Equivalent cluster aggregation method considering distributed resource operating state

Abstract

This application belongs to the field of smart grid dispatching, and specifically discloses an equivalent cluster aggregation method that takes into account the operating state of distributed resources. This application first obtains the parameters of the distributed resources and performs constraint modeling based on the equivalent cluster aggregation model, including the operating state of the distributed resources. Then, it performs robust optimization modeling on the constraint results to obtain its aggregation formula. Next, by replacing Lagrange relaxation and dual operations, the robust optimization problem is transformed into a solvable single form. Finally, the aggregated feasible region is calculated through a computational layer. This solves the shortcomings of existing equivalent cluster methods that lack robust optimization and solution mechanisms adapted to the operating state of distributed resources and network constraints, avoids calculation bias in the aggregated feasible region, and simplifies the solution complexity, thereby achieving an accurate characterization of the aggregated feasible region and enabling the equivalent cluster to respond more reliably to the dispatching needs of the smart grid.

Description

Technical Field

[0001] This application belongs to the field of smart grid dispatching, and more specifically, relates to an equivalent cluster aggregation method that takes into account the operating status of distributed resources. Background Technology

[0002] Currently, common distributed resources, such as micro gas turbines, battery energy storage units, thermal storage tanks, HVAC systems, electric vehicles, and distributed photovoltaics, share the common characteristics of being "small in number, large in total quantity, widely distributed, and diverse in type." Due to limitations in communication capabilities and data processing capacity, directly managing and utilizing distributed resources is not feasible. In the smart grid industry, equivalent clustering, as a means of intelligently scheduling distributed resources through aggregation, is increasingly favored by industry and academia.

[0003] Compared to a large number of distributed resources, the equivalent cluster, through aggregation, has only one aggregated feasible region at the common coupling point of its power interaction with the upper-level grid. This region is physically a constraint on the operation of the interaction power. Therefore, the equivalent cluster can use this aggregated feasible region to participate in grid dispatch and the electricity market, while the upper-level grid only needs to use this aggregated feasible region to issue dispatch instructions, greatly simplifying grid dispatch and operation and the management of distributed resources—a win-win situation.

[0004] However, the aggregation process of equivalent clusters generally lacks real-time consideration of whether individual distributed resources should participate. This is because the aggregation of equivalent clusters needs to take into account the different operating characteristics of different types of distributed resources and the network constraints between various distributed resources. At the same time, in reality, distributed resources in equivalent clusters generally have 0-1 operating states, such as the start-stop state of micro gas turbines and the on / off state of temperature-controlled loads, which will affect the size of the feasible region for aggregation.

[0005] Even considering the 0-1 operating states of distributed resources, the lack of effective problem modeling methods makes it difficult to achieve robust optimization solutions for distributed resources with multiple uncertainties. Therefore, the aggregated feasible region obtained through existing technologies is generally inaccurate, resulting in limited scheduling instructions. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this application aims to propose an equivalent cluster aggregation method that takes into account the operating state of distributed resources. This method addresses the problem that existing equivalent clusters lack modeling and optimization methods that are adapted to the operating state and network constraints of distributed resources, resulting in inaccurate calculation of the aggregation feasible region and limited scheduling instructions.

[0007] To achieve the above objectives, in a first aspect, this application provides an equivalent cluster aggregation method that considers the operational state of distributed resources. The method includes: obtaining parameters of each distributed resource in the equivalent cluster; based on the equivalent cluster aggregation model, constraining the distributed resources according to the parameters of the distributed resources to obtain equivalent cluster constraints including the operational state of the distributed resources; based on the equivalent cluster aggregation model, performing robust optimization modeling on the equivalent cluster constraints, and performing dual operations on the robust optimization modeling results to obtain the aggregation constraint results of the aggregated feasible region through computation; wherein, the equivalent cluster aggregation model includes: a modeling layer, a robust optimization layer based on robust optimization with an objective function of 0, a dual layer based on the equivalent cluster constraints replacing the duality of Lagrange relaxation, and a computation layer; and obtaining the aggregated feasible region of the equivalent cluster based on the aggregation constraint results.

[0008] In one embodiment, based on an equivalent cluster aggregation model, constraint modeling is performed on the distributed resources according to their parameters to obtain equivalent cluster constraints including the operating state of the distributed resources. This includes: based on the modeling layer, constraining modeling is performed on the distributed resources according to their parameters to obtain the operating constraints and network constraints of the distributed resources; the operating constraints and network constraints are rearranged to obtain inequality constraints, equality constraints, and mixed integer linear constraints including the operating state of the distributed resources; based on the modeling layer, aggregation constraints of the aggregated feasible region are obtained according to the parameters of the distributed resources; and equivalent cluster constraints are obtained based on the mixed integer linear constraints, inequality constraints, equality constraints, and aggregation constraints.

