Virtual power plant distributed resource aggregation and cooperative scheduling method

By establishing feasible domain solution models and collaborative optimization models in virtual power plants, using vertex enumeration method and ADMM algorithm, the problems of interaction and privacy protection between multiple virtual power plants are solved, and the flexibility of distributed resources and grid operation efficiency are improved. A safe and economical power grid scheduling has been achieved.

CN119010030APending Publication Date: 2024-11-22HUIZHOU POWER SUPPLY BUREAU OF GUANGDONG POWER GRID CO LTD
View PDF 0 Cites 5 Cited by

Patent Information

Application Number
CN202410691632.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-05-30
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The flexibility of existing technologies is difficult to effectively manage and utilize distributed energy, especially in terms of interaction and privacy protection between multiple virtual power plants, and the support role of reactive regulation capabilities in power grid operation is not fully evaluated.

Method used

A distributed resource aggregation and collaborative scheduling method for virtual power plants is proposed. By establishing a feasible domain solution model for virtual power plants that consider the uncertainty of renewable energy generation, the vertex enumeration method and ADMM algorithm are used to realize collaborative optimization scheduling of multiple virtual power plants. Ensure the security and economical system operation while protecting privacy information.

Benefits of technology

Effectively portray the feasible domain of multiple types of distributed resources in a single virtual power plant, improves its flexibility to participate in the power grid operation, realizes unified coordination and independent autonomous scheduling of multiple virtual power plants, with good convergence and scalability, ensuring Information security.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119010030A_ABST
    Figure CN119010030A_ABST
Patent Text Reader

Abstract

The invention discloses a virtual power plant distributed resource aggregation and cooperative scheduling method. The method mainly comprises the following steps: establishing a virtual power plant feasible region solving model considering renewable energy power generation uncertainty; solving the feasible region model of the virtual power plant by adopting a vertex enumeration method to obtain a feasible region of the virtual power plant, and representing the active-reactive regulation capability of a flexible resource aggregate in the virtual power plant; based on the virtual power plant feasible region, establishing an active power distribution network multi-virtual power plant collaborative optimization model; and solving the active power distribution network multi-virtual power plant collaborative optimization model by adopting an ADMM algorithm. According to the method, resource aggregation of distributed flexible resources and cooperative regulation and control of the system on the multiple virtual power plants are realized, and the privacy security of each virtual power plant can be protected.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention mainly relates to a method for aggregating and coordinating distributed resources of a virtual power plant, and belongs to the technical field of power system dispatching. Background Art

[0002] Numerous distributed energy resources in the distribution network, such as energy storage systems, photovoltaics, wind turbines, etc., have great flexibility potential. However, due to the small scale, wide distribution and heterogeneity of distributed energy resources, it is difficult to manage them and utilize their power flexibility. Virtual power plants provide an effective way to manage distributed resources in a clustered manner. Operators of virtual power plants can aggregate the flexibility of distributed resources as a whole to provide energy and ancillary services to the power grid. Therefore, it is of great significance to evaluate the equivalent power flexibility of virtual power plants by aggregating distributed resources. It can provide a scientific decision-making basis for power system operation and scheduling and power market bidding.

[0003] Power flexibility can be described as the feasible domain of active and reactive output power that satisfies network constraints and equipment operation constraints in all time periods. From a geometric point of view, the flexibility aggregation process can be regarded as a projection from the power flexibility area of ​​all distributed resources to the power flexibility area of ​​the virtual power plant spanned by the output power of the common coupling point. Therefore, this problem can be described as a projection from a high-dimensional origin polyhedron to a reduced-dimensional projection polyhedron. It is a mathematically difficult problem to obtain the exact analytical form of the projection polyhedron.

[0004] In the field of power grid regulation research, the collaborative optimization dispatching model of the feasible domain of flexible resource aggregation on the distribution network side is gradually becoming a research hotspot. It provides a useful model design reference for resource aggregates to participate in power grid regulation. However, the current resource modeling on the distribution network side is generally weak and fails to fully consider the volatility of renewable energy generation. Therefore, it is necessary to further enrich the modeling content to more accurately reflect the flexible regulation capability of the distribution network. By aggregating more types of flexible resources to form a virtual power plant and solving the feasible domain model containing active and reactive coupling, we can have a deeper understanding of the operation of the power grid. However, it is worth noting that in terms of resource aggregate regulation, most studies have not fully evaluated the supporting role of the reactive regulation capability of the feasible domain on the operation of the power grid, which is still an area to be further explored. At the same time, the correlation between the resource aggregation model and the aggregate regulation method is weak, and the application of resource aggregate upload parameters is not sufficient. These are all problems in current research.

