A method for modeling and aggregating the characteristics of flexible resource regulation

Abstract

This invention discloses a method for modeling and aggregating the adjustment characteristics of flexible resources. A resource aggregation platform collects the virtual battery model coefficient matrices reported by individual flexible resources, calculates the number of primitive polyhedral systems, and sends it to each individual. Each individual calculates its parameters, and the resource aggregation platform then calculates the final equivalent aggregate based on these parameters and sends it to the dispatch center for decision-making. Finally, the decision results are de-aggregated and distributed to each individual for execution. This invention ensures the consistency of group behavior; by controlling the final equivalent aggregate to prevent simultaneous charging and discharging, each individual can satisfy the constraints. This more accurately reflects the actual behavior of the energy storage system, improves the overall utilization efficiency of flexible resources, and enhances the power system's ability to cope with the volatility and uncertainty of renewable energy.

Description

Technical Field

[0001] This invention relates to the field of electrical engineering and automation, and in particular to a method for modeling and aggregating the characteristics of flexible resource regulation. Background Technology

[0002] As the penetration rate of renewable energy in the power system continues to increase, the role of flexible resources such as energy storage systems and electric vehicles in maintaining grid stability and reliability is becoming increasingly important. These resources provide crucial support for balancing electricity supply and demand, effectively mitigating the volatility and uncertainty brought about by renewable energy. However, in practical applications, effectively integrating and utilizing these flexible resources, especially regarding their regulation characteristics and aggregation methods, still faces many challenges.

[0003] Existing research emphasizes the importance of detailed modeling of the regulation characteristics of individual flexibility resources to optimize their collective performance within the power system. For example, some studies suggest that the charging and discharging efficiency of energy storage systems directly impacts the accurate prediction of their behavior and their potential contribution to grid service. Accurate modeling of the charging and discharging processes of energy storage systems can help power system operators better maintain grid stability and improve the absorption capacity of renewable energy. Furthermore, aggregating numerous flexibility resources into virtual power plants or similar structures can effectively improve their operational efficiency and economic feasibility. This aggregation approach integrates dispersed resources into a unified whole, making them more competitive in the electricity market.

[0004] Therefore, the configuration and control of distributed energy storage has become a key research focus. Some studies suggest that to achieve the optimal configuration strategy, the high cost of energy storage and the need to improve grid flexibility must be considered, which is crucial for the efficient integration of renewable energy. Other studies have proposed real-time clustering algorithms that can dynamically aggregate energy storage systems into virtual power plants, thereby improving operational efficiency and flexibility. These studies demonstrate that optimizing the configuration and control of energy storage systems can significantly improve the overall performance of the power system.

[0005] However, existing aggregation methods typically rely on polyhedral models to approximate the behavior of flexibility resources. These methods often neglect the non-simultaneous charging and discharging constraints that energy storage systems must adhere to during charging and discharging, leading to suboptimal or even infeasible solutions in practical applications. In other words, traditional methods often ignore the constraint that energy storage systems cannot charge and discharge simultaneously, which is impractical in real-world operations. Therefore, existing aggregation methods have significant limitations when dealing with energy storage systems.

[0006] Recent research has focused on developing more complex aggregation techniques. While these techniques have improved the overall utility of flexibility resources to some extent, they have not yet solved the problem of non-simultaneous charge and discharge constraints in energy storage systems. Therefore, there is an urgent need for a new approach that can ensure that each flexibility resource adheres to this key constraint during the aggregation process. Summary of the Invention

[0007] To address the technical problem that current methods for aggregating flexible resources neglect the constraint that they cannot be charged and discharged simultaneously, leading to suboptimal or even infeasible solutions, this invention provides a method for modeling and aggregating the adjustment characteristics of flexible resources that ensures each flexible resource adheres to the non-simultaneous charging and discharging constraint during the aggregation process.

[0008] To achieve the above-mentioned technical objectives, the technical solution of the present invention is as follows:

[0009] A method for modeling and aggregating the adjustment characteristics of flexible resources includes the following steps:

[0010] Step 1: The virtual battery model coefficient matrix reported by each individual flexible resource is collected by the resource aggregation platform of the same type. The category of the flexible resource to which each individual belongs is either energy storage system (ES) or electric vehicle (EV).

