Cluster regulation capacity feasible region aggregation method based on similar feasible region regularization

Abstract

The application provides a cluster regulation capacity feasible region aggregation method based on similar feasible region regularization, and belongs to the technical field of distributed power supply.The technical scheme is as follows: S1, flexible resource regulation feasible region modeling;S2, construction of feasible region similarity quantization index;S3, setting of the benchmark model by considering the difference of device regulation capacity;S4, feasible region approximation regularization based on the benchmark model;S5, calculation of the overall aggregation feasible region of the cluster.The beneficial effects of the application are as follows: the application can take into account the feasible region boundary characteristics of each adjustable resource and the difference of device regulation capacity, effectively improve the overall feasible region approximation accuracy of the device, and is beneficial to fully tap the regulation potential of the flexible resource and enhance the ability of the flexible resource cluster to participate in the optimal dispatch of the distribution network.

Description

Technical Field

[0001] This invention relates to the field of distributed power technology, and in particular to a method for aggregation of feasible regions based on similar feasible region regularization to improve the flexibility of resource cluster adjustment capabilities. Background Technology

[0002] While there are numerous available flexible resources within a power grid cluster, the supply capacity of a single flexible device is limited, and directly obtaining information from each flexible resource to participate in regional power grid regulation is challenging. Therefore, aggregating diverse distributed flexible resources is the primary method for participating in distribution network regulation. To fully utilize the adjustability of flexible resources within the cluster and reduce the difficulty of the cluster's participation in distribution network regulation, it is necessary to characterize the overall regulation capability of the flexible resource cluster.

[0003] To address the issues of low efficiency and long processing time associated with direct aggregation, existing research typically employs models such as the Chino polyhedron, hyperellipsoid, and fixed polyhedron to characterize the feasible region of flexible devices. However, due to the diverse types of flexible resources and the significant differences in their characteristics, their feasible regions also exhibit considerable variations. Existing methods tend to overlook the morphological characteristics of the resource's feasible region itself, leading to a reduction in aggregation accuracy.

[0004] How to solve the above-mentioned technical problems is the challenge facing this invention. Summary of the Invention

[0005] The purpose of this invention is to provide a feasible domain aggregation method for the adjustment capability of flexible resource clusters based on similar feasible domain regularization. The method uses an adjacent time period feasible domain description method to segment and characterize the feasible domain within the entire scheduling cycle. By setting a benchmark model and internal approximate regularization for the adjacent time period feasible domains of similar equipment, the method achieves efficient aggregation of the feasible domain of cluster adjustment capability, effectively improves the overall feasible domain approximation accuracy of equipment, and is conducive to fully exploring the adjustment potential of flexible resources and enhancing the ability of flexible resource clusters to participate in the optimal scheduling of distribution networks.

[0006] To achieve the aforementioned objectives, the present invention employs the following technical solution: a feasible region aggregation method for flexible resource cluster adjustment capabilities based on similar feasible region regularization, comprising the following steps:

[0007] S1: Modeling the feasible region for flexible resource adjustment;

[0008] S2: Construct a similarity metric for feasible regions;

[0009] S3: Set the benchmark model considering the differences in equipment adjustment capabilities;

[0010] S4: Approximate regularization within the feasible region based on the baseline model;

[0011] S5: Overall aggregated feasible region of the computing cluster.

[0012] Preferably, the flexibility resources mentioned in step S1 include gas generator sets, battery energy storage devices, and electric vehicles; the mathematical model for the feasible region of the gas generator set's adjustment capability is as follows:

[0013] P g,min ≤P g,t ≤P g,max

[0014]

[0015] In the formula, P g,t P represents the output power of the gas generator set during time period t. g,max and P g,min These represent the upper and lower limits of the output power of the gas generator set; P g,t-1 Let ν represent the power output of the gas generator set during the time period t-1, where Δt is the duration of a single time period within the time scale of interest, and ν is the power output of the gas generator set during the time period t-1. g,up and ν g,down These are the equipment's maximum upward and downward ramp rates, respectively.

[0016] The mathematical model for the feasible region of the battery energy storage device's regulation capability is as follows:

[0017] -P es,dc,max <P es,t ≤P es,c,max

[0018] E es,min ≤E es,t ≤E es,max

[0019]

[0020] In the formula, P es,t P represents the active power output of the energy storage device during time period t. es,c,max and P es,dc,max These are the maximum charging and discharging power of the energy storage device, respectively; E es,max and E es,min The upper and lower limits of the allowable energy storage capacity; E es,t and E es,t-1 Let η represent the energy state of the energy storage device during time period t and time period t-1, respectively. c and η dc These represent the charging and discharging efficiency of the energy storage device.

