are merely preferred embodiments of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various improvements and substitutions without departing from the principles of the present invention, and these improvements and substitutions should also be considered within the scope of protection of the present invention. Therefore, the scope of protection of this invention should be determined by the scope of the claims.

Claims

1. A method for multi-scenario collaborative control of virtual power plants for power supply security, characterized in that, The virtual power plant comprises multiple resource aggregation regions, which include various flexible resources. The method includes: The system periodically acquires virtual power plant supply and demand time-series data, analyzes the supply and demand fluctuation characteristics of the virtual power plant supply and demand time-series data, and generates corresponding supply and demand fluctuation time-series data. The virtual power plant supply and demand time-series data includes new energy supply time-series data and load demand time-series data. The supply and demand fluctuation time-series data includes energy supply fluctuation time-series data and load demand fluctuation time-series data. Based on the time-series data of supply and demand fluctuations, scenario matching analysis is performed on a pre-built multi-scenario control strategy library to obtain the corresponding initial scenario control strategy. The multi-scenario control strategy library includes benchmark feature vectors of supply and demand fluctuations under various fluctuation scenarios and benchmark control strategies for each scenario. The benchmark control strategy for each scenario includes control parameters, control constraints, and control objectives. The initial scenario control strategy is subjected to scenario adaptive simulation analysis, and the initial scenario control strategy is optimized based on the corresponding simulation analysis results to obtain the target scenario control strategy. The scenario adaptive simulation analysis is based on a virtual power plant dedicated simulation tool to construct a virtual simulation environment that includes new energy sources, loads, energy storage and grid operation constraints, and to verify the fluctuation scenario control effect of the initial scenario control strategy through simulation. The control objectives in the target scenario control strategy are decomposed into hierarchical and partitioned control objectives to obtain hierarchical and partitioned control objectives, and virtual power plant control is executed according to the hierarchical and partitioned control objectives for the corresponding period. The step of obtaining the corresponding initial scenario control strategy by performing scenario matching analysis based on the supply and demand fluctuation time series data and a pre-built multi-scenario control strategy library includes: The corresponding fluctuation features are extracted from the time series data of energy supply fluctuation and the time series data of load demand fluctuation, respectively, to generate corresponding energy supply fluctuation features and load demand fluctuation features. Based on the energy supply fluctuation characteristics and the load demand fluctuation characteristics, corresponding energy supply fluctuation feature vectors and load demand fluctuation feature vectors are generated respectively. The energy supply fluctuation feature vector and the load demand fluctuation feature vector are concatenated to obtain the supply and demand fluctuation feature vector. The similarity analysis is performed between the supply and demand fluctuation feature vector and the supply and demand fluctuation benchmark feature vector corresponding to each scenario control strategy in the multi-scenario regulation strategy library to obtain the corresponding scenario similarity value. Based on all the scene similarity values, several relevant scene control strategies are obtained based on a preset similarity threshold; The scene similarity values ​​corresponding to each of the relevant scene control strategies are normalized to obtain the corresponding similarity weights. Based on the similarity weights of the various related scenario control strategies, all the related scenario control strategies are weighted and fused to obtain the initial scenario control strategy.

2. The method for multi-scenario collaborative control of virtual power plants for power supply assurance as described in claim 1, characterized in that, The construction steps of the multi-scenario control strategy library include: Source, load and storage information is extracted from the historical operation data of the virtual power plant to obtain the power plant's historical energy production sequence, power plant's historical load demand sequence and power plant's energy storage charging and discharging power sequence. According to the preset time period, the power plant's historical energy production sequence, the power plant's historical load demand sequence, and the power plant's energy storage charging and discharging power sequence are divided into time periods to obtain the corresponding sets of energy production subsequences, load demand subsequences, and energy storage charging and discharging power subsequences. Fluctuation features are extracted from each subsequence in the energy output subsequence set, the load demand subsequence set, and the energy storage charging and discharging power subsequence set to obtain the corresponding energy fluctuation feature dataset, load fluctuation feature dataset, and energy storage charging and discharging power fluctuation feature dataset. Cluster analysis was performed on the energy fluctuation feature dataset, the load fluctuation feature dataset, and the energy storage charging and discharging power fluctuation feature dataset to obtain multiple source-load-storage interaction scenarios. The energy fluctuation characteristic data and load fluctuation characteristic data of each of the source-load-storage interaction scenarios are spliced ​​together to obtain the corresponding supply and demand fluctuation benchmark feature vector. Based on the historical control strategies of each of the source-load-storage interaction scenarios, the corresponding scenario benchmark control strategy is obtained. The supply and demand fluctuation benchmark feature vectors and scenario benchmark control strategies of each source-load-storage interaction scenario are summarized to obtain the multi-scenario control strategy library.

