A resource aggregation correction updating and effect evaluation method for a virtual power plant

Abstract

The present application relates to the technical field of virtual power plant operation control and resource aggregation analysis in new power systems, in particular to a resource aggregation correction updating and effect evaluation method for virtual power plants, which comprises obtaining real-time operation state information, prediction information and safety risk information of various resources in the virtual power plant, and establishing resource operation boundaries; according to resource state quantity, price signal, prediction deviation and disturbance risk, the operation boundaries are online corrected; according to the corrected boundaries, the feasible output set of various resources is constructed, the feasible region is aggregated, the available space is calculated by combining the current reference power to adjust the available space up and down, and the global aggregated feasible region is updated; according to the two-way adjustment space and the aggregated output dynamic response process, the equivalent adjustment capacity, response speed and stability index are calculated to complete the aggregation effect evaluation. The present application can realize dynamic characterization of virtual power plant aggregation adjustment capacity, real-time updating of feasible region and quantitative evaluation of aggregation effect.

Description

Technical Field

[0001] This invention relates to the field of virtual power plant operation control and resource aggregation analysis technology in novel power systems, and particularly to a method for resource aggregation correction, updating, and effect evaluation for virtual power plants. Background Technology

[0002] With the continuous expansion of distributed power sources, energy storage devices, and adjustable loads in new power systems, virtual power plants can enhance the system's flexible regulation capabilities and participate in ancillary services such as frequency regulation and peak shaving by aggregating and controlling various types of dispersed resources. Existing virtual power plant resource aggregation methods mostly establish aggregation models based on resource rated parameters or static boundaries to describe the regulation capabilities of aggregated resources.

[0003] However, in actual operation, the adjustability of various resources changes dynamically with factors such as energy storage state of charge, renewable energy forecasting bias, price signals, and operational risks. Traditional static aggregation methods cannot accurately reflect the real-time availability boundary of resources, easily leading to inconsistencies between aggregation results and actual execution capabilities. At the same time, existing aggregation effect assessments typically focus on capacity size, lacking a comprehensive characterization of regulation response speed and output stability, making it difficult to fully reflect the regulation performance of virtual power plants in actual operating scenarios.

[0004] Therefore, it is necessary to propose an online correction method for resource aggregation, a feasible domain update method, and an effect evaluation method for virtual power plants, so as to achieve dynamic correction, real-time update, and comprehensive evaluation of the aggregation and regulation capabilities of virtual power plants. Summary of the Invention

[0005] To overcome the problems raised in the background art, the present invention provides a method for resource aggregation, correction, updating, and effect evaluation for virtual power plants, comprising the following steps:

[0006] S1. Collect the operational data of resources within the virtual power plant, and establish the initial operational boundaries of the resources based on their physical and security constraints.

[0007] S2. Based on the resource's operational data, the initial operational boundary is corrected in real time to obtain the resource's real-time operational boundary;

[0008] S3. Based on real-time operational boundaries, construct a feasible set of resource outputs;

[0009] S4. Perform Minkowski sum operation on the feasible output set of resources to obtain the global aggregate feasible region of the virtual power plant;

[0010] S5. Collect the current baseline power of the virtual power plant, combine it with the global aggregated feasible region, calculate the upward and downward available space of the virtual power plant, and update the global aggregated feasible region when the real-time operating boundary of the resource changes.

[0011] S6. Calculate the equivalent regulating capacity, response speed index, and stability index of the aggregated virtual power plant to complete the effect evaluation of virtual power plant resource aggregation.

[0012] Preferably, in step S1, the resources include distributed new energy power generation resources, energy storage resources, controllable generating units, and demand response resources;

[0013] Operational data includes real-time operational data and predicted operational data. Real-time operational data includes current status parameters, external price signals, and system safety risk indicators. Predictive operational data includes output prediction deviation.

[0014] The operational data is preprocessed, including imputation of missing values, data standardization, and extraction of data closely related to carbon emissions through grey relational analysis or correlation coefficient screening.

[0015] Preferably, in step S2, the correction includes: adopting a boundary shrinkage mechanism based on prediction confidence intervals for distributed new energy power generation resources;

[0016] An adaptive boundary correction mechanism based on the state of charge (SOC) is adopted for energy storage resources: when the SOC is higher than the preset upper limit threshold, the charging boundary shrinks; when the SOC is lower than the preset lower limit threshold, the discharging boundary shrinks.