[0021] The feasible domain model for adjusting the charging power of the electric vehicle is as follows:

[0022] -P ev,dc,max ≤Pev,t ≤P ev,c,max

[0023] E ev,min ≤E ev,t ≤E ev,max

[0024]

[0025] E0+E ev,c,sum -E ev,dc,sum ≥E P

[0026]

[0027] In the formula, P ev,t P represents the charging and discharging load of the electric vehicle during time period t. ev,c,max and P ev,dc,max These are the maximum charging and discharging power of the electric vehicle, respectively; E ev,t and E ev,t-1 E represents the electric vehicle's state of charge during time period t and time period t-1, respectively. ev,max and E ev,min η represents the upper and lower limits of the allowable battery capacity based on battery health considerations. ev,c and η ev,dc These represent the charging efficiency and discharging efficiency of the charging pile, respectively; E0 is the initial energy level of the electric vehicle when connected to the grid, E... ev,c,sum and E ev,dc,sum These represent the total charging and discharging power of the electric vehicle during grid connection, respectively. P The expected battery capacity set for electric vehicle owners; φ c and φ dc These are sets of all charging and discharging periods, respectively.

[0028] Preferably, in step S2, the calculation formula for the feasible region similarity metric is as follows:

[0029]

[0030] In the formula, d i,j Let be the similarity distance between objects i and j, and let matrix N be the key parameter matrix affecting the morphology of the feasible region of the equipment. n is the dimension of parameter matrix N. Morphological features are the sole criterion for similarity clustering. To avoid the influence of spatial location and area size on the judgment of shape similarity, normalized relative parameters can be used to construct vectors representing the characteristics of each equipment's feasible region. For power generation equipment, the calculation formulas for its various parameters are as follows:

[0031]

[0032] In the formula, νu％ ν d％ and P 0％ These represent the relative upward ramp rate, relative downward ramp rate, and relative initial output of the equipment, respectively. min and P max These are the lower and upper limits of the device's output power, respectively. up and ν down These represent the maximum upward and downward ramp rates of the equipment, respectively, and P0 is the initial output power of the equipment.

[0033] Similarly, for energy storage devices, their relative parameters can be set as follows:

[0034]

[0035] In the formula, E 0％ ν c％ and ν dc％ These represent the relative initial energy level, relative charging rate, and relative discharging rate of the energy storage device, respectively. min and E max These represent the maximum and minimum allowable power levels for device operation, P. c and P dc These are the device's maximum charging power and maximum discharging power, respectively.

[0036] Preferably, in step S3, to improve clustering accuracy and reduce the loss of flexibility during the clustering process, the approximate accuracy of devices with stronger adjustability should be given priority. Therefore, the weights are set to account for the differences in device adjustability, thereby obtaining a benchmark model that takes into account both the overall boundary characteristics of the feasible domain of the devices and the differences in individual adjustability. The expression of the optimization objective function S for benchmark selection is as follows:

[0037]

[0038] In the formula, φ s R is the set of all baseline nodes. i R is the output adjustment range of device i. max It is the maximum adjustment range of the equipment in the system.

[0039] Preferably, in step S4, the feasible region of adjacent time periods of the device is approximated and normalized to obtain a device feasible region representation based on the benchmark model. Taking device n as an example... b Taking the feasible region as a baseline model, let's assume device n b The feasible region centroid is represented by the set of feasible region vertices after being moved to the origin, and by the device n b The device n1 for the baseline model is characterized as follows:

[0040] φ nb ={(xb1 ,y b1 ),…,(x bm ,y bm )}

[0041]

[0042] In the formula, φ nb For device n b The feasible domain vertex set, m is the device n b The number of feasible domain vertices; φ n1 Let be the set of feasible domain vertices of device n1, and let a be the number of feasible domain vertices of device n1.

[0043] The feasible region information of device n1 after internal approximate normalization can be obtained through the centroid coordinates (x n1 ,y n1 and scaling factor k n1 To characterize, where the scaling factor refers to the change in the centroid of the reference feasible region when it moves to coordinate (x... n1 ,y n1 After that, the maximum magnification that can be achieved within the feasible region of device n1, and the centroid coordinates (x) n1 ,y n1 The formula for solving ) is as follows:

[0044]

[0045] Scaling factor k n1 The size of can be transformed into a simple optimization problem to be solved. The optimization solution model is as follows:

[0046]

[0047] In the formula, Z n1 Let (x) be the feasible region of device n1. n1 +k n1 x bi ,y n1 +k n1 x bi ) represents the coordinates of vertex i of the normalized feasible region.