[0005] In the existing technology, significant progress has been made in the research on the participation of massive distributed resources in power grid dispatching through the access of virtual power plants. These studies have shown that by effectively dispatching and controlling distributed resources, their flexibility can be fully utilized, thereby improving the efficiency of new energy consumption and load power supply satisfaction. However, there are still some shortcomings in the existing technology. On the one hand, there is still insufficient research on the interaction between multiple virtual power plants, and there is a lack of effective interaction strategies. Different virtual power plants aggregate diverse distributed resources. By strengthening the collaborative interaction between multiple virtual power plants, the regulation capabilities of these resources can be better utilized, and the overall flexibility and economy can be improved. On the other hand, the privacy issues of virtual power plants when participating in dispatching have been ignored. The privacy information of virtual power plants involves commercial secrets. Therefore, it is necessary to explore effective privacy protection optimization methods to ensure information security and confidentiality. Summary of the invention

[0006] In order to solve the above problems, the present invention proposes a method for distributed resource aggregation and coordinated scheduling of a virtual power plant, which can achieve the safety and economy of the overall system operation while taking into account the privacy and interests of each community.

[0007] To achieve the above object, the present invention adopts the following technical solution:

[0008] A method for aggregating and coordinating distributed resources of a virtual power plant comprises the following steps:

[0009] S1. Establish a feasible domain solution model for virtual power plants considering the uncertainty of renewable energy generation;

[0010] S2. The feasible domain model of the virtual power plant is solved by using the vertex enumeration method to obtain the feasible domain of the virtual power plant, which represents the active-reactive regulation capability of the distributed resource aggregate in the virtual power plant;

[0011] S3. Based on the feasible region of virtual power plants, a collaborative optimization model of multiple virtual power plants in active distribution networks is established;

[0012] S4. The ADMM algorithm is used to solve the collaborative optimization model of multiple virtual power plants in active distribution networks.

[0013] The construction process of the feasible domain solution model of the virtual power plant considering the uncertainty of renewable energy generation in step 1 is as follows:

[0014] S11. Establish a distributed resource model in a virtual power plant. The distributed resource equipment in a virtual power plant can be divided into two categories: energy storage-like distributed resources and generator-like distributed resources. Among them, energy storage-like distributed resources mainly include distributed energy storage and electric vehicles and other resources with energy accumulation; generator-like distributed resources include traditional distributed generators represented by micro gas turbines, and renewable distributed generators represented by distributed wind power and distributed photovoltaics; the models of various types of distributed resources are:

[0015] The model of the ES is:

[0016]

[0017] In the formula, renewable distributed generators are referred to as RDG in English; virtual power plant is referred to as VPP in English; energy storage is referred to as ES in English; conventional distributed generators are referred to as CDG in English; renewable distributed generators are referred to as RDG in English; is the set of all ES nodes belonging to the ath VPP, is the minimum permissible capacity of the energy storage connected to node i, is the maximum permissible capacity of the energy storage connected to node i, is the energy of the energy storage connected to node i at time t, is the charging power of the energy storage connected to node i at time t, is the discharge power of the energy storage connected to node i at time t, is the charging efficiency of the energy storage connected to node i, is the discharge efficiency of the energy storage connected to node i, ΔT is the control period, is the state variable corresponding to the energy storage connected to node i at time t, 0 for discharge and 1 for charge, is the upper limit of the energy storage charging power connected to node i, is the upper limit of the energy storage discharge power connected to node i;

[0018] The model of the CDG is:

[0019]

[0020] In the formula, is the set of all nodes of the CDG belonging to the ath VPP, is the active output of the CDG connected to node i at time t, is the reactive power output of the CDG connected to node i at time t, is the upper limit of the hourly ramp rate of the CDG connected to node i, is the upper limit of the active output of the CDG connected to node i, is the lower limit of the active output of the CDG connected to node i, is the upper limit of the reactive power output of the CDG connected to node i, is the lower limit of the reactive power output of the CDG connected to node i.