[0011] Step 2: The similar resource aggregation platform uses the coefficient matrix of each individual collected as the polyhedron representation, and then calculates the primitive polyhedron based on the maximum inner approximation algorithm of similar polyhedra, and sends the coefficients of the calculated primitive polyhedron to each individual.

[0012] Step 3: The resource aggregation platform receives the parameters calculated independently by each individual. The parameters are calculated by each individual according to the category of its own flexible resources and using the corresponding parameter calculation method.

[0013] Step 4: The resource aggregation platform calculates the final equivalent aggregate based on the parameters of each individual and sends it to the scheduling center for decision-making based on a preset decision algorithm, and then receives the decision result.

[0014] Step 5: The resource aggregation platform of the same type de-aggregates the decision results and finally distributes them to each individual for execution.

[0015] The method described above, in step 1, refers to the virtual battery model coefficient matrix reported by the i-th flexibility resource individual. for:

[0016]

[0017] Where K represents the category of the resource individual, K∈{ES,EV}, As decision variables, The coefficient matrix, Let τ represent the real number space, and τ be the maximum offline time interval.

[0018] In the method described above, step 1, the virtual battery model coefficient matrix is ​​obtained based on a virtual battery model issued by a similar resource aggregation platform. The virtual battery model is as follows:

[0019]

[0020] in, Indicates matrix transpose. This represents the column vector of charging power for the i-th individual. Let be the charging power of the i-th individual at time t. This represents the column vector of discharge power for the i-th individual. Let be the discharge power of the i-th individual at time t. This represents the upper bound of the charging power of the i-th individual, and T represents the offline time interval, T = {1, 2, ..., τ}. This represents the upper bound of the discharge power of the i-th individual. This represents the maximum downhill climbing speed of the i-th individual. This represents the power of the i-th individual at time t. This represents the maximum upward climbing speed of the i-th individual. This represents the lower bound of the energy value of the i-th individual at time t+1. This represents the energy value of the i-th individual at time t+1. This represents the upper bound of the energy value of the i-th individual at time t+1.

[0021] In the method described above, step 1, the virtual battery model is obtained by uniformly modeling based on the individual models of ES and EV, and then based on the adjustment characteristics.

[0022] The individual model of ES described in the method is as follows:

[0023]

[0024] in, Let be the charging power of the i-th ES at time t. Let be the discharge power of the i-th ES at time t. Let be the upper bound of the charging power of the i-th ES at time t. Let be the upper bound of the discharge power of the i-th ES at time t; Let be the external power of the i-th ES; Let i be the maximum downhill ramp rate of the i-th ES. The maximum upward climbing rate of the i-th ES; For the energy stored in the i-th ES, The self-discharge rate is denoted by Δt, and the time interval is Δt. Let i be the charging efficiency of the i-th ES. Let be the discharge efficiency of the i-th ES; Let i be the upper bound of the allowed energy for the i-th ES. This is the lower bound of the allowed energy for the i-th ES;

[0025] The individual model of EV is:

[0026] When the EV is charging

[0027]

[0028] in, Let be the charging power of the i-th EV at time t. Let be the discharge power of the i-th EV at time t; Let i be the upper bound of the charging power of the i-th EV. This is the upper bound of the discharge power of the i-th EV; Let be the external power of the i-th EV; For the maximum downhill climbing rate, The maximum uphill climbing rate; Let i be the energy of the i-th EV at time t. Let be the self-discharge rate of the i-th EV at time t, and Δt be the time interval; Let be the charging efficiency of the i-th EV at time t. Let be the discharge efficiency of the i-th EV at time t; Let i be the upper bound of the allowed energy for the i-th EV. Let i be the lower bound of the allowed energy for the i-th EV. Let i be the expected energy value of the i-th EV at time t+1;

[0029] When the EV is not charging

[0030]

[0031] The method described above, in step 2, includes calculating the primitive polyhedron:

[0032] Using the coefficient matrix of the virtual battery model As a polyhedral representation, the average value of each category of flexibility resources is taken. To be used as a basic polyhedron:

[0033]

[0034] in, for The average value, N K The number of individuals representing a certain type of flexible resource. for The average value, The decision variables represent the primitive polyhedrons;

[0035] Then similar resource aggregation platforms will and Send to each individual.