[0048] Preferably, the formula for calculating the overall feasible region of the cluster in step S5 is as follows:

[0049]

[0050] In the formula, Z sum Let K be the overall aggregate feasible region of the cluster, and K be the number of baseline models for the cluster. to This represents the aggregated feasible region for each baseline model device group.

[0051] The vertex set representation of the aggregate feasible region of the device group can be obtained through simple calculation based on the scaling factor and centroid coordinates, using device n. b The vertex set of the aggregated feasible region of each device, based on the feasible region, is represented as:

[0052]

[0053] In the formula, ∑k nj The sum of scaling factors for all devices determines the final feasible domain size, (∑x nj ,∑y nj This determines the location of the aggregated feasible region in geometric space.

[0054] Compared with the prior art, the beneficial effects of the present invention are:

[0055] 1. The feasible region similarity metric proposed in this invention, based on the Euclidean distance of relative parameter vectors, not only ensures good similarity calculation results between devices with similar original parameters, but also ensures similarity calculation results between devices with large differences in original parameters but similar feasible region boundary shapes, thereby improving the accuracy of device feasible region similarity metric.

[0056] 2. The cluster regulation capability aggregation method based on similar feasible region regularization proposed in this invention can take into account the feasible region boundary characteristics of each adjustable resource and the differences in equipment regulation capability, effectively improving the overall feasible region approximation accuracy of the equipment, which is conducive to fully tapping the regulation potential of flexible resources and enhancing the ability of flexible resource clusters to participate in the optimal scheduling of distribution networks.

[0057] 3. The cluster aggregation adjustment capability characterization method based on the power feasible domain of adjacent time periods proposed in this invention realizes the visualization of the cluster's adjustable range and flexibility adjustment margin, providing theoretical support for the subsequent participation of flexible resource clusters in distribution network optimization scheduling. Attached Figure Description

[0058] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0059] Figure 1 This is a flowchart of the feasible region solution for cluster adjustment capability based on similar feasible region regularization in this invention.

[0060] Figure 2 This is a schematic diagram showing the parameter distribution of the 30 battery energy storage devices in this invention.

[0061] Figure 3 This is a schematic diagram of the clustering results of similar feasible regions for the battery energy storage device in this invention.

[0062] Figure 4 This is a schematic diagram comparing the approximate accuracy of the aggregate feasible region under different benchmark setting schemes in this invention. Detailed Implementation

[0063] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. Of course, the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0064] Example 1

[0065] Figure 1 The flowchart for solving the feasible region of cluster adjustment capability based on similar feasible region regularization is shown in this embodiment. The flexibility resources include gas generator sets, battery energy storage devices and electric vehicles.

[0066] The mathematical model for the feasible region of the gas generator set's regulation capability is as follows:

[0067] P g,min ≤P g,t ≤P g,max

[0068]

[0069] In the formula, P g,t P represents the output power of the gas generator set during time period t. g,max and P g,min These represent the upper and lower limits of the output power of the gas generator set; P g,t-1 Let ν represent the power output of the gas generator set during the time period t-1, where Δt is the duration of a single time period within the time scale of interest, and ν is the power output of the gas generator set during the time period t-1. g,up and ν g,down These are the equipment's maximum upward and downward ramp rates, respectively.

[0070] The mathematical model for the feasible region of the regulation capability of battery energy storage devices is as follows:

[0071] -P es,dc,max <P es,t ≤P es,c,max

[0072] E es,min ≤E es,t ≤E es,max

[0073]

[0074] In the formula, P es,t P represents the active power output of the energy storage device during time period t. es,c,max and P es,dc,maxThese are the maximum charging and discharging power of the energy storage device, respectively; E es,max and E es,min The upper and lower limits of the allowable energy storage capacity; E es,t and E es,t-1 Let η represent the energy state of the energy storage device during time period t and time period t-1, respectively. c and η dc These represent the charging and discharging efficiency of the energy storage device.