[0021] The model of the RDG is:

[0022]

[0023] In the formula, is the set of all nodes of the RDG belonging to the ath VPP, is the active output of the RDG connected to node i at time t, is the reactive power output of the RDG connected to node i at time t, is the maximum active output that can be provided by the RDG connected to node i at time t, is the minimum power factor angle of the RDG connected to node i, is the maximum power factor angle of the RDG connected to node i.

[0024] The Gaussian mixture model will include random variables The chance constraint is transformed into a deterministic constraint. First, the random variable The constraint is transformed into a chance constraint form:

[0025]

[0026] In the formula, For events The probability of establishment, κ RDG is the confidence corresponding to the chance constraint.

[0027] Secondly, transform the above chance constraints into deterministic constraints:

[0028]

[0029] In the formula, is the quantile point, is the inverse of the cumulative distribution function.

[0030] S12. Establish a network model. The power flow analysis of the power network topology adopts the DistFlow linear power flow model, which is:

[0031]

[0032]

[0033] Where: Θ a is the set of all nodes belonging to the ath VPP, PCC is the set of lines connecting PCC, is the active power of node j at time t, is the reactive power of node j at time t, is the number of CDGs connected to node j, is the number of RDGs connected at node j, is the number of ES connected to node j, is the active output of the lth CDG connected to node j at time t, is the reactive power output of the lth CDG connected to node j at time t, is the active output of the lth RDG connected to node j at time t, is the reactive power output of the lth RDG connected to node j at time t, is the charging power of the lth ES connected to node j at time t, is the discharge power of the lth ES connected to node j at time t, is the active power of line ij at time t, is the reactive power of line ij at time t, r ij is the resistance of line ij, x ij is the reactance of line ij, I ij,t is the square value of the current in line ij, is the upper limit of the square of the current, I i j is the lower limit of the square of the current, v i,t is the square value of the voltage at node i at time t; is the upper limit of the square of the voltage at node i, v i is the lower limit of the square of the voltage at node i, P a,t is the output active power of the ath VPP at PCC at time t, Q a,t They are the output reactive power of the ath VPP at PCC at time t respectively.

[0034] The step 2 adopts the vertex enumeration method to solve the feasible domain model of the virtual power plant, wherein, under the premise of considering the network constraints and distributed resource constraints of the virtual power plant, the power adjustable range of each distributed resource in the virtual power plant at each time is mapped to the point of common coupling (PCC) between the virtual power plant and the distribution network, so that the power adjustable range at each time at the PCC is used to represent the feasible domain of the virtual power plant. The process is:

[0035] 1) The objective function for enumerating the vertices of the feasible region of the virtual power plant is:

[0036]

[0037] In the formula, The unit direction vector for enumerating new vertices, is the power factor angle at time t, W a,t =[Q a,t ,P a,t ] is a point in the feasible region at time t, is the feasible domain of the ath VPP at time t.

[0038] 2) Determine the constraints for enumerating the vertices of the feasible domain of the virtual power plant, including the distributed resource model and network model in the virtual power plant of claim 2, and further, determine the specific power factor angle For the constraints of the feasible domain, set the constraints as follows:

[0039]

[0040] Then determine the optimization problem F0 for solving the feasible domain of the virtual power plant.

[0041] 3) Solve for different power factor angles The optimization problem F0 is solved to obtain all boundary vertices of the feasible domain of the virtual power plant, and finally the feasible domain of the virtual power plant is obtained by forming a convex hull based on each vertex.

[0042] The construction process of the active distribution network multi-virtual power plant collaborative optimization model in step S3 is as follows:

[0043] S31. Aggregate the active power costs of each distributed resource within the virtual power plant to obtain a piecewise quadratic function aggregation cost model for the active power output of the virtual power plant:

[0044]

[0045]

[0046] In the formula, is the active cost of the ath VPP at time t, is the cost of CDG in the ath VPP at time t, is the cost of the RDG in the ath VPP at time t, is the cost of the ES in the ath VPP at time t, is the quadratic cost coefficient of CDG in the ath VPP, is the linear cost coefficient of CDG in the ath VPP, is the constant cost coefficient of CDG in the ath VPP, is the first-order cost coefficient of RDG in the ath VPP, is the constant cost coefficient of the RDG in the ath VPP, is the first-order cost coefficient of ES in the ath VPP, is the constant cost coefficient of ES in the ath VPP, The ath VPP output is active at N B The quadratic cost coefficient for the segment, The ath VPP output is active at N B The first-order cost coefficient of the segment, The ath VPP output is active at N B The constant cost coefficient for the segment, The ath VPP output is active at N B The boundary value of the segment.