[0017] For controllable generating units, a boundary boosting mechanism based on price signals is adopted: the real-time electricity price signal is mapped to the unit output boundary increment through an S-shaped price incentive function, and the boundary increment is superimposed on the initial operating boundary of the unit.

[0018] A dynamic boundary adjustment mechanism based on user participation intentions is adopted for demand response resources.

[0019] Preferably, in step S4, the global aggregation feasible region expression is: ;

[0020] in, Let be the global aggregate feasible region of the virtual power plant at time t. The total number of all resource categories, For the first The set of feasible outputs of resource class at time t: ;

[0021] in, , The first The upper and lower boundaries of the output of the resource class after online correction at time t. For the first The output of a resource at time t, with positive and negative values ​​representing discharging and charging, respectively. It is the set of real numbers.

[0022] Preferably, in step S5, the available space is adjusted upward to the remaining adjustment space from the current reference power to the upper boundary of the global aggregated feasible region, and the available space is adjusted downward to the remaining adjustment space from the current reference power to the lower boundary of the global aggregated feasible region; the global aggregated feasible region is updated online using a support function or half-space representation; for demand response resources with cross-time period coupling constraints, the feasible output set is updated online using a rolling time window method.

[0023] Preferably, in step S6, the equivalent adjustment capacity is the smaller of the available upward adjustment space and the available downward adjustment space; the response speed index includes: the response time for the aggregated output to reach 90% of the target amplitude under the sudden adjustment command, and the maximum rate of change of the aggregated output between adjacent sampling times under the continuous adjustment signal; the stability index is the variance of the aggregated output power.

[0024] The beneficial effects of this invention are as follows:

[0025] This invention incorporates energy storage state of charge, price signals, prediction bias, and disturbance risks into the resource boundary correction process, enabling online dynamic updates of the adjustable boundaries of various resources within the virtual power plant. This improves the resource aggregation model's ability to track actual operating conditions. Furthermore, by constructing feasible output sets for each resource and implementing online updates of the aggregation feasible domain based on Minkowski dynamics, this invention can characterize the virtual power plant's bidirectional regulation capability near the current operating point in real time, improving the accuracy and timeliness of the aggregation regulation capability description. Finally, by introducing equivalent regulation capacity, response speed indicators, and stability indicators, this invention comprehensively evaluates the resource aggregation effect of the virtual power plant. This not only reflects the expansion of the regulation space after aggregation but also characterizes the dynamic response characteristics and operational stability of the aggregation output, thus providing a basis for virtual power plant scheduling, control decisions, and aggregation effect analysis. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the overall process of an embodiment of the present invention. Detailed Implementation

[0027] This invention provides a method for resource aggregation correction and update, and effect evaluation for virtual power plants, applicable to virtual power plant aggregation scenarios consisting of distributed new energy power generation resources, energy storage resources, controllable generating units, and demand response resources. The virtual power plant employs a collaborative model between the information layer and the physical layer. The information layer is responsible for real-time state perception, short-term prediction, boundary correction, aggregation feasible region calculation, and effect evaluation, while the physical layer is responsible for executing power regulation of various resources. During operation, the information layer continuously receives feedback from the physical layer and updates resource state estimates and capacity boundaries when the executed response deviates from the target, maintaining consistency between the model and actual execution capabilities.

[0028] like Figure 1 As shown, the method includes the following steps:

[0029] S1. Collect real-time operating status data and predicted operating data of various adjustable resources in the virtual power plant, and establish the initial operating boundaries of various resources at time t based on the physical constraints and safety constraints of various resources.

[0030] The collected data is preprocessed, including imputing missing values, standardizing the data, and extracting features closely related to carbon emissions through grey relational analysis or correlation coefficient screening as data input for the next step.

[0031] Adjustable resources include at least distributed new energy power generation resources, energy storage resources, controllable generating units, and demand response resources.

[0032] Real-time operational status data includes distributed renewable energy real-time output, energy storage state of charge (SOC), current output of controllable units, current load adjustment of demand response resources, grid connection point switching power, and safety risk indicators. Safety risk indicators include at least grid connection point switching power constraints, power flow safety status, voltage deviation tolerance, and extreme disturbance risk indicators.

[0033] The forecasted operational data includes predicted output values ​​for new energy sources in the future and predicted load change trends.