[0036] In the method described above, in step 3, when each individual calculates parameters independently, if the individual is ES, the parameters are calculated using the following formula:

[0037]

[0038] in Minimize st represents the constraint condition. Represents the scaling factor transformation coefficient, G represents a nonnegative matrix, and r i K Represents the translation factor transformation vector;

[0039] When the individual is EV, the parameters are calculated using the following formula:

[0040]

[0041] In step 3 of the method described above, the scaling factor and translation factor are converted through the following variable substitution:

[0042]

[0043] in, This represents the translation factor for the i-th individual. This represents the scaling factor of the i-th individual; and after the individual completes the calculation, it sends its own scaling factor or translation factor and scaling factor to the same resource aggregation platform.

[0044] In the method described above, in step 4, the resource aggregation platform calculates the final equivalent aggregate based on the parameters of each individual entity:

[0045]

[0046] The method described above, in step 5, involves the similar resource aggregation platform de-aggregating the decision results, including:

[0047] Similar resource aggregation platforms perform de-aggregation based on the following formula:

[0048]

[0049] in Scaling factor β K U represents the sum of scaling factors. K Represents aggregated decision variables. This represents the sum of translation factors. The translation factor is...

[0050] Then similar resource aggregation platforms will Distribute to each individual.

[0051] The technical advantages of this invention are as follows: First, by introducing non-simultaneous charging and discharging constraints, this invention can more accurately reflect the actual behavior of energy storage systems, improving the accuracy and reliability of the model. Second, by employing a translation-free approximation method, this invention ensures that each flexible resource meets key constraints, achieving optimal resource scheduling and utilization, thereby significantly improving the overall utilization efficiency of flexible resources and optimizing power system operation. Furthermore, by aggregating flexible resources, this invention can effectively balance the supply and demand relationship of the power system, alleviate the volatility and uncertainty brought by renewable energy, enhance the stability and reliability of the power grid, and improve the economic benefits of the system. This invention proposes a novel translation-free approximation aggregation method based on similar polyhedra for energy storage resources. By ensuring the consistency of group behavior, as long as the final equivalent aggregate is controlled to not charge and discharge simultaneously, each individual can be controlled to satisfy the constraints. This method can more accurately reflect the actual behavior of energy storage systems, improve the overall utilization efficiency of flexible resources, and enhance the power system's ability to cope with the volatility and uncertainty of renewable energy. Attached Figure Description

[0052] Figure 1 This is a flowchart illustrating an embodiment of the present invention.

[0053] Figure 2 The diagram shows the upper and lower bounds of power and energy for the corresponding energy storage units in the prior art; where (a) is a diagram of the upper bound of power and energy, and (b) is a diagram of the lower bound of power and energy.

[0054] Figure 3 The diagram shows the upper and lower bounds of power and energy of the corresponding energy storage unit in this invention; where (a) is a diagram of the upper bound of power and energy, and (b) is a diagram of the lower bound of power and energy. Detailed Implementation

[0055] In this embodiment, individual models are first created based on different categories of flexible resources (FRs), primarily including energy storage (ES) and electric vehicles (EVs). Let N be the number of individuals for a particular type of flexible resource. K Given that K∈{ES,EV} and the offline time interval is T={1,2,…,τ}, the models for these two types of flexibility resources are:

[0056] 1) ES:

[0057]

[0058] In the formula t∈T. Equation (1.a) represents the charging (discharging) power constraint of ES. Let be the charging (discharging) power at the i-th ES time t. This represents the upper limit of charging (discharging) power. Equation (1.b) indicates the external power of the ES individual. The equation relating charging and discharging power. Equation (1.c) represents the ramp rate constraint of the ES. These represent the maximum downward and upward climbing rates, respectively. Equation (1.d) expresses the equation relating ES energy and power. For the energy stored in ES, The self-discharge rate is denoted by Δt, and the time interval is Δt. The charging (discharging) efficiency is given by equation (1.e), which represents the energy constraint of the energy storage element (ES). These represent the upper and lower bounds of the allowed energy, respectively.