3. The method for multi-scenario collaborative control of virtual power plants for power supply assurance as described in claim 2, characterized in that, The virtual power plant's historical operation data includes the historical sequence of regional energy production, regional load demand, and regional energy storage charging and discharging power for each resource aggregation area. The steps of extracting source-load-storage information from the historical operation data of the virtual power plant to obtain the historical energy production sequence of the power plant, the historical load demand sequence of the power plant, and the energy storage charging and discharging power sequence of the power plant include: The historical sequences of regional energy production, regional load demand, and regional energy storage charging and discharging power in each resource aggregation region are sequentially cleaned and time-aligned to obtain the corresponding regional energy production sequence, regional load demand sequence, and regional energy storage charging and discharging power sequence to be analyzed. The energy production sequences of all resource aggregation regions to be analyzed are summarized according to the sampling time to obtain the corresponding historical energy production sequences of power plants. The load demand sequences of all resource aggregation regions to be analyzed are summarized according to the sampling time to obtain the corresponding historical load demand sequences of power plants. The energy storage charging and discharging power sequences of each resource aggregation region to be analyzed are weighted and summarized with the corresponding regional energy storage capacity ratio as the weight to obtain the corresponding power plant energy storage charging and discharging power sequence.

4. The method for multi-scenario collaborative control of virtual power plants for power supply assurance as described in claim 1, characterized in that, The control parameters include new energy control parameters, energy storage control parameters, and load control parameters; The steps of performing scene adaptability simulation analysis on the initial scene control strategy and optimizing the initial scene control strategy based on the corresponding simulation analysis results to obtain the target scene control strategy include: Based on the initial scenario control strategy and the virtual power plant supply and demand time series data, a virtual power plant strategy scheduling simulation is performed to obtain the corresponding simulation operation data. Based on the simulation data, corresponding scenario adaptability evaluation indicators are obtained; the scenario adaptability evaluation indicators include supply and demand balance error rate, response delay time, regulation resource utilization rate, and over-limit frequency; the regulation resource utilization rate includes new energy utilization rate, load utilization rate, and energy storage utilization rate. The scene adaptability evaluation indexes are compared and analyzed with the corresponding scene adaptability evaluation index thresholds to obtain the corresponding scene adaptability evaluation results. Obtain the scenario adaptability assessment index items that do not meet the standards in the scenario adaptability assessment results, and obtain the control parameters to be optimized from the control parameters based on the preset correlation analysis algorithm; Based on preset adjustment rules, each of the control parameters to be optimized in the control parameters is adjusted to obtain the target scene control strategy.

5. The method for multi-scenario collaborative control of virtual power plants for power supply assurance as described in claim 1, characterized in that, The control targets include control values ​​for new energy production, control values ​​for load demand, and control values ​​for energy storage charging and discharging power. The hierarchical and regional control objectives include the regional control objectives of each resource aggregation area and the resource control objectives of each flexible resource within each resource aggregation area. The step of decomposing the control targets in the target scene control strategy into hierarchical and partitioned control targets includes: The control targets in the target scenario control strategy are hierarchically split based on the preset regional target allocation principle to obtain the regional control targets of each resource aggregation region. Based on the regional regulation objectives of each resource aggregation area, the regional objectives are decomposed according to the resource scheduling optimization model within the preset area to obtain the resource regulation objectives of each flexible resource within the corresponding resource aggregation area; the resource scheduling optimization model within the preset area is constructed with the optimization objective of balancing new energy consumption, regulation costs and grid operation risks.