[0009] In one embodiment, based on the equivalent cluster aggregation model, robust optimization modeling of the equivalent cluster constraints is performed, and a dual operation is executed on the robust optimization modeling result to obtain the aggregation constraint result of the aggregated feasible region through computation, including:

[0010] Based on the robust optimization layer, the equivalent cluster constraint is modeled as an objective function; based on the dual layer, the objective function is dualized according to the equivalent cluster constraint to obtain the objective dual function; based on the computation layer, the objective dual function is parsed to obtain the worst-case scheduling instruction; based on the computation layer, the aggregated constraint result of the aggregated feasible region is generated iteratively according to the worst-case scheduling instruction.

[0011] In one embodiment, based on the dual layer, the objective function is subjected to dual operations according to the equivalent cluster constraints to obtain the objective dual function, including: performing relaxation and dual operations on the inner functions of the objective function through mixed integer linear constraints to obtain a first dual function; performing dual operations on inequality constraints and equality constraints to obtain a second dual function; and obtaining the objective dual function based on the first dual function, the second dual function, and the objective function.

[0012] In one embodiment, a first dual function is obtained by relaxing and dualizing the inner functions of the objective function through mixed integer linear constraints, including: relaxing the inner functions of the objective function by replacing the Lagrange multiplication with mixed integer linear constraints to obtain a relaxed function; and performing a dual operation on the relaxed function and transforming it into a strongly dual form to obtain the first dual function.

[0013] In one embodiment, based on the computation layer, the aggregation constraint result of the aggregated feasible region is generated iteratively according to the worst scheduling instruction, including: based on the computation layer, obtaining the feasible scheduling instruction closest to the worst scheduling instruction according to the worst scheduling instruction; iterating according to the closest feasible scheduling instruction until the result of the objective dual function is 0, thereby generating the aggregation constraint result of the aggregated feasible region.

[0014] In one embodiment, after obtaining the aggregated feasible domain of the equivalent cluster based on the aggregation constraint results, the method further includes: transmitting the aggregated feasible domain to the scheduling center; receiving intelligent scheduling instructions generated by the scheduling center based on the aggregated feasible domain; and dynamically adjusting the distributed resources in the equivalent cluster based on the intelligent scheduling instructions.

[0015] In a second aspect, this application provides an electronic device, comprising: at least one memory for storing a program; and at least one processor for executing the program stored in the memory, wherein when the program stored in the memory is executed, the processor is configured to execute the method described in the first aspect or any possible implementation thereof.

[0016] Thirdly, this application provides a computer-readable storage medium storing a computer program that, when run on a processor, causes the processor to perform the method described in the first aspect or any possible implementation thereof.

[0017] Fourthly, this application provides a computer program product that, when run on a processor, causes the processor to perform the method described in the first aspect or any possible implementation thereof.

[0018] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here.

[0019] Overall, the technical solutions conceived in this application have the following beneficial effects compared with the prior art:

[0020] This application first obtains distributed resource parameters to provide accurate input for modeling, avoiding model distortion caused by missing parameters. Second, it incorporates operational state variables, such as start / stop and power limits, into constraint modeling, enabling the model to accurately adapt to the actual physical characteristics of the resources and overcoming the feasible region bias caused by neglecting state constraints in traditional methods. Furthermore, it encapsulates the cluster model with operational state constraints into a feasible set framework through robust optimization with an objective function of 0, and uses alternative Lagrange relaxation and dual operations to equivalently transform the complex high-dimensional constraint system into a single low-dimensional aggregation formula to solve the aggregation constraint results. This fundamentally solves the problem that existing methods rely on approximate simplification or complex iterations because they cannot handle dimensional explosion.

[0021] Meanwhile, the model's hierarchical structure—modeling layer, robust optimization layer, dual layer, and computation layer—systematically ensures the rigor of this transformation process and guarantees the crucial dimensionality reduction role of the dual layer. Finally, based on the simplified aggregation constraint results, the accurate feasible region is directly output, significantly reducing computational complexity and resource consumption. This enables a rapid and accurate characterization of the feasible region under the full-state constraints of the equivalent cluster, allowing scheduling instructions to be executed safely and reliably.