[0075] The feasible region model for adjusting the charging power of electric vehicles is as follows:

[0076] -P ev,dc,max ≤P ev,t ≤P ev,c,max

[0077] E ev,min ≤E ev,t ≤E ev,max

[0078]

[0079] E0+E ev,c,sum -E ev,dc,sum ≥E P

[0080]

[0081] In the formula, P ev,t P represents the charging and discharging load of the electric vehicle during time period t. ev,c,max and P ev,dc,max These are the maximum charging and discharging power of the electric vehicle, respectively; E ev,t and E ev,t-1 E represents the electric vehicle's state of charge during time period t and time period t-1, respectively. ev,max and E ev,min η represents the upper and lower limits of the allowable battery capacity based on battery health considerations. ev,c and η ev,dc These represent the charging efficiency and discharging efficiency of the charging pile, respectively; E0 is the initial energy level of the electric vehicle when connected to the grid, E... ev,c,sum and E ev,dc,sum These represent the total charging and discharging power of the electric vehicle during grid connection, respectively. P The expected battery capacity set for electric vehicle owners; φ c and φ dc These are sets of all charging and discharging periods, respectively.

[0082] The formula for calculating the feasible region similarity metric is as follows:

[0083]

[0084] In the formula, d i,j Let be the similarity distance between objects i and j, and let matrix N be the key parameter matrix affecting the morphology of the feasible region of the equipment, where n is the dimension of parameter matrix N. Morphological features are the sole criterion for similarity clustering. To avoid the influence of spatial location and area size on the judgment of shape similarity, normalized relative parameters can be used to construct vectors representing the characteristics of each equipment's feasible region. For power generation equipment, the calculation formulas for its various parameters are as follows:

[0085]

[0086] In the formula, ν u％ ν d％ and P 0％ These represent the relative upward ramp rate, relative downward ramp rate, and relative initial output of the equipment, respectively. min and P max These are the lower and upper limits of the device's output power, respectively. up and ν down These represent the maximum upward and downward ramp rates of the equipment, respectively, and P0 is the initial output power of the equipment.

[0087] Similarly, for energy storage devices, their relative parameters can be set as follows:

[0088]

[0089] In the formula, E 0％ ν c％ and ν dc％ These represent the relative initial energy level, relative charging rate, and relative discharging rate of the energy storage device, respectively. min and E max These represent the maximum and minimum allowable power levels for device operation, P. c and P dc These are the device's maximum charging power and maximum discharging power, respectively.

[0090] The expression for the optimization objective function S set by the benchmark is as follows:

[0091]

[0092] In the formula, φ s R is the set of all baseline nodes. i R is the output adjustment range of device i. max It is the maximum adjustment range of the equipment in the system.

[0093] With device n b Taking the feasible region as a baseline model, let's assume device n bThe feasible region centroid is represented by the set of feasible region vertices after being moved to the origin, and by the device n b The device n1 for the baseline model is characterized as follows:

[0094] φ nb ={(x b1 ,y b1 ),…,(x bm ,y bm )}

[0095]

[0096] In the formula, φ nb For device n b The feasible domain vertex set, m is the device n b The number of feasible domain vertices; Let be the set of feasible domain vertices of device n1, and let a be the number of feasible domain vertices of device n1.

[0097] The feasible region information of device n1 after internal approximate normalization can be obtained through the centroid coordinates (x n1 ,y n1 and scaling factor k n1 To characterize, where the scaling factor refers to the change in the centroid of the reference feasible region when it moves to coordinate (x... n1 ,y n1 After that, the maximum magnification that can be achieved within the feasible region of device n1, and the centroid coordinates (x) n1 ,y n1 The formula for solving ) is as follows:

[0098]

[0099] Scaling factor k n1 The size of can be transformed into a simple optimization problem to be solved. The optimization solution model is as follows:

[0100]

[0101] In the formula, Z n1 Let (x) be the feasible region of device n1. n1 +k n1 x bi ,y n1 +k n1 x bi ) represents the coordinates of vertex i of the normalized feasible region.

[0102] The formula for calculating the overall feasible region of the cluster is as follows:

[0103]

[0104] In the formula, Z sumLet K be the overall aggregate feasible region of the cluster, and K be the number of baseline models for the cluster. to This represents the aggregated feasible region for each baseline model device group.

[0105] The vertex set representation of the aggregate feasible region of the device group can be obtained through simple calculation based on the scaling factor and centroid coordinates, using device n. b The vertex set of the aggregated feasible region of each device, based on the feasible region of the baseline, is represented as:

[0106]

[0107] In the formula, ∑k nj The sum of scaling factors for all devices determines the final feasible domain size, (∑x nj ,∑y nj This determines the location of the aggregated feasible region in geometric space.