[0047] S32. The reactive power cost of a virtual power plant is the cost of active power generation lost by the virtual power plant due to providing reactive services, and its expression is:

[0048]

[0049] In the formula, is the reactive power cost of the ath VPP at time t, is the reactive power cost coefficient of the ath VPP, is the maximum active power of the ath VPP, is the maximum apparent power of the ath VPP.

[0050] S33. The goal of collaborative optimization of multiple virtual power plants in active distribution networks is to minimize the total cost of the system, which is expressed as:

[0051]

[0052] Where C is the total cost of the system, C a,t is the cost of the ath VPP at time t, N is the number of virtual power plants included in the distribution network, and Γ is the scheduling period;

[0053] S34. Applying active distribution network with multiple virtual power plants

[0054] The constraints for collaborative optimization are as follows:

[0055]

[0056] In the formula, is the total active power demand of the system at time t, is the total reactive power demand of the system at time t, is the total active power loss of the system at time t, is the total reactive power loss of the system at time t, u i,t is the voltage of node i at time t, is the maximum value of the node voltage, u is the minimum value of the node voltage, and Θ is the node set of the distribution network.

[0057] The process of using the ADMM algorithm to solve the active distribution network multi-virtual power plant collaborative optimization model in step S4 is:

[0058] S41. Decouple the coupling constraints by introducing auxiliary variables, expressed as:

[0059]

[0060]

[0061] In the formula, and P a,t and Q a,t The corresponding auxiliary variables.

[0062] S42. Establishing multiple virtual power plants in active distribution networks

[0063] The augmented Lagrangian equation of the collaborative optimization model is:

[0064]

[0065] Where L is the augmented Lagrangian equation, λ a,t For Constraint The corresponding Lagrange multiplier, μ a,t For Constraint The corresponding Lagrange multiplier, ρ is the penalty factor.

[0066] S43. Use the ADMM algorithm to perform distributed iterative optimization on the optimization scheduling model.

[0067] The process of using the ADMM algorithm to perform distributed iterative optimization on the optimization scheduling model in step S43 is as follows:

[0068] S430. Each virtual power plant independently and in parallel solves its own autonomous optimization sub-problem, which is expressed as:

[0069]

[0070] In the formula, is the output active power value of the ath VPP at the PCC at time t of the k+1th iteration, is the output reactive power value of the ath VPP at the PCC at time t of the k+1th iteration, Auxiliary variables The value of the kth iteration, Auxiliary variables The value of the kth iteration, is the Lagrange multiplier λ a,t The value of the kth iteration, is the Lagrange multiplier μ a,t The value of the kth iteration.

[0071] S431. The upper distribution system operator (DSO) updates the auxiliary variables, which are expressed as:

[0072]

[0073] In the formula, Auxiliary variables The value of the k+1th iteration, Auxiliary variables The value of the k+1th iteration.

[0074] S432. Update the Lagrange multiplier, expressed as:

[0075]

[0076] In the formula, is the Lagrange multiplier λ a,t The value of the k+1th iteration, is the Lagrange multiplier μ a,t The value of the k+1th iteration.

[0077] S433. Determine whether the optimization result meets the convergence condition. If the original residual and the dual residual of the optimization result are less than the set threshold, it is determined that the optimal scheduling strategy is achieved at this time, and the optimization result is output; if the optimization result does not meet the convergence condition, the number of iterations is increased by one, and the above process is repeated. The convergence condition is:

[0078]

[0079] In the formula, r k and d k are the original residual and dual residual of the kth iteration, ∈ pri and ∈ dual are the convergence criteria of the primal residual and the dual residual, respectively. Auxiliary variables The value of the k-1th iteration, Auxiliary variables The value of the k-1th iteration.

[0080] The beneficial effects of the present invention are:

[0081] 1. The method of the present invention can effectively characterize the feasible domain of multiple types of distributed resources in a single virtual power plant, fully utilize the active-reactive regulation capability of distributed resources, and enhance their flexibility in participating in power grid operation.