[0022] Compared with existing technologies, this method is the first to deeply embed the operating state of distributed resources into the equivalent cluster model. Through a robust and dual collaborative transformation mechanism, it completely avoids the dimensionality curse problem of solving high-dimensional constraints while ensuring equivalence. This not only eliminates the feasible region bias caused by empirical fitting, but also avoids the computational bottleneck of traditional optimization methods, providing efficient and reliable cluster scheduling support for smart grids. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of an equivalent cluster instance that includes distributed resources;

[0024] Figure 2 This is one of the flowcharts provided in this application for an equivalent cluster aggregation method embodiment that takes into account the operating state of distributed resources;

[0025] Figure 3 This is the second flowchart provided in the embodiments of the method of this application;

[0026] Figure 4 This is the third flowchart provided in the embodiments of the method of this application;

[0027] Figure 5 This is the fourth flowchart provided in the embodiments of the method of this application;

[0028] Figure 6 This is the fifth flowchart provided in the embodiments of the method of this application;

[0029] Figure 7This is a topology diagram of a specific implementation of a virtual power plant based on the IEEE-33 node system according to an embodiment of the method of this application;

[0030] Figure 8 This is a diagram of the intraday distributed photovoltaic power output of a specific embodiment of this application;

[0031] Figure 9 This is a daily temperature map of a specific embodiment of this application;

[0032] Figure 10 This is a daily load chart of a specific embodiment of this application;

[0033] Figure 11 This is a feasible intraday aggregation domain diagram of a specific implementation of this application;

[0034] Figure 12 This is a schematic diagram of scheduling instructions based on aggregated feasible domains in a specific implementation of this application;

[0035] Figure 13 This is a schematic diagram of the structure of an electronic device provided in a specific embodiment of this application. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. The technical terms used in the embodiments of this application are introduced below.

[0037] Equivalent cluster: This refers to aggregating multiple distributed resources (such as photovoltaics, energy storage, and loads) and using a unified control strategy to transform them into a single virtual entity that can be centrally scheduled and respond to market and model demands (such as grid demands). Figure 1 As shown, an equivalent cluster of distributed resources is illustrated, namely a virtual power plant.

[0038] Aggregate feasible region: For example, a virtual power plant will exchange bidirectional power with the upper-level power grid at a common coupling point; the aggregate feasible region refers to the capacity range of the equivalent cluster that can exchange power.

[0039] The embodiments of this application are described below with reference to the accompanying drawings. Please refer to... Figure 2 This embodiment provides an equivalent cluster aggregation method that takes into account distributed resources in their operational state. Referring to the virtual power plant description, the method includes steps S10 to S40:

[0040] Step S10: Obtain the parameters of each distributed resource in the equivalent cluster.

[0041] It should be noted that, in this embodiment, the parameters of distributed resources refer to the parameters of power generation units, energy storage units, or load units, such as the upper and lower limits of the output power of photovoltaic, micro gas turbines, or other power generation units; the upper and lower limits of transmission line power and voltage in the transmission line; the upper and lower limits of the state of charge and charging / discharging power of energy storage units; and the real-time power, electricity consumption, and power factor of load units. The operating status of distributed resources refers to the 0-1 operating status of each unit.

[0042] It is understood that the above data can be automatically collected in real time through on-site monitoring equipment such as sensors, smart meters or protocol gateways, or it can be supplemented by manual input such as equipment inspection records or offline debugging data. This embodiment does not limit the means of acquisition, but mainly focuses on integrating and processing the parameters of these multi-source heterogeneous distributed resources.

[0043] Step S20: Based on the equivalent cluster aggregation model, constrain the distributed resources according to the parameters of the distributed resources to obtain the equivalent cluster constraints including the operating status of the distributed resources.

[0044] Step S30: Based on the equivalent cluster aggregation model, robust optimization modeling of the equivalent cluster constraints is performed, and dual operation is executed on the robust optimization modeling results to obtain the aggregation constraint results of the aggregate feasible region through computation.

[0045] It should be noted that the equivalent cluster aggregation model includes: a modeling layer, a robust optimization layer based on robust optimization with an objective function of 0, a dual layer based on equivalent cluster constraints replacing the duality of Lagrange relaxation, and a computation layer.

[0046] Understandably, once operational status data, including the start-up and shutdown states of distributed resources, is obtained, it is necessary to perform constraint modeling on this complex multidimensional data. Robust optimization is used to build a problem model seeking the range of the constraint intersection. This problem model seeking the range of constraint intersection includes two aspects: first, the decision variables of each distributed energy source and network within the equivalent cluster; second, the power interaction between the equivalent cluster and the upper-level power grid, or the dispatch instructions from the upper-level power grid. The interaction power between the equivalent cluster and the upper-level power grid is the dispatch instruction from the upper-level power grid, which is then made based on the aggregated feasible region obtained by aggregating the equivalent cluster. The aggregated feasible region is part of the constraints of the equivalent cluster.