[0059] 2) EV:

[0060]

[0061]

[0062] In the formula t∈T. When the EV is not charging, i.e., t≤t i,arrive or t≥t i,leave At this time, the charging and discharging power can only be 0, as described by equation (2.a). When the EV is charging, equation (2.b) represents the EV charging (discharging) power constraint. Let be the charging (discharging) power at the i-th EV time t. This represents the upper limit of charging (discharging) power. Equation (2.c) indicates the external power of an individual EV. The equation relating charging and discharging power. Equation (2.d) represents the ramp rate constraint for the EV. These represent the maximum downward and upward ramp rates, respectively. Equation (2.e) expresses the equation relating EV energy and power. For energy, The self-discharge rate is denoted by Δt, and the time interval is Δt. The charging (discharging) efficiency is given by equation (2.f), which represents the energy constraint of the EV. These are the upper and lower bounds of the allowed energy, respectively, since EV leaves at time t. i,leave Its energy must be greater than or equal to Therefore, we also need to consider the time from t to t. i,leave All at maximum power When charging, Minimum energy required to convert to time t+1

[0063] Next, we will model the individual adjustment characteristics of flexible resources in a unified manner, and transform the individual model into an aggregated reference model, which is the virtual battery model.

[0064] The virtual battery model is described as follows:

[0065]

[0066] In the formula These are column vectors representing charge / discharge power. They describe the charge / discharge power constraints, ramp-up constraints, and energy state constraints. This can be further written in matrix form.

[0067]

[0068] In the formula: As decision variables, It is a coefficient matrix.

[0069] Then, aggregation is performed on each individual. Since the general Minkowski sum is an NP-hard problem, there is no exact algorithm with polynomial time complexity; only an approximate solution is possible. Similar FRs share similar characteristics, differing only in parameters. The main challenge in aggregation lies in the high computational cost due to the large number of individuals. Therefore, an efficient aggregation method supporting parallel computation is needed, and the maximum inner approximation method based on similar polyhedra can well meet this requirement.

[0070] As shown in equation (4), FR individuals are represented in the form of polyhedra. When selecting the primitive polyhedra for each type of flexibility resource, the average value of each type is usually taken:

[0071]

[0072] In the formula:

[0073] Each individual is approximated as a similar polyhedron of the primitive polyhedron:

[0074]

[0075] In the formula: It is a scaling factor. It is the translation factor.

[0076] Then the Minkowski sums of the same type of FR simplify to:

[0077]

[0078] By Use Ω K express, Substitution Constraints, Ω K This can be further expressed as:

[0079]

[0080] If similar polyhedra belong to the original polyhedra, that is This is an internal approximation. If the original polyhedron is a similar polyhedron, that is... This is considered an external approximation.

[0081] To ensure scheduling feasibility, an internal approximation method is used. Find... Maximum internal approximation included The mathematical expression is:

[0082]

[0083] Through variable substitution The above problem can be equivalent to solving a linear programming problem:

[0084]

[0085] Where: G≥0 means that all elements are non-negative.

[0086] This aggregation method supports parallel computing when the aggregation platform distributes data to individuals. and Afterwards, each individual can independently calculate their own corresponding... The results are then reported to the aggregation platform for simple summation, making it highly suitable for a large number of similar FR aggregations. Furthermore, this aggregation method inherently includes de-aggregation properties; after obtaining the overall decision variables, there's no need to calculate individual decision variables through optimization or collaborative control methods. The de-aggregation method is as follows:

[0087]

[0088] Taking ES as an example, a very important constraint is that the storage device cannot be charged and discharged simultaneously, that is... Introducing 0-1 variables into the decision variables makes the feasible region discrete and non-convex, greatly complicating aggregation. Furthermore, to ensure this constraint, some studies use only one... The excitation energy (ES) is represented by a value greater than zero for charging and less than zero for discharging. However, this approach ignores charging and discharging efficiency, assuming that the efficiency is always 1, which has a significant deviation from current practical applications. Other studies use a cost function to... Including it as a penalty term in the objective function can relax this nonlinear constraint. However, this method has its applicable scenarios; essentially, it increases the net discharge power of the energy storage system (ES) or decreases the net charging power of the ES under the same energy change. It only achieves good results when the original objective aligns with the penalty term's objective. When the original objective aims to increase the energy storage charging power, it contradicts the penalty term's objective, making it impossible to obtain the desired solution. Therefore, existing aggregation methods struggle to guarantee that when scheduling an equivalent aggregate of ESs, each individual ES will satisfy the constraint of simultaneous charging and discharging.

[0089] In the original approximation method, without considering simultaneous charging and discharging, the maximum internal approximation of the original polyhedron is obtained by solving the optimization problem of equation (10):

[0090]

[0091] Although The lower bound for the charging and discharging power is 0, but because it is an internal approximation, The existence of this constraint will cause the lower bound of the charging and discharging power to be greater than 0 at certain times. The originally expected maximum inner approximation will instead cause the constraint of non-simultaneous charging and discharging to be violated at certain times, and the obtained approximate feasible region will be practically unusable.