[0082] 2. The distributed collaborative scheduling method based on ADMM proposed in the method of the present invention can realize the unified coordination of multiple virtual power plants and the independent autonomous scheduling of each virtual power plant, and has good convergence and scalability. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 is a flow chart of the method of the present invention;

[0084] Figure 2 A flowchart of solving the feasible domain of a virtual power plant using the vertex enumeration method of the present invention;

[0085] Figure 3 This is a schematic diagram of a distributed resource aggregation and collaborative scheduling scenario of a virtual power plant according to an embodiment of the present invention. DETAILED DESCRIPTION

[0086] Below, the present invention is further described in conjunction with the accompanying drawings and specific embodiments:

[0087] Reference Figure 1 The present invention provides a method for aggregating and coordinating distributed resources of a virtual power plant, which specifically includes:

[0088] S1. Establish a feasible domain solution model for virtual power plants considering the uncertainty of renewable energy generation;

[0089] S2. The feasible domain model of the virtual power plant is solved by using the vertex enumeration method to obtain the feasible domain of the virtual power plant, which represents the active-reactive regulation capability of the distributed resource aggregate in the virtual power plant;

[0090] S3. Based on the feasible region of virtual power plants, a collaborative optimization model of multiple virtual power plants in active distribution networks is established;

[0091] S4. The ADMM algorithm is used to solve the collaborative optimization model of multiple virtual power plants in active distribution networks.

[0092] In an embodiment of the present invention, the construction process of the feasible region solution model of a virtual power plant considering the uncertainty of renewable energy power generation is as follows:

[0093] S11. Establish a distributed resource model in a virtual power plant. The distributed resource equipment in a virtual power plant can be divided into two categories: energy storage-like distributed resources and generator-like distributed resources. Among them, energy storage-like distributed resources mainly include distributed energy storage and electric vehicles and other resources with energy accumulation; generator-like distributed resources include traditional distributed generators represented by micro gas turbines, and renewable distributed generators represented by distributed wind power and distributed photovoltaics; the models of various types of distributed resources are:

[0094] The model of the ES is:

[0095]

[0096] In the formula, renewable distributed generators are referred to as RDG in English; virtual power plant is referred to as VPP in English; energy storage is referred to as ES in English; conventional distributed generators are referred to as CDG in English; renewable distributed generators are referred to as RDG in English; is the set of all ES nodes belonging to the ath VPP, is the minimum permissible capacity of the energy storage connected to node i, is the maximum permissible capacity of the energy storage connected to node i, is the energy of the energy storage connected to node i at time t, is the charging power of the energy storage connected to node i at time t, is the discharge power of the energy storage connected to node i at time t, is the charging efficiency of the energy storage connected to node i, is the discharge efficiency of the energy storage connected to node i, ΔT is the control period, is the state variable corresponding to the energy storage connected to node i at time t, 0 for discharge and 1 for charge, is the upper limit of the energy storage charging power connected to node i, is the upper limit of the energy storage discharge power connected to node i;

[0097] The model of the CDG is:

[0098]

[0099]

[0100] In the formula, is the set of all nodes of the CDG belonging to the ath VPP, is the active output of the CDG connected to node i at time t, is the reactive power output of the CDG connected to node i at time t, is the upper limit of the hourly ramp rate of the CDG connected to node i, is the upper limit of the active output of the CDG connected to node i, is the lower limit of the active output of the CDG connected to node i, is the upper limit of the reactive power output of the CDG connected to node i, is the lower limit of the reactive power output of the CDG connected to node i.

[0101] The model of the RDG is:

[0102]

[0103] In the formula, is the set of all nodes of the RDG belonging to the ath VPP, is the active output of the RDG connected to node i at time t, is the reactive power output of the RDG connected to node i at time t, is the maximum active output that can be provided by the RDG connected to node i at time t, is the minimum power factor angle of the RDG connected to node i, is the maximum power factor angle of the RDG connected to node i.

[0104] The Gaussian mixture model will include random variables The chance constraint is transformed into a deterministic constraint. First, the random variable The constraint is transformed into a chance constraint form:

[0105]

[0106] In the formula, For events The probability of establishment, κ RDG is the confidence corresponding to the chance constraint.

[0107] Secondly, transform the above chance constraints into deterministic constraints:

[0108]

[0109] In the formula, is the quantile point, is the inverse of the cumulative distribution function.