[0047] It should be noted that this embodiment selects a robust optimization layer based on an objective function of 0. The core idea is that the decision variables of each distributed energy source and network within the equivalent cluster, and the power of the equivalent cluster interacting with the external grid, represent different optimization directions. For the variables within the equivalent cluster, the goal is to expand the aggregate feasible region to fully utilize its efficiency; while for the power of the equivalent cluster interacting with the external grid, or the dispatch instructions from the upper-level grid, the goal is to narrow the aggregate feasible region to obtain an absolutely feasible dispatch instruction.

[0048] It should be noted that, mathematically, the aggregation problem of equivalent clusters is used to verify whether the given robust optimization problem has a solution, so it is sufficient to set the objective function to 0. If the robust optimization problem can achieve a result of 0, it means that under the current equivalent cluster constraints, the dispatching instructions issued by the upper-level power grid can definitely be executed by the equivalent cluster, which means that the aggregation feasible region has been obtained.

[0049] As is understandable, Lagrange relaxation is a technique for handling constrained optimization problems. In optimization problems, there are usually constraints that limit the range of feasible solutions. Lagrange relaxation introduces Lagrange multipliers to incorporate the constraints into the objective function, thereby transforming the original constrained optimization problem into an unconstrained or less constrained optimization problem.

[0050] Understandably, replacing Lagrange relaxation with equivalent cluster constraints for robust optimization problems simplifies the modeling and solution process. For example, when considering the power interaction optimization problem between the equivalent cluster and the upper-level grid, using Lagrange relaxation might require constructing a complex objective function containing multiple constraint-related multipliers. However, with equivalent cluster constraints, the objective function can be more directly expressed as a cost minimization or benefit maximization problem under conditions such as equivalent cluster power output constraints and energy storage constraints, avoiding the complex multiplier adjustment process.

[0051] Understandably, for an optimization problem, more information about the primal problem can be obtained by constructing its dual problem. The dual problem usually has a different form than the primal problem, but under certain conditions, there is a specific relationship between the optimal solutions of the primal and dual problems. For example, the strong duality theorem states that under certain conditions, the optimal value of the primal problem is equal to the optimal value of the dual problem. Through duality theory, optimization problems can be analyzed and solved from different perspectives, and sometimes it can even provide lower or upper bounds for the primal problem, helping to evaluate the quality of the solution. Solving the dual problem in this case yields the aggregated constraints of the aggregated feasible region, i.e., the parameter values ​​of the aggregated constraints of the aggregated feasible region.

[0052] Step S40: Based on the aggregation constraint results, obtain the aggregation feasible region of the equivalent cluster.

[0053] It is understandable that, based on the parameter values ​​of the aggregation constraints of the aggregation feasible region, the capacity range of interactive power can be obtained, which describes the maximum power adjustment range that the cluster can provide within a given time scale. The equivalent net output power that the cluster can increase (such as energy storage discharge, photovoltaic power generation) and the net output power that the cluster can reduce (such as energy storage charging, micro gas turbine output reduction, photovoltaic curtailment) are the aggregation feasible region.

[0054] In this embodiment, firstly, distributed resource parameters are obtained to provide accurate input for modeling, avoiding model distortion caused by missing parameters. Secondly, operational state variables, such as start / stop and power limits, are incorporated into constraint modeling, enabling the model to accurately adapt to the actual physical characteristics of the resources and overcoming the deviation in feasible region caused by neglecting state constraints in traditional methods. Furthermore, robust optimization with an objective function of 0 encapsulates the cluster model containing operational state constraints into a feasible set framework. Using alternative Lagrange relaxation and dual operations, the complex high-dimensional constraint system is equivalently transformed into a single low-dimensional aggregation formula to solve the aggregation constraint results. This fundamentally solves the problem of existing methods relying on approximate simplification or complex iterations due to their inability to handle dimensionality explosion. Simultaneously, the hierarchical structure of the model—modeling layer, robust optimization layer, dual layer, and computation layer—systematically ensures the rigor of this transformation process, guaranteeing the crucial dimensionality reduction effect of the dual layer. Finally, based on the mathematically simplified aggregation constraint results, the accurate feasible region is directly output, significantly reducing computational complexity and resource consumption. This achieves a rapid and accurate characterization of the feasible region under equivalent cluster full-state constraints, enabling the safe and reliable execution of scheduling instructions.

[0055] Compared to existing technologies, this method deeply embeds the operating state of distributed resources into the equivalent cluster model and completely avoids the dimensionality curse problem of solving high-dimensional constraints through a robust and dual collaborative transformation mechanism, while ensuring equivalence. It eliminates the feasible region bias caused by empirical fitting and avoids the computational bottleneck of traditional optimization methods, providing efficient and reliable cluster scheduling support for smart grids.