[0108] Example 2

[0109] To verify the effectiveness of the proposed cluster adjustment capability aggregation method based on similar feasible region regularization, the aggregation method proposed in this invention is used to solve the overall aggregation feasible region of 30 battery energy storage devices.

[0110] Figure 2 This shows the parameter distribution of the initial charge and relative charge / discharge rate of each energy storage device.

[0111] Figure 3 This section defines the set of energy storage devices and its corresponding benchmark when the number of clusters in the similar feasible region is 3. In this embodiment, the relative charge and discharge rates of the energy storage devices are relatively concentrated, so the differences in the relative charge and discharge rates of the selected benchmark devices are not significant. The devices are mainly divided based on the differences in their relative electrical capacity. Figure 2 The relative charge levels of the three benchmark devices are approximately 20%, 50%, and 70%, respectively, representing three states where the devices have strong charging capabilities, relatively balanced charging and discharging capabilities, and strong discharging capabilities.

[0112] Figure 4 A comparison of the approximate accuracy of the aggregate feasible region under different benchmark setting schemes.

[0113] In this embodiment, to verify the performance of the proposed feasible region aggregation method, three schemes were used to select the feasible region benchmark for the energy storage device set:

[0114] Q1: Select the baseline using the original basic parameters of the equipment;

[0115] Q2: Select the feasible region baseline using relative parameters;

[0116] Q3: The benchmark model setting method considering device weights proposed in this invention is adopted;

[0117] Based on the above different benchmark models, the feasible region of individual equipment is approximated, and the feasible region of the entire equipment is finally obtained. The comparison of the approximation accuracy under different numbers of benchmarks and schemes is as follows: Figure 4 As shown in the figure, the overall approximate accuracy of the feasible region increases significantly with the increase in the number of benchmarks. Finding feasible region benchmarks that can adapt to individual device differences, and thus using multiple benchmarks for feasible region aggregation, can significantly improve the approximate accuracy of the aggregation. Comparing three different benchmark selection methods, it can be seen that compared to scheme Q1 based on original parameters, schemes Q2 and Q3, which use relative parameters to describe the characteristics of the device's feasible region, have a significant advantage in approximate accuracy, with accuracy improvements of 3.9%, 5.4%, and 8.8% or more, respectively, under different numbers of benchmarks. This indicates that relative parameters perform better in describing the characteristics of the device's feasible region, and the selected feasible region benchmarks are more representative, effectively improving the overall approximate accuracy. Scheme Q3 introduces device weights to interfere with the direction of benchmark selection based on the benchmarks using relative parameters, mainly to improve the approximate accuracy of feasible regions for devices with strong adjustment capabilities. In the single-benchmark scenario of this embodiment, scheme Q3 does not show an advantage over scheme Q2 because the feasible region benchmarks selected by both are similar. In two-benchmark and three-benchmark scenarios, Scheme Q3 improves the approximate accuracy by approximately 2.7% and 1.3% compared to Scheme Q2, respectively, demonstrating the effectiveness of Scheme Q3 proposed in this invention in improving the approximate accuracy of the aggregation results. In summary, the benchmark model setting method considering the differences in equipment adjustment capabilities proposed in this invention can effectively improve the approximate accuracy of the overall feasible region of the equipment, preserving the feasible region of individual equipment to the greatest extent possible.

[0118] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A feasible region aggregation method for flexible resource cluster adjustment capability based on similar feasible region regularization, characterized in that, Includes the following steps: S1: Modeling the feasible region for flexible resource adjustment; S2: Construct a similarity metric for feasible regions; S3: Set the benchmark model considering the differences in equipment adjustment capabilities; S4: Approximate regularization within the feasible region based on the baseline model; In step S4, the feasible region of adjacent time periods of the device is approximated and normalized to obtain the device feasible region representation based on the benchmark model, with device n as an example. b Using the feasible region as the baseline model, device n b The feasible region centroid is represented by the set of feasible region vertices after being moved to the origin, and by the device n b The device n1 for the baseline model is characterized as follows: ； ； In the formula, For device n b The feasible domain vertex set, m is the device n b The number of feasible domain vertices; Let be the set of vertices in the feasible region of device n1, and let a be the number of vertices in the feasible region of device n1. The feasible region information of device n1 after internal approximate normalization is obtained through centroid coordinates ( , and scaling factor k n1 To characterize, where the scaling factor refers to the change in the centroid of the reference feasible region when it moves to the coordinate system ( , After that, the maximum magnification reached within the feasible region of device n1, centroid coordinates ( , The formula for solving ) is as follows: ； Scaling factor k n1 The size of the problem is transformed into an optimization problem to be solved. The optimization solution model is as follows: ； In the formula, Let n1 be the feasible region. Let i be the coordinates of vertex i of the normalized feasible region; S5: Overall aggregated feasible region of the computing cluster.