[0110] S12. Establish a network model. The power flow analysis of the power network topology adopts the DistFlow linear power flow model, which is:

[0111]

[0112] Where: Θ a is the set of all nodes belonging to the ath VPP, PCC is the set of lines connecting PCC, is the active power of node j at time t, is the reactive power of node j at time t, is the number of CDGs connected to node j, is the number of RDGs connected at node j, is the number of ES connected to node j, is the active output of the lth CDG connected to node j at time t, is the reactive power output of the lth CDG connected to node j at time t, is the active output of the lth RDG connected to node j at time t, is the reactive power output of the lth RDG connected to node j at time t, is the charging power of the lth ES connected to node j at time t, is the discharge power of the lth ES connected to node j at time t, is the active power of line ij at time t, is the reactive power of line ij at time t, r ij is the resistance of line ij, x ij is the reactance of line ij, I ij,t is the square value of the current in line ij, is the upper limit of the square of the current, I i j is the lower limit of the square of the current, v i,t is the square value of the voltage at node i at time t; v i is the upper limit of the square of the voltage at node i, v i is the lower limit of the square of the voltage at node i, P a,t is the output active power of the ath VPP at PCC at time t, Q a,t They are the output reactive power of the ath VPP at PCC at time t respectively.

[0113] The feasible domain model of the virtual power plant is solved by the vertex enumeration method. Under the premise of considering the network constraints and distributed resource constraints of the virtual power plant, the power adjustable range of each distributed resource in the virtual power plant at each time is mapped to the point of common coupling (PCC) between the virtual power plant and the distribution network, so that the power adjustable range at each time at the PCC is used to represent the feasible domain of the virtual power plant. The process is as follows:

[0114] 1) The objective function for enumerating the vertices of the feasible region of the virtual power plant is:

[0115]

[0116] In the formula, The unit direction vector for enumerating new vertices, is the power factor angle at time t, W a,t =[Q a,t ,P a,t ] is a point in the feasible region at time t, is the feasible domain of the ath VPP at time t.

[0117] 2) Determine the constraints for enumerating the vertices of the feasible domain of the virtual power plant, including the distributed resource model and network model in the virtual power plant of claim 2, and further, determine the specific power factor angle For the constraints of the feasible domain, set the constraints as follows:

[0118]

[0119] Then determine the optimization problem F0 for solving the feasible domain of the virtual power plant.

[0120] 3) Solve for different power factor angles The optimization problem F0 is solved to obtain all the boundary vertices of the feasible region of the virtual power plant, and finally the feasible region of the virtual power plant is obtained by forming a convex hull based on each vertex. The specific process is as follows Figure 2 shown.

[0121] The active distribution network multi-virtual power plant collaborative optimization model in step S3 specifically includes:

[0122] S31. Aggregate the active power costs of each distributed resource within the virtual power plant to obtain a piecewise quadratic function aggregation cost model for the active power output of the virtual power plant:

[0123]

[0124] In the formula, is the active cost of the ath VPP at time t, is the cost of CDG in the ath VPP at time t, is the cost of the RDG in the ath VPP at time t, is the cost of the ES in the ath VPP at time t, is the quadratic cost coefficient of CDG in the ath VPP, is the linear cost coefficient of CDG in the ath VPP, is the constant cost coefficient of CDG in the ath VPP, is the first-order cost coefficient of RDG in the ath VPP, is the constant cost coefficient of the RDG in the ath VPP, is the first-order cost coefficient of ES in the ath VPP, is the constant cost coefficient of ES in the ath VPP, The ath VPP output is active at N B The quadratic cost coefficient for the segment, The ath VPP output is active at N B The first-order cost coefficient of the segment, The ath VPP output is active at N B The constant cost coefficient for the segment, The ath VPP output is active at N B The boundary value of the segment.

[0125] S32. The reactive power cost of a virtual power plant is the cost of active power generation lost by the virtual power plant due to providing reactive services, and its expression is:

[0126]

[0127] In the formula, is the reactive power cost of the ath VPP at time t, is the reactive power cost coefficient of the ath VPP, is the maximum active power of the ath VPP, is the maximum apparent power of the ath VPP.

[0128] S33. Active distribution network with multiple virtual power plants

[0129] The goal of collaborative optimization is to minimize the total cost of the system, which is expressed as:

[0130]

[0131] Where C is the total cost of the system, C a,t is the cost of the ath VPP at time t, N is the number of virtual power plants included in the distribution network, and Γ is the scheduling period;

[0132] S34. Apply the constraints for collaborative optimization of multiple virtual power plants in active distribution networks as follows:

[0133]

[0134]

[0135] In the formula, is the total active power demand of the system at time t, is the total reactive power demand of the system at time t, is the total active power loss of the system at time t, is the total reactive power loss of the system at time t, u i,t is the voltage of node i at time t, is the maximum value of the node voltage, u is the minimum value of the node voltage, and Θ is the node set of the distribution network.