[0056] Further, please refer to Figure 3 This embodiment provides specific steps S20, including S21 to S24:

[0057] Step S21: Based on the modeling layer, constrain the distributed resources according to their parameters to obtain the operational and network constraints of the distributed resources.

[0058] Specifically, this embodiment uses a virtual power plant in an equivalent cluster as an example. The distributed resources in the virtual power plant include micro gas turbines, battery storage units, temperature-controlled loads, and distributed photovoltaics. Network constraints exist within the virtual power plant, and the distributed resources are connected via transmission lines. The model of the virtual power plant consists of the operational constraints of its distributed resources and network constraints. By substituting the operational status data into the modeling layer, the corresponding constraints can be obtained, as shown below:

[0059] (1)

[0060] (2)

[0061] (3)

[0062] Equations (1) to (3) represent the operating constraints of the micro gas turbine.

[0063] (4)

[0064] (5)

[0065] (6)

[0066] (7)

[0067] Equations (4) to (7) represent the operational constraints of the battery energy storage unit.

[0068] (8)

[0069] (9)

[0070] Equations (8) and (9) represent the output constraints of distributed photovoltaic systems.

[0071] (10)

[0072] (11)

[0073] Equations (10) and (11) are the operating constraints of the temperature-controlled load.

[0074] (12)

[0075] (13)

[0076] (14)

[0077] (15)

[0078] (16)

[0079] (17)

[0080] Equations (12) to (17) are network constraints.

[0081] Specifically, This represents the active power / reactive power output of the micro gas turbine. The operating state of the micro gas turbine is described by Boolean variables; This represents the active power / reactive power output of the battery energy storage unit. This represents the stored energy value. This represents the active power / reactive power output of distributed photovoltaic systems. This represents the active power / reactive power load value for the temperature-controlled load. The operating status of the temperature-controlled load is described by Boolean variables; The active power / reactive power flowing on the transmission line between nodes. The square of the node voltage; For time period, This refers to the node number in the virtual power plant.

[0082] Specifically, These represent the upper and lower limits of the power output of the micro gas turbine. The upper and lower limits of power ramp-up for micro gas turbines; This refers to the upper limit of the charging and discharging power of the battery energy storage unit. This is the upper limit of energy storage. The upper and lower limits of the state of charge to ensure the healthy operation of the battery energy storage unit; This refers to the power generation of distributed photovoltaic systems. To ensure the economic viability of distributed photovoltaic operation, the curtailment rate is capped; These represent the lower and upper limits of indoor temperature. Outdoor temperature The equivalent thermal resistance of the temperature-controlled load, The operating coefficient for temperature-controlled loads; The active / reactive load of the virtual power plant. These are the upper and lower limits of the active power of the transmission line. These are the upper and lower limits of the reactive power of the transmission line. These are the upper and lower limits of the square of the node voltage. For the resistance and inductance of the transmission line; The power factors are for micro gas turbines, battery energy storage units, distributed photovoltaics, and temperature-controlled loads, respectively.

[0083] Step S22: Perform a restructuring operation on the runtime constraints and network constraints to obtain inequality constraints, equality constraints, and mixed-integer linear constraints including the start and stop states of distributed resources. Restructuring the constraints into a compact form:

[0084] (18)

[0085] (19)

[0086] (20)

[0087] Specifically, Let be a vector consisting of all decision variables. Let be a vector consisting of continuous decision variables in a mixed-integer linear constraint. Let be a vector of Boolean variables in the mixed-integer linear constraints; Equation (18) corresponds to all mixed-integer linear constraints, Equation (19) corresponds to all inequality constraints, and Equation (20) corresponds to all equality constraints; parameter matrix sum coefficient vector All are determined by the respective constraints.

[0088] Step S23: Based on the modeling layer, obtain the aggregation constraints of the aggregation feasible region according to the parameters of the distributed resources. The aggregation operational region of the virtual power plant is as follows:

[0089] (twenty one)

[0090] (twenty two)

[0091] (twenty three)

[0092] (twenty four)

[0093] (25)

[0094] Specifically, Total number of time periods The power exchanged between the virtual power plant and the upper-level power grid is represented by equations (21) and (22), which are the constraints corresponding to the aggregated operation domain of the virtual power plant. For parameter matrices, This is a coefficient vector, corresponding to the upper and lower bounds of the virtual power plant's output power and the upper and lower bounds of its output power ramp-up. This is the quantity that the virtual power plant aggregation ultimately seeks, and its relationship is shown in equations (23) to (25).

[0095] Step S24: Based on mixed-integer linear constraints, inequality constraints, equality constraints, and aggregation constraints, obtain equivalent cluster constraints. Equivalent cluster constraints refer to a set of constraints including mixed-integer linear constraints, inequality constraints, equality constraints, and aggregation constraints.