2. The method for flexible resource cluster adjustment capability feasible region aggregation based on similar feasible region regularization according to claim 1, characterized in that, In step S1, the flexibility resources include gas generator sets, battery energy storage devices, and electric vehicles; The mathematical model for the feasible region of the gas generator set's regulation capability is as follows: ； ； In the formula, P g,t P represents the output power of the gas generator set during time period t. g,max and P g,min These represent the upper and lower limits of the output power of the gas generator set; P g,t-1 This represents the power output of the gas generator set during the time period t-1. The duration of a single time period within the timescale of interest. and These are the equipment's maximum upward ramp rate and downward ramp rate, respectively. The mathematical model for the feasible region of the battery energy storage device's regulation capability is as follows: ； ； ； In the formula, P es,t P represents the active power output of the energy storage device during time period t. es,c,max and P es,dc,max These are the maximum charging and discharging power of the energy storage device, respectively; E es,max and E es,min The upper and lower limits of the allowable energy storage capacity; E es,t and E es,t-1 These represent the electrical status of the energy storage device during time period t and time period t-1, respectively. and These refer to the charging and discharging efficiency of the energy storage device; The feasible domain model for adjusting the charging power of the electric vehicle is as follows: ； ； ； ； ； ； In the formula, The charging and discharging load of the electric vehicle during time period t. and These are the maximum charging and discharging power of the electric vehicle, respectively. and E represents the electric vehicle's state of charge during time period t and time period t-1, respectively. ev,max and E ev,min These are the upper and lower limits of the allowable battery capacity, taken into consideration for battery health. and These represent the charging efficiency and discharging efficiency of the charging pile, respectively; E0 represents the initial charge level of the electric vehicle when connected to the grid. and These represent the total charging and discharging power of the electric vehicle during grid connection, respectively. P The expected battery capacity set for electric vehicle owners; φ c and φ dc These are sets of all charging and discharging periods, respectively.

3. The feasible region aggregation method for flexible resource cluster adjustment capability based on similar feasible region regularization according to claim 1, characterized in that, In step S2, the calculation formula for the feasible region similarity metric is as follows: ； In the formula, d i,j Let be the similarity distance between objects i and j, and let matrix N be the key parameter matrix affecting the feasible region morphology of the equipment. Let n be the dimension of parameter matrix N. Normalized relative parameters are used to construct vectors representing the feasible region characteristics of each equipment. For power generation equipment, the calculation formulas for each parameter are as follows: ； In the formula, , as well as These represent the relative upward ramp rate, relative downward ramp rate, and relative initial output of the equipment, respectively. min and P max These are the lower and upper limits of the device's output power, respectively. and These are the equipment's maximum upward ramp rate and downward ramp rate, respectively. This refers to the initial output power of the device. For energy storage devices, the relative parameter settings are as follows: ； In the formula, , as well as These represent the relative initial energy level, relative charging rate, and relative discharging rate of the energy storage device, respectively. min and E max These represent the maximum and minimum allowable power levels for device operation, P. c and P dc These are the device's maximum charging power and maximum discharging power, respectively.

4. The method for flexible resource cluster adjustment capability feasible region aggregation based on similar feasible region regularization according to claim 1, characterized in that, In step S3, the expression for the optimization objective function S selected as the benchmark is as follows: ； In the formula, R is the set of all reference nodes. i R is the output adjustment range of device i. max It is the maximum adjustment range of the equipment in the system.

5. The feasible region aggregation method for flexible resource cluster adjustment capability based on similar feasible region regularization according to claim 1, characterized in that, In step S5, the formula for calculating the overall feasible region of the cluster is as follows: ； In the formula, Let K be the overall aggregate feasible region of the cluster, and K be the number of baseline models for the cluster. to For each baseline model device group, the aggregated feasible region is defined. The vertex set representation of the aggregated feasible region of the device group is obtained by calculation based on the scaling factor and centroid coordinates, using device n. b The vertex set of the aggregated feasible region of each device, based on the feasible region, is represented as: ； In the formula, The sum of scaling factors for all devices determines the final feasible domain size. This determines the location of the aggregated feasible region in geometric space.

Images