[0136] In this embodiment, based on the above-established active distribution network multi-virtual power plant collaborative optimization model, the ADMM algorithm is used in step S4 to solve the active distribution network multi-virtual power plant collaborative optimization model, which specifically includes:

[0137] S41. Decouple the coupling constraints by introducing auxiliary variables, expressed as:

[0138]

[0139] In the formula, and P a,t and Q a,t The corresponding auxiliary variables.

[0140] S42. The augmented Lagrangian equation for establishing a collaborative optimization model of multiple virtual power plants in active distribution networks is:

[0141]

[0142] Where L is the augmented Lagrangian equation, λ a,t For Constraint The corresponding Lagrange multiplier, μ a,t For Constraint The corresponding Lagrange multiplier, ρ is the penalty factor.

[0143] S43. Use the ADMM algorithm to perform distributed iterative optimization on the optimization scheduling model.

[0144] The distributed iterative optimization using the ADMM algorithm in step S43 includes the following steps:

[0145] S430. Each virtual power plant independently and in parallel solves its own autonomous optimization sub-problem, which is expressed as:

[0146]

[0147] In the formula, is the output active power value of the ath VPP at the PCC at time t of the k+1th iteration, is the output reactive power value of the ath VPP at the PCC at time t of the k+1th iteration, Auxiliary variables The value of the kth iteration, Auxiliary variables The value of the kth iteration, is the Lagrange multiplier λ a,t The value of the kth iteration, is the Lagrange multiplier μ a,t The value of the kth iteration.

[0148] S431. The upper distribution system operator (DSO) updates the auxiliary variables, which are expressed as:

[0149]

[0150] In the formula, Auxiliary variables The value of the k+1th iteration, Auxiliary variables The value of the k+1th iteration.

[0151] S432. Update the Lagrange multiplier, expressed as:

[0152]

[0153] In the formula, is the Lagrange multiplier λ a,t The value of the k+1th iteration, is the Lagrange multiplier μ a,t The value of the k+1th iteration.

[0154] S433. Determine whether the optimization result meets the convergence condition. If the original residual and the dual residual of the optimization result are less than the set threshold, it is determined that the optimal scheduling strategy is achieved at this time, and the optimization result is output; if the optimization result does not meet the convergence condition, the number of iterations is increased by one, and the above process is repeated. The convergence condition is:

[0155]

[0156] In the formula, rk and d k are the original residual and dual residual of the kth iteration, ∈ pri and ∈ dual are the convergence criteria of the primal residual and the dual residual, respectively. Auxiliary variables The value of the k-1th iteration, Auxiliary variables The value of the k-1th iteration.

[0157] At this point, the distributed resource aggregation and collaborative scheduling model of the virtual power plant has been built. The schematic diagram of the distributed resource aggregation and collaborative scheduling scenario of the virtual power plant described in the embodiment is as follows: Figure 3 shown.

[0158] The virtual power plant distributed resource aggregation and coordination method of the present invention establishes a feasible domain of virtual power plants that takes into account the uncertainty of renewable energy generation, proposes a distributed collaborative scheduling strategy for multiple virtual power plants based on the ADMM algorithm, completes the distributed solution of the distributed resource aggregation scheduling of the power grid, can converge quickly and maintain high accuracy, and when the algorithm converges, all constraints are satisfied, and the objective function is optimized. The proposed distributed collaborative scheduling method based on ADMM can achieve unified coordination of multiple virtual power plants and independent autonomous scheduling of each virtual power plant, and has good convergence and scalability.

[0159] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A method for distributed resource aggregation and coordinated scheduling of a virtual power plant, characterized in that: The following steps are involved: S1. Establish a feasible domain solution model for virtual power plants considering the uncertainty of renewable energy generation; S2. The feasible domain model of the virtual power plant is solved by using the vertex enumeration method to obtain the feasible domain of the virtual power plant, which represents the active-reactive regulation capability of the flexible resource aggregate in the virtual power plant; S3. Based on the feasible region of virtual power plants, a collaborative optimization model of multiple virtual power plants in active distribution networks is established; S4. The ADMM algorithm is used to solve the collaborative optimization model of multiple virtual power plants in active distribution networks.