[0096] Further, please refer to Figure 3 This embodiment provides specific steps S30, including steps S31 to S34:

[0097] Step S31: Based on the robust optimization layer, the equivalent cluster constraints are modeled as an objective function. The operating domain that the virtual power plant needs to solve corresponds to a min-max robust optimization problem with an objective function of 0:

[0098] (26)

[0099] (27)

[0100] Specifically, the objective function set in equation (26) is independent of the decision variables. If the constraints in equation (27) are valid, the optimal solution to the problem can be 0.

[0101] Step S32: Based on the dual layer, perform dual operations on the objective function according to the equivalent cluster constraints to obtain the objective dual function.

[0102] It should be noted that step S32 includes: relaxing and dualizing the inner function of the objective function through mixed integer linear constraints to obtain the first dual function; dualizing the inequality constraints and equality constraints to obtain the second dual function; and obtaining the objective dual function based on the first dual function, the second dual function, and the objective function.

[0103] Specifically, by performing relaxation and dual operations on the inner functions of the objective function through mixed integer linear constraints, the first dual function is obtained. This includes: relaxing the inner functions of the objective function by replacing the Lagrange multiplication with mixed integer linear constraints to obtain a relaxed function; and performing dual operations on the relaxed function and transforming it into a strongly dual form to obtain the first dual function.

[0104] Specifically, For matrix The number of columns, For vectors The number of elements; the inner-layer max problem and corresponding mixed-integer linear constraints in the virtual power plant aggregation model are proposed as follows:

[0105] (28)

[0106] (29)

[0107] Specifically, let the Lagrange multipliers be By replacing the Lagrange relaxation with mixed-integer linear constraints, the constraints are relaxed as follows:

[0108] (30)

[0109] (31)

[0110] Preferably, it can be written as follows Lagrange dual function as follows:

[0111] (32)

[0112] (33)

[0113] Specifically, The inner max problem and positive Lagrange multipliers It is irrelevant, The following is proposed and rewritten:

[0114] (34)

[0115] (35)

[0116] Preferably, because ,so ;and Therefore It is actually a direct proportional function, and only when... At the optimal value ,and Since they are the same, they satisfy strong duality.

[0117] Preferably, equations (34) and (35) are transformed into the following strongly dual form, namely the first dual function:

[0118] (36)

[0119] (37)

[0120] Specifically, Let be the coefficient vector, and obtain it as follows:

[0121] (38)

[0122] (39)

[0123] Preferably, the solution is an iterative process, in the first iteration... In this iteration, the constraints of the virtual power plant are:

[0124] (40)

[0125] Specifically, the initial values ​​of the parameters in the virtual power plant aggregation solution iteration. The solution method is as follows:

[0126] (41)

[0127] (42)

[0128] (43)

[0129] (44)

[0130] (45)

[0131] (46)

[0132] Preferably, the remaining linear constraints in the aggregation model are subjected to dual transformation. Since the remaining constraints do not include start and stop states, only conventional dual transformation is needed to obtain the second dual function. Combining the first dual function, the original problem is obtained. The dual problem, i.e., the objective dual function, is as follows:

[0133] (47)

[0134] (48)

[0135] (49)

[0136] Specifically, As dual variables, Lagrange multipliers under KKT conditions Let M be a sufficiently large positive number. Used as a Boolean variable, it transforms complex constraints into a form that can be handled by linear programming / mixed integer programming.

[0137] Step S33: Based on the computation layer, the objective dual function is analyzed to obtain the worst-case scheduling instruction. Specifically, the computation layer solves problems (47) and (48) to obtain... and , That is, the current The worst-case scheduling instruction issued.

[0138] Step S34: Based on the computation layer, the aggregated constraint result of the aggregated feasible region is generated iteratively according to the worst-case scheduling instruction. It can be understood that obtaining the worst-case scheduling instruction provides an initial boundary for the iterative process. Through iterative adjustments to the scheduling instruction, an aggregated constraint result of the aggregated feasible region that satisfies all constraints is generated.

[0139] Further, please refer to Figure 3 This embodiment provides the specific steps of S34, including steps S341 to S342:

[0140] Step S341: Based on the computation layer, obtain the feasible scheduling instruction closest to the worst-case scheduling instruction according to the worst-case scheduling instruction. Solve for the distance Recent feasible scheduling instructions as follows:

[0141] (50)

[0142] (51)

[0143] Specifically, according to and The obtained first Parameters in the next iteration as follows:

[0144] (52)

[0145] (53)

[0146] Specifically, Let the variable be a Boolean variable, and the solution is obtained for the first... Parameters of the next iteration Afterwards, order and .