[0034] S2. Based on the current state parameters of resources, external price signals, output prediction deviations, and safety risk indicators, the initial operating boundary is dynamically and in real-time corrected online to obtain the real-time operating boundary of various resources at time t.

[0035] A boundary contraction mechanism based on prediction confidence intervals is adopted for distributed renewable energy power generation resources: the output of photovoltaic and wind power is subject to uncertainty due to weather conditions. The output range is first defined by prediction confidence intervals; when the actual output deviates from the predicted value and exceeds the interval, the output operation boundary is automatically contracted to avoid misjudgment of regulation capacity due to renewable energy fluctuations and to ensure operational safety.

[0036] An adaptive boundary correction mechanism based on State of Charge (SOC) is adopted for energy storage resources: when the SOC is higher than the preset upper limit threshold, the charging boundary is contracted; when the SOC is lower than the preset lower limit threshold, the discharging boundary is contracted. This is to limit the charging output of energy storage resources when the SOC is too high and limit the discharging output when the SOC is too low. The SOC adaptive correction of the operating boundary always matches the real-time safety state.

[0037] A boundary enhancement mechanism based on price signals is adopted for controllable generating units: the real-time electricity price signal is mapped to the unit output boundary increment through the S-shaped price incentive function, and the boundary increment is superimposed on the initial operating boundary of the unit to enhance the unit's callability during high electricity price periods.

[0038] A dynamic boundary adjustment mechanism based on user participation intention is adopted for demand response resources: if users have a high willingness to participate in load adjustment, the boundary is widened and more adjustable space is released; if the willingness is low, the boundary is narrowed.

[0039] When the power flow exceeds the limit, voltage deviation, or the risk of extreme disturbance increases, the operating boundaries of various resources are simultaneously contracted according to the preset safety contraction coefficient.

[0040] S3. Based on the real-time operating boundary, construct the feasible output set of various resources at time t, i.e., the individual feasible domain, which is used to characterize the adjustable power range of such resources.

[0041] For energy storage resources, let the state variables of electrical or thermal energy storage be uniformly denoted as St, and its adjustable power is determined by the charging and discharging directions. To describe the online feasible region of energy storage at time t, the set of feasible outputs of the energy storage resource at time t is defined. Represented as: ;

[0042] in, , These are the upper and lower boundaries of the actual output of the energy storage resource at time t (derived from step S1). , The upper and lower boundaries of the energy storage resource output at time t after online correction (derived from step S2). Let SOC be the energy storage resource at time t. and It is not a fixed constant, but a safety boundary that varies with the state of charge (SOC) of the energy storage. When the SOC is too high, it limits continued charging; when the SOC is too low, it limits continued discharging, in order to avoid saturation or exceeding the limit during frequency regulation.

[0043] For renewable power generation resources and controllable generating units, this embodiment adopts a unified boundary representation method. For resources such as photovoltaic and wind power that are affected by weather conditions, their boundaries are mainly given by predicted output and confidence intervals; for controllable generating units such as gas turbines, their boundaries are not only constrained by the upper limit of static output, but also related to external price signals or opportunity costs. Therefore, the online boundary of the unit is written as the sum of the original boundary and the state-cost coupling correction.

[0044] For demand response resources, this embodiment considers three types of load adjustment: reduction, transfer, and shift. Let the load adjustment amount be... The feasible region constraint is used to characterize the load-side boundary, which not only constrains the adjustment amplitude at a single moment, but also ensures the conservation of total energy before and after load transfer through zero-sum conservation constraints. For shiftable and transferable loads with cross-time coupling characteristics, linear constraints are used to represent them as multidimensional polyhedra within the rolling time window, and they are projected onto the single-moment adjustment space at the current moment during subsequent aggregation updates.

[0045] Further implement online boundary correction.

[0046] First, implement SOC-based adaptive boundary contraction or relaxation for energy storage resources. Let... Given the SOC at time t, and upper and lower safety thresholds. , Constructing a smooth contraction factor: ; ;

[0047] in, , These are the charging and discharging smoothing contraction factors, respectively. , These are the upper and lower safety threshold coefficients, respectively.

[0048] A continuous mapping correction is applied to the energy storage boundary to ensure the continuity and interpretability of boundary changes. The corrected online boundary of energy storage can be represented as: ; ;

[0049] in, , These represent the upper and lower boundaries of the energy storage resource's operating output at time t after online correction.