[0092] Based on the above understanding, this invention proposes a translation-free approximation method to address this problem. In the inner approximation, the translation factor is removed, and only the scaling factor is retained:

[0093]

[0094] This method not only ensures that the lower bound of the charging and discharging power is 0, but also guarantees the consistency of the overall behavior of the ES cluster. As long as the ES clusters after equivalent aggregation are not simultaneously charging and discharging, each EV will also not charge and discharge simultaneously. This will be achieved by imposing simultaneous charging and discharging constraints on the ES cluster when aggregating different types of flexibility resources. The corresponding optimization problem is modified as follows:

[0095]

[0096] This method can be used not only for energy storage, but also for any flexible resource with varying simultaneous charge / discharge constraints, considering charge / discharge efficiency. Furthermore, because it ensures consistent charge / discharge cycles, it also facilitates the addition of constraints on the number of charge / discharge cycles.

[0097] See Figure 1 Based on the above, the implementation steps of this embodiment are as follows:

[0098] Step 1: The resource aggregation platform will distribute the template and calculation method of Equation (4) to each individual resource and require them to report the corresponding model coefficient matrix.

[0099] Step 2: Each resource entity needs to calculate its model coefficient matrix using a template and report it to the resource aggregation platform of the same type.

[0100] Step 3: The resource aggregation platform calculates the primitive polyhedron according to equation (5) and distributes the coefficients to each individual resource. Depending on whether it is an energy storage resource, the parameter calculation method of equation (10) or equation (14) is distributed to each individual resource.

[0101] Step 4: Each resource entity independently calculates its parameters according to formula (10) or formula (14) and reports them to the resource aggregation platform of the same type.

[0102] Step 5: The final equivalent aggregate calculated by the same type of aggregation platform according to equation (8) is reported to the scheduling center.

[0103] Step 6: The scheduling center makes a decision based on the final equivalent aggregate and sends the decision result to similar aggregation platforms.

[0104] Step 7: The similar aggregation platform deaggregates the decision results according to formula (11) and distributes them to each resource individual for execution.

[0105] See Figure 2 In existing technologies, energy storage devices exhibit simultaneous charging and discharging at both the upper and lower limits to relax energy constraints and achieve higher power ceilings. However, in reality, energy storage devices cannot charge and discharge simultaneously. Figure 2 The power limit specified in the figure is unattainable in practical applications.

[0106] See Figure 3 The translation-free approximation method proposed in this invention can effectively ensure that each individual meets the constraints of charging and discharging at different times, thereby accurately giving the upper limit of power in practical application scenarios.

Claims

1. A method for modeling and aggregating the adjustment characteristics of flexible resources, characterized in that, Includes the following steps: Step 1: The virtual battery model coefficient matrix reported by each individual flexible resource is collected by the resource aggregation platform of the same type. The category of the flexible resource to which each individual belongs is either energy storage system (ES) or electric vehicle (EV). Step 2: The similar resource aggregation platform uses the coefficient matrix of each individual collected as the polyhedron representation, and then calculates the primitive polyhedron based on the maximum inner approximation algorithm of similar polyhedra, and sends the coefficients of the calculated primitive polyhedron to each individual. Step 3: The resource aggregation platform receives the parameters calculated independently by each individual. The parameters are calculated by each individual according to the category of its own flexible resources and using the corresponding parameter calculation method. Step 4: The resource aggregation platform calculates the final equivalent aggregate based on the parameters of each individual and sends it to the scheduling center for decision-making based on a preset decision algorithm, and then receives the decision result. Step 5: The resource aggregation platform of the same type de-aggregates the decision results and finally distributes them to each individual for execution.

2. The method according to claim 1, characterized in that, In step 1, the virtual battery model coefficient matrix reported by the i-th flexibility resource individual for: Where K represents the category of the resource individual, K∈{ES,EV}, For decision variables, F i K , The coefficient matrix, Let τ represent the real number space, and τ be the maximum offline time interval.