[0147] Step S342: Based on the most recent feasible scheduling instruction, iterate until the objective dual function result is 0, generating the aggregated constraint result of the aggregated feasible region. That is, repeat the formula steps until the solution of equation (47) is obtained. The aggregated feasible domain parameters of the virtual power plant can be obtained. The virtual power plant's aggregated feasible region parameters, which are the aggregated constraints of the aggregated feasible region, fully express the aggregated constraints of the aggregated feasible region. This allows the aggregated feasible region to be transformed from an abstract concept into a concrete mathematical model, facilitating the subsequent generation and verification of scheduling instructions.

[0148] In one embodiment, please refer to Figure 6 Step S40 is followed by steps S50 to S70:

[0149] Step S50: Transmit the aggregated feasible domain to the scheduling center.

[0150] Step S60: Receive the intelligent scheduling instruction generated by the scheduling center based on the aggregated feasible domain.

[0151] It is understandable that the equivalent cluster aggregation constraint model will be used to constrain... The application is submitted to the higher-level power grid dispatch center, which can then issue corresponding dispatch instructions. .

[0152] Step S70: Based on intelligent scheduling instructions, dynamically regulate the distributed resources in the equivalent cluster.

[0153] It is understandable that the control signals for specific distributed resources must satisfy the condition that the sum of the output / adjustment of all individual resources equals... At the same time, the output of each resource does not exceed the limits, such as the maximum power of the power generation unit and the SOC range of energy storage. It can be allocated according to preset priorities or protocols, such as photovoltaic priority and energy storage balanced allocation according to SOC.

[0154] Here, this embodiment combines the following specific application scenario, such as Figure 7 As shown, a specific implementation method is given, and the beneficial effects that can be achieved by this embodiment are further described. In this embodiment, a virtual power plant based on the IEEE-33 node system is presented, which includes 3 micro gas turbines, 2 energy storage units, 2 temperature-controlled loads, and 6 distributed photovoltaic units. The aggregation results and operating conditions considering and not considering the operating conditions of distributed resources are compared. Specifically, the daily distributed photovoltaic output is as follows: Figure 8 As shown, the intraday temperature is as follows Figure 9 As shown, the intraday load is as follows Figure 10 As shown in the table. The operating status data for the micro gas turbine, battery energy storage unit, and temperature-controlled load are shown in Tables 1, 2, and 3, respectively:

[0155] Table 1 Parameters of Micro Gas Turbine

[0156]

[0157] Table 2 Battery Energy Storage Unit Parameters

[0158]

[0159] Table 3 Temperature control load parameters

[0160]

[0161] A program was written in MATLAB R2024a to call the Yalmip solver with Gurobi. The optimization problem was solved using a Legion laptop with an Intel Core i7-10510U processor, 32GB of RAM, and running Windows 11 Professional. The aggregated feasible region of the virtual power plant, considering / not considering the 0-1 operating states of distributed resources, is as follows: Figure 11 , Figure 12 As shown in Table 4, the net peak-shaving benefits obtained by using the aggregated feasible domain of the virtual power plant to participate in the main grid scheduling operation are as follows:

[0162] Table 4 Main Network Scheduling Results

[0163]

[0164] It is evident that, compared to the virtual power plant aggregation method that does not consider the 0-1 operating state of distributed resources, the equivalent cluster aggregation method that takes into account the operating state of distributed resources provided in this application can increase the aggregation feasible region and more accurately depict the aggregation feasible region when applied to virtual power plant aggregation, thereby effectively improving the economic benefits of virtual power plants participating in main grid scheduling.

[0165] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from the spirit or essential characteristics of this application. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of this application is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within this application. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0166] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

[0167] Based on the methods in the above embodiments, this application provides an electronic device, which may include: a processor 100, a communications interface 200, a memory 300, and a communication bus 400, wherein the processor 100, the communications interface 200, and the memory 300 communicate with each other through the communication bus 400. The processor 100 can call logical instructions in the memory 300 to execute the methods in the above embodiments.

[0168] Furthermore, the logical instructions in the aforementioned memory 300 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application.

[0169] Based on the methods in the above embodiments, this application provides a computer-readable storage medium storing a computer program that, when run on a processor, causes the processor to execute the methods in the above embodiments.

[0170] Based on the methods in the above embodiments, this application provides a computer program product that, when run on a processor, causes the processor to execute the methods in the above embodiments.

[0171] It is understood that the processor in the embodiments of this application can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor can be a microprocessor or any conventional processor.

[0172] The method steps in this application embodiment can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, portable hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can reside in an ASIC.

[0173] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in or transmitted through a computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0174] It is understood that the various numerical designations used in the embodiments of this application are merely for descriptive convenience and are not intended to limit the scope of the embodiments of this application.