[0050] When the State of Charge (SOC) approaches the upper threshold, the adjustable boundary of the charging direction is reduced to prevent overcharging; when the SOC approaches the lower threshold, the adjustable boundary of the discharging direction is reduced to prevent over-discharging. This method prevents the energy storage from losing its subsequent regulatory capability due to maximum charging and discharging during frequency regulation.

[0051] Secondly, a price-driven boundary escalation mechanism is implemented for controllable generating units. In this embodiment, a price incentive function is introduced, and an S-shaped function is used to map the external price signal to the unit boundary increment, resulting in: ;

[0052] in, Let g be the upper bound of the output of the g-th controllable unit at time t after price incentive adjustment. The external price signal at time t (electricity market price, ancillary service price, or dispatch incentive price signal). This is the price incentive inflection point threshold (the midpoint price of the S-shaped incentive function). The slope coefficient of the S-shaped function. This represents the maximum output increment that price incentives bring to controllable generating units.

[0053] Implement safety correction boundaries and define risk indicators to address prediction bias and extreme disturbance risks. , used to reflect the The power flow, voltage, or grid connection security risks corresponding to a resource at the current moment are assessed. Using risk indicators, the adjustable boundary is safely contracted to obtain the following: ;

[0054] in, Let this be the upper / lower bound of the adjustable output force of the k-th resource at time t before safe contraction. Let this be the upper / lower bound of the adjustable output force of the k-th resource at time t after safe contraction. The safety shrinkage coefficient. When As the price rises, the boundary automatically contracts, forcing the adjustment strategy back to the conservative feasible region; when When the price decreases, the boundary recovers, releasing the regulatory potential.

[0055] After completing the above boundary correction, the corrected boundary is written back to the feasible output set of each resource. For any resource... At any given moment The feasible set of outputs is written as: ;

[0056] in, , The first The upper and lower boundaries of the output of the resource class after online correction at time t. For the first The output of a resource at time t, with positive and negative values ​​representing discharging and charging, respectively. It is the set of real numbers.

[0057] S4. Perform Minkowski sum operations on the individual feasible regions of various resources, considering only the reachability of net power, to obtain the global aggregate feasible region of the virtual power plant at time t. : ;

[0058] in, The total number of all resource categories, For the first The set of feasible outputs for each type of resource at time t. If each If the interval is defined as an interval, the aggregation result will still be an interval, and the endpoints of the interval will be equal to the sum of the endpoints. If the mutual exclusion constraints of energy storage charging and discharging and the continuity constraints of demand response are considered, the feasible region of aggregation will be represented as a polyhedron or its projection.

[0059] To avoid rebuilding the convex hull every time the boundary changes, the global aggregate feasible region is updated online using a support function or half-space representation. Taking advantage of the Minkowski principle and the property of the additivity of the support function, when the boundary correction of any resource causes a change in its individual feasible region, only the incremental part of the aggregate feasible region corresponding to that resource is updated, without updating the global aggregate feasible region.

[0060] S5. Collect the current base power of the virtual power plant, combine it with the global aggregated feasible region, calculate the up-adjustment available space and down-adjustment available space of the virtual power plant at time t, and update the global aggregated feasible region online when the real-time operating boundary of any resource changes.

[0061] Upward adjustment of available space at time t Current reference power Reduce the available space to the remaining adjustment space at the upper boundary of the global aggregate feasible domain. Current reference power The remaining adjustment space up to the lower boundary of the global aggregation feasible region.

[0062] Let the upper and lower boundaries of the aggregate feasible region at time t in the net power dimension be respectively... and ,but: ; ;

[0063] When SOC, price, or risk triggers boundary contraction, the and The system will be synchronously reduced to enable real-time updates of the available space for virtual power plant regulation, avoiding misjudgments of frequency regulation command reachability. For demand response resources coupled across time periods, at each new time, the rolling time window will be advanced by one time period, the feasible set for unexecuted time periods will be recalculated, and projected onto the current time regulation space to balance cross-time period coupling constraints with single-time control docking requirements.

[0064] For demand response resources with cross-time period coupling constraints, a rolling time window method is used to update the feasible output set online, and the updated result is projected into the adjustment space at the current moment.

[0065] S6. Based on the adjustment of available space, the adjustment of available space, and the dynamic response process of aggregated output, calculate the equivalent regulation capacity, response speed index and stability index after aggregation, and complete the effect evaluation of virtual power plant resource aggregation.