6. The method for multi-scenario collaborative control of virtual power plants for power supply assurance as described in claim 5, characterized in that, The regional regulation targets include regional new energy production regulation targets, regional load demand regulation targets, and regional energy storage charging and discharging power regulation targets. The step of splitting the control targets in the target scenario control strategy into hierarchical targets based on a preset regional target allocation principle to obtain the regional control targets of each resource aggregation region includes: Based on the proportion of new energy installed capacity and current output capacity of each resource aggregation region, the new energy production regulation and allocation ratio of the corresponding resource aggregation region is obtained, and based on the new energy production regulation and allocation ratio and the new energy production regulation value, the regional new energy production regulation target of the corresponding resource aggregation region is obtained. Based on the load capacity and preset control potential coefficient of different control levels in each resource aggregation area, the regional load control potential is summarized and analyzed to obtain the corresponding regional load control potential ratio. Based on the regional load control potential ratio and the load demand control value, the corresponding regional load demand control target is obtained. Based on the energy storage power regulation type of the energy storage charging and discharging power regulation value, the energy storage charging and discharging power allocation ratio of each resource aggregation area is obtained, and based on the energy storage charging and discharging power allocation ratio and the energy storage charging and discharging power regulation value, the corresponding regional energy storage charging and discharging power regulation target is obtained.

7. The method for multi-scenario collaborative control of virtual power plants for power supply assurance as described in claim 6, characterized in that, The step of obtaining the energy storage charging and discharging power allocation ratio of each resource aggregation area based on the energy storage power regulation type of the energy storage charging and discharging power regulation value includes: When the energy storage power regulation type is energy storage charging power regulation, the energy storage charging efficiency, unit charging cost and allowable charging power ratio of each resource aggregation area are obtained respectively. Based on the principle of setting the charging allocation ratio, the energy storage charging efficiency, unit charging cost, and allowable charging power ratio of each resource aggregation area are weighted and fused to obtain the corresponding regional energy storage charging power allocation ratio. The principle of setting the charging allocation ratio includes that the regional energy storage charging power allocation ratio is directly proportional to the energy storage charging efficiency and the allowable charging power ratio, and the regional energy storage charging power allocation ratio is inversely proportional to the unit charging cost.

8. The method for multi-scenario collaborative control of virtual power plants for power supply assurance as described in claim 6, characterized in that, The step of obtaining the energy storage charging and discharging power allocation ratio of each resource aggregation area based on the energy storage power regulation type of the energy storage charging and discharging power regulation value further includes: When the energy storage charging and discharging power regulation value is the energy storage discharging power regulation value, the energy storage discharging efficiency, unit discharging cost and allowable discharging power ratio of each resource aggregation area are obtained respectively. Based on the principle of setting the discharge allocation ratio, the energy storage discharge efficiency, unit discharge cost, and allowable discharge power ratio of each resource aggregation area are weighted and fused to obtain the corresponding regional energy storage discharge power allocation ratio. The principle of setting the discharge allocation ratio includes that the regional energy storage discharge power allocation ratio is directly proportional to the energy storage discharge efficiency and the allowable discharge power ratio, and the regional energy storage discharge power allocation ratio is inversely proportional to the unit discharge cost.

9. A virtual power plant multi-scenario collaborative control system for power supply security, characterized in that, The virtual power plant comprises multiple resource aggregation areas, each including various flexible resources. Applying the multi-scenario collaborative control method for virtual power plants oriented towards power supply security as described in claim 1, the system includes: The fluctuation analysis module is used to periodically acquire virtual power plant supply and demand time-series data, and to analyze the supply and demand fluctuation characteristics of the virtual power plant supply and demand time-series data to generate corresponding supply and demand fluctuation time-series data; the virtual power plant supply and demand time-series data includes new energy supply time-series data and load demand time-series data; The scenario matching module is used to perform scenario matching analysis based on the supply and demand fluctuation time series data and a pre-built multi-scenario control strategy library to obtain the corresponding initial scenario control strategy. The multi-scenario control strategy library includes supply and demand fluctuation benchmark feature vectors and scenario benchmark control strategies under various fluctuation scenarios. The scenario benchmark control strategy includes control parameters, control constraints, and control targets. The strategy acquisition module is used to perform scene adaptability simulation analysis on the initial scene control strategy, and optimize the initial scene control strategy according to the corresponding simulation analysis results to obtain the target scene control strategy. The target decomposition module is used to decompose the control targets in the target scenario control strategy into hierarchical and partitioned control targets to obtain hierarchical and partitioned control targets, and to execute virtual power plant control for the corresponding period according to the hierarchical and partitioned control targets.

Images