[0175] Those skilled in the art will readily understand that the above are merely preferred embodiments of this application and are not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. An equivalent cluster aggregation method considering the operational status of distributed resources, characterized in that, The method includes: Obtain the parameters of each distributed resource in the equivalent cluster; Based on the equivalent cluster aggregation model, the distributed resources are constrained and modeled according to their parameters to obtain equivalent cluster constraints including the operating state of the distributed resources. Based on the equivalent cluster aggregation model, robust optimization modeling is performed on the equivalent cluster constraints, and dual operation is performed on the robust optimization modeling results to obtain the aggregation constraint results of the aggregate feasible region through computation. The equivalent cluster aggregation model includes: a modeling layer, a robust optimization layer based on a robust optimization with an objective function of 0, a dual layer based on the equivalent cluster constraint replacing the Lagrange relaxation, and a computation layer. Specifically, based on the equivalent cluster aggregation model, robust optimization modeling is performed on the equivalent cluster constraints, and dual operations are executed on the robust optimization modeling results to obtain the aggregation constraint results of the aggregated feasible region through computation. This includes: modeling the equivalent cluster constraints as an objective function based on the robust optimization layer; performing dual operations on the objective function according to the equivalent cluster constraints based on the dual layer to obtain the objective dual function; parsing the objective dual function based on the computation layer to obtain the worst-case scheduling instruction; and iteratively generating the aggregation constraint results of the aggregated feasible region based on the worst-case scheduling instruction using the computation layer. Specifically, based on the dual layer and according to the equivalent cluster constraints, the objective function is subjected to dual operations to obtain the objective dual function, including: performing relaxation and dual operations on the inner functions of the objective function through the mixed integer linear constraints of the equivalent cluster constraints to obtain a first dual function, that is, relaxing the inner functions of the objective function by replacing the Lagrange multiplication with the mixed integer linear constraints to obtain a relaxed function; performing dual operations on the relaxed function and transforming it into a strongly dual form to obtain the first dual function; performing dual operations on the inequality constraints and equality constraints of the equivalent cluster constraints to obtain a second dual function; and obtaining the objective dual function based on the first dual function, the second dual function, and the objective function. Based on the aggregation constraint results, the aggregation feasible region of the equivalent cluster is obtained.

2. The equivalent cluster aggregation method considering the operating status of distributed resources as described in claim 1, characterized in that, Based on the equivalent cluster aggregation model, the distributed resources are constrained according to their parameters to obtain equivalent cluster constraints including the operating state of the distributed resources, including: Based on the modeling layer, constraint modeling is performed on the distributed resources according to the parameters of the distributed resources to obtain the operational constraints and network constraints of the distributed resources. The operational constraints and network constraints are processed to obtain inequality constraints, equality constraints, and mixed integer linear constraints including the operational state of distributed resources; Based on the modeling layer, the aggregation constraints of the aggregate feasible region are obtained according to the parameters of the distributed resource; Based on the mixed integer linear constraint, the inequality constraint, the equality constraint, and the aggregation constraint, the equivalent cluster constraint is obtained.

3. The equivalent cluster aggregation method considering the operating status of distributed resources as described in claim 1, characterized in that, Based on the aforementioned computational layer, the aggregated constraint result of the aggregated feasible region is generated iteratively according to the worst-case scheduling instruction, including: Based on the computation layer, the feasible scheduling instruction closest to the worst-case scheduling instruction is obtained according to the worst-case scheduling instruction; Based on the most recent feasible scheduling instruction, iterate until the result of the objective dual function is 0, and generate the aggregated constraint result of the aggregated feasible region.

4. The equivalent cluster aggregation method considering the operating status of distributed resources as described in claim 1, characterized in that, After obtaining the aggregate feasible region of the equivalent cluster based on the aggregation constraint results, the process further includes: Transmit the aggregated feasible domain to the scheduling center; Receive the intelligent scheduling instruction generated by the scheduling center based on the aggregated feasible domain; Based on the intelligent scheduling instructions, the distributed resources in the equivalent cluster are dynamically controlled.

5. An electronic device, characterized in that, include: At least one memory for storing computer programs; At least one processor is configured to execute a program stored in the memory, wherein when the program stored in the memory is executed, the processor is configured to execute an equivalent cluster aggregation method taking into account the operating state of distributed resources as described in any one of claims 1 to 4.

6. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is run on a processor, the processor performs the equivalent cluster aggregation method that takes into account the operating state of distributed resources as described in any one of claims 1 to 4.

7. A computer program product, characterized in that, When the computer program product is run on a processor, the processor performs the equivalent cluster aggregation method as described in any one of claims 1 to 4, taking into account the operating state of distributed resources.

Images