[0066] Equivalent regulation capacity at time t The expression for adjusting the available space by the smaller of the upward and downward adjustments is: ;

[0067] in, , These represent the adjustable available space above and below time t, respectively. This definition avoids miscalculating a one-sided space as a two-way available capability and better reflects the true adjustable space near the current operating point.

[0068] For individual resources that are not aggregated, before aggregation, the space around their own baseline power is still defined within the set of individual resources: ; ;

[0069] in, Let be the adjustment space of the k-th type of resource at time t, which represents the adjustable power range from the resource's own base power to the upper boundary of the feasible output. Let be the adjustment space for the k-th type of resource at time t, representing the adjustable power range from the resource's own base power to the lower boundary of feasible output.

[0070] The corresponding unpolymerized equivalent capacity is: ;

[0071] The capacity change resulting from polymerization is expressed as a relative quantity: ;

[0072] in, The equivalent adjustment capacity relative change rate characterizes the magnitude of change in the equivalent capacity after aggregation relative to the total unaggregated capacity. This indicator is used to characterize whether aggregation expands the adjustable space through coordinated scheduling or simply redistributes the original capacity. When boundary contraction occurs, the aggregated capacity will decrease synchronously, reflecting the early removal of unsustainable capacity rather than simple capacity degradation.

[0073] The response speed indicators include: the response time for the aggregated output to reach 90% of the target amplitude under a sudden adjustment command issued by the power grid dispatch center, and the maximum rate of change of the aggregated output between adjacent sampling times under a continuous adjustment signal from the real-time power grid dispatch center (the ratio of the power change amplitude of the aggregated output power between adjacent sampling times to the time step). Under a sudden command, the shorter the time to reach 90% of the target amplitude, the better the rapid response capability; under a continuous adjustment signal, the larger the maximum rate of change between adjacent times, the stronger the continuous tracking capability. A comprehensive judgment is made on whether the power grid frequency regulation speed requirements are met. Let the frequency regulation command be... (The frequency regulation target regulation power issued by the grid to the virtual power plant at time t), the net regulation output achieved by the aggregator within the current feasible region is For mutated instructions, the time of instruction change is used as the reference. Starting from this point, the shortest time for the aggregated output to reach 90% of the target amplitude is defined as the response time: ;

[0074] in, To reach 90% of the target amplitude, For 90% amplitude response time, for The virtual power plant's net regulation output power at all times. When the instruction changes The virtual power plant's baseline actual output power; this time describes the time scale from the triggering of the adjustment action to its basic completion, and can reflect the degree of matching between weight allocation and resource dynamic capabilities.

[0075] For a continuously changing adjustment signal, let the sampling time step be... Define the maximum rate of change of the aggregate output between adjacent sampling times as: ;

[0076] in, for The virtual power plant's net regulated output power at any given moment; this metric characterizes the dynamic achievability of the aggregated output under rapidly changing signals. A higher rate of achievability should be observed when the fast resource carries more high-frequency components.

[0077] The stability metric is the variance of the aggregated output power (the sum of the output of all resources): ;

[0078] in, The time sequence number of the sampling point within the sliding window. This represents the total number of sampling points within the sliding window. All within the sliding window The time-series average of the aggregated output power at each sampling point; the smaller the variance, the smoother the output and the higher the operational stability; a large variance indicates large output fluctuations, which can easily lead to grid safety risks. When the boundary is not corrected in advance, saturation after space exhaustion will introduce discontinuous changes, making... Increase; boundary correction reduces such abrupt changes by shrinking the feasible region in advance. Capacity structure and dynamic behavior are intrinsically linked here.

[0079] Excellent performance across all three indicators indicates good aggregation effect, strong adjustability, fast response, and stable operation; high capacity but slow speed or poor stability indicates suitability for peak shaving but not for rapid frequency regulation; fast speed but small capacity indicates suitability for short-term high-frequency regulation but not for large-scale peak shaving. The aggregation effect evaluation results are used for virtual power plants to participate in bidding for ancillary services in the electricity market, optimize the allocation of dispatch instructions, and dynamically adjust resource aggregation strategies.

[0080] This implementation method first collects real-time operation and forecast data of distributed new energy sources, energy storage, controllable generating units, and demand response resources to achieve resource aggregation. Then, it establishes initial operating boundaries based on physical and safety constraints and makes online corrections for different resources: energy storage adaptively shrinks / relaxes charging and discharging boundaries based on SOC, and controllable generating units increase output boundaries during high electricity price periods through price incentive functions, while shrinking boundaries by combining forecast bias and safety risks. Subsequently, it constructs feasible output sets for each resource, obtains the aggregated feasible region through Minkowski sum calculation, and achieves local online updates using support functions / half-space representation. Finally, it completes the effect evaluation from three dimensions: equivalent regulation capacity, response speed, and stability.