3. The method according to claim 2, characterized in that, In step 1, the virtual battery model coefficient matrix is ​​obtained based on the virtual battery model issued by a similar resource aggregation platform. The virtual battery model is as follows: in, Indicates matrix transpose. P i K,ch This represents the column vector of charging power for the i-th individual. Let be the charging power of the i-th individual at time t. P i K,dis This represents the column vector of discharge power for the i-th individual. Let be the discharge power of the i-th individual at time t. This represents the upper bound of the charging power of the i-th individual, and T represents the offline time interval, T = {1, 2, ..., τ}. This represents the upper bound of the discharge power of the i-th individual. This represents the maximum downhill climbing speed of the i-th individual. This represents the power of the i-th individual at time t. This represents the maximum upward climbing speed of the i-th individual. This represents the lower bound of the energy value of the i-th individual at time t+1. This represents the energy value of the i-th individual at time t+1. This represents the upper bound of the energy value of the i-th individual at time t+1.

4. The method according to claim 3, characterized in that, In step 1, the virtual battery model is obtained by combining the individual models of ES and EV with a unified model based on the adjustment characteristics.

5. The method according to claim 4, characterized in that, The individual model of ES is: in, Let be the charging power of the i-th ES at time t. Let be the discharge power of the i-th ES at time t. Let be the upper bound of the charging power of the i-th ES at time t. Let be the upper bound of the discharge power of the i-th ES at time t; Let be the external power of the i-th ES; Let i be the maximum downhill ramp rate of the i-th ES. The maximum upward climbing rate of the i-th ES; For the energy stored in the i-th ES, The self-discharge rate is denoted by Δt, and the time interval is Δt. Let i be the charging efficiency of the i-th ES. Let be the discharge efficiency of the i-th ES; Let i be the upper bound of the allowed energy for the i-th ES. This is the lower bound of the allowed energy for the i-th ES; The individual model of EV is: When the EV is charging in, Let be the charging power of the i-th EV at time t. Let be the discharge power of the i-th EV at time t; Let i be the upper bound of the charging power of the i-th EV. This is the upper bound of the discharge power of the i-th EV; Let be the external power of the i-th EV; For the maximum downhill climbing rate, The maximum uphill climbing rate; Let i be the energy of the i-th EV at time t. Let be the self-discharge rate of the i-th EV at time t, and Δt be the time interval; Let be the charging efficiency of the i-th EV at time t. Let be the discharge efficiency of the i-th EV at time t; Let i be the upper bound of the allowed energy for the i-th EV. Let i be the lower bound of the allowed energy for the i-th EV. Let i be the expected energy value of the i-th EV at time t+1; When the EV is not charging 6. The method according to claim 2, characterized in that, In step 2, calculating the primitive polyhedron includes: Using the coefficient matrix of the virtual battery model As a representation of a polyhedron, the average value of each category of flexibility resources is taken. To be used as a basic polyhedron: in, For F i K The average value, N K The number of individuals representing a certain type of flexible resource. for The average value, The decision variables represent the primitive polyhedrons; Then similar resource aggregation platforms will and Send to each individual.

7. The method according to claim 4, characterized in that, In step 3, when each individual calculates its parameters independently, if the individual is ES, the parameters are calculated using the following formula: in Minimize Indicates constraints. Represents the scaling factor transformation coefficient, G represents a nonnegative matrix, and r i K Represents the translation factor transformation vector; When the individual is EV, the parameters are calculated using the following formula:

8. The method according to claim 7, characterized in that, In step 3, the scaling factor and translation factor are transformed through the following variable substitutions: in, This represents the translation factor for the i-th individual. This represents the scaling factor of the i-th individual; and after the individual completes the calculation, it sends its own scaling factor or translation factor and scaling factor to the same resource aggregation platform.

9. The method according to claim 8, characterized in that, In step 4, the resource aggregation platform calculates the final equivalent aggregate based on the parameters of each individual resource. Where Ω K It is an equivalent polymer. This is the coefficient matrix of the virtual battery model. They are similar polyhedra; It is a primitive polyhedron; β K This represents the sum of scaling factors; This represents the sum of translation factors; Minkowski and N represent Minkowski and N. K The number of individuals for a certain type of flexible resource.

10. The method according to claim 9, characterized in that, In step 5, the de-aggregation of decision results by the similar resource aggregation platform includes: Similar resource aggregation platforms perform de-aggregation based on the following formula: Where β i K Scaling factor β K U represents the sum of scaling factors. K Represents aggregated decision variables. This represents the sum of translation factors. The translation factor is... Then similar resource aggregation platforms will Distribute to each individual.

Images