[0081] This invention has been described through embodiments. Those skilled in the art will understand that various changes or equivalent substitutions can be made to these features and embodiments without departing from the spirit and scope of the invention. Furthermore, under the teachings of this invention, these features and embodiments can be modified to adapt to specific situations and materials without departing from the spirit and scope of the invention. Therefore, this invention is not limited to the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the protection scope of this invention.

Claims

1. A method for resource aggregation, modification, updating, and effect evaluation for virtual power plants, characterized in that, Includes the following steps: S1. Collect the operational data of resources within the virtual power plant, and establish the initial operational boundaries of the resources based on their physical and security constraints. The resources include distributed new energy power generation resources, energy storage resources, controllable generating units, and demand response resources; S2. Based on the resource's operational data, the initial operational boundary is corrected in real time to obtain the resource's real-time operational boundary; The correction includes adopting a boundary shrinkage mechanism based on prediction confidence intervals for distributed new energy power generation resources; An adaptive boundary correction mechanism based on the state of charge (SOC) is adopted for energy storage resources. When the SOC is higher than the preset upper limit threshold, the charging boundary shrinks; when the SOC is lower than the preset lower limit threshold, the discharging boundary shrinks. For controllable generating units, a boundary boosting mechanism based on price signals is adopted. The real-time electricity price signal is mapped to the unit output boundary increment through an S-shaped price incentive function, and the boundary increment is superimposed on the initial operating boundary of the unit. A dynamic boundary adjustment mechanism based on user participation intentions is adopted for demand response resources; S3. Based on real-time operational boundaries, construct a feasible set of resource outputs; S4. Perform Minkowski sum operation on the feasible output set of resources to obtain the global aggregate feasible region of the virtual power plant; S5. Collect the current baseline power of the virtual power plant, combine it with the global aggregated feasible region, calculate the upward and downward available space of the virtual power plant, and update the global aggregated feasible region when the real-time operating boundary of the resource changes. S6. Calculate the equivalent regulating capacity, response speed index, and stability index of the aggregated virtual power plant to complete the effect evaluation of virtual power plant resource aggregation.

2. The method for resource aggregation, correction, updating, and effect evaluation for virtual power plants according to claim 1, characterized in that, In step S1, the operating data includes real-time operating data and predicted operating data. The real-time operating data includes current status parameters, external price signals, and system safety risk indicators. The predicted operating data includes output prediction deviation. The operational data is preprocessed, including imputation of missing values, data standardization, and extraction of data closely related to carbon emissions through grey relational analysis or correlation coefficient screening.

3. The method for resource aggregation, correction, updating, and effect evaluation for virtual power plants according to claim 1, characterized in that, In step S4, the global aggregation feasible region expression is: ； in, Let be the global aggregate feasible region of the virtual power plant at time t. The total number of all resource categories, For the first The set of feasible outputs of resource class at time t: ； in, , The first The upper and lower boundaries of the output of the resource class after online correction at time t. For the first The output of a resource at time t, with positive and negative values ​​representing discharging and charging, respectively. It is the set of real numbers.

4. The method for resource aggregation, correction, updating, and effect evaluation for virtual power plants according to claim 1, characterized in that, In step S5, the upward adjustment available space is the remaining adjustment space from the current reference power to the upper boundary of the global aggregated feasible region, and the downward adjustment available space is the remaining adjustment space from the current reference power to the lower boundary of the global aggregated feasible region; the global aggregated feasible region is updated online using a support function or half-space representation; for demand response resources with cross-time period coupling constraints, the feasible output set is updated online using a rolling time window method.

5. The method for resource aggregation, correction, updating, and effect evaluation for virtual power plants according to claim 1, characterized in that, In step S6, the equivalent adjustment capacity is the smaller of the available space for upward adjustment and the available space for downward adjustment; the response speed index includes: the response time for the aggregated output to reach 90% of the target amplitude under the sudden adjustment command, and the maximum rate of change of the aggregated output between adjacent sampling times under the continuous adjustment signal; the stability index is the variance of the aggregated output power.

Images