2. A method for distributed resource aggregation and coordinated scheduling of a virtual power plant according to claim 1, characterized in that: The process of step S1 includes: S11. Establish a distributed resource model in a virtual power plant. The distributed resource equipment in a virtual power plant can be divided into two categories: energy storage-like distributed resources and generator-like distributed resources. Among them, energy storage-like distributed resources mainly include distributed energy storage and electric vehicles and other resources with energy accumulation; generator-like distributed resources include traditional distributed generators represented by micro gas turbines, and renewable distributed generators represented by distributed wind power and distributed photovoltaics; the models of various types of distributed resources are: The model of the ES is: In the formula, renewable distributed generators are referred to as RDG in English; virtual power plant is referred to as VPP in English; energy storage is referred to as ES in English; conventional distributed generators are referred to as CDG in English; renewable distributed generators are referred to as RDG in English; is the set of all ES nodes belonging to the ath VPP, is the minimum permissible capacity of the energy storage connected to node i, is the maximum permissible capacity of the energy storage connected to node i, is the energy of the energy storage connected to node i at time t, is the charging power of the energy storage connected to node i at time t, is the discharge power of the energy storage connected to node i at time t, is the charging efficiency of the energy storage connected to node i, is the discharge efficiency of the energy storage connected to node i, ΔT is the control period, is the state variable corresponding to the energy storage connected to node i at time t, 0 for discharge and 1 for charge, is the upper limit of the energy storage charging power connected to node i, is the upper limit of the energy storage discharge power connected to node i; The model of the CDG is: In the formula, is the set of all nodes of the CDG belonging to the ath VPP, is the active output of the CDG connected to node i at time t, is the reactive power output of the CDG connected to node i at time t, ΔP i CHP is the upper limit of the hourly ramp rate of the CDG connected to node i, is the upper limit of the active output of the CDG connected to node i, P i CHP is the lower limit of the active output of the CDG connected to node i, is the upper limit of the reactive power output of the CDG connected to node i, is the lower limit of the reactive power output of the CDG connected to node i. The model of the RDG is: In the formula, is the set of all nodes of the RDG belonging to the ath VPP, is the active output of the RDG connected to node i at time t, is the reactive power output of the RDG connected to node i at time t, is the maximum active output that can be provided by the RDG connected to node i at time t, is the minimum power factor angle of the RDG connected to node i, is the maximum power factor angle of the RDG connected to node i. S12. Establish a network model. The power flow analysis of the power network topology adopts the DistFlow linear power flow model, which is: Where: Θ a is the set of all nodes belonging to the ath VPP, PCC is the set of lines connecting PCC, is the active power of node j at time t, is the reactive power of node j at time t, is the number of CDGs connected to node j, is the number of RDGs connected at node j, is the number of ES connected to node j, is the active output of the lth CDG connected to node j at time t, is the reactive power output of the lth CDG connected to node j at time t, is the active output of the lth RDG connected to node j at time t, is the reactive power output of the lth RDG connected to node j at time t, is the charging power of the lth ES connected to node j at time t, is the discharge power of the lth ES connected to node j at time t, is the active power of line ij at time t, is the reactive power of line ij at time t, r ij is the resistance of line ij, x ij is the reactance of line ij, I ij,t is the square value of the current in line ij, is the upper limit of the square of the current, I i j is the lower limit of the square of the current, v i,t is the square value of the voltage at node i at time t; is the upper limit of the square of the voltage at node i, v i is the lower limit of the square of the voltage at node i, P a,t is the output active power of the ath VPP at PCC at time t, Q a,t They are the output reactive power of the ath VPP at PCC at time t respectively.

3. A method for distributed resource aggregation and coordinated scheduling of a virtual power plant according to claim 2, characterized in that: The Gaussian mixture model is used to establish the RDG model considering the uncertainty of renewable energy generation. The process is as follows: 1) Contain random variables The constraint is transformed into a chance constraint form: In the formula, For events The probability of establishment, κ RDG is the confidence corresponding to the chance constraint; 2) Convert the above chance constraints into deterministic constraints: In the formula, is the quantile point, is the inverse of the cumulative distribution function.

4. A method for distributed resource aggregation and coordinated scheduling of a virtual power plant according to claim 1, characterized in that: The feasible domain model of the virtual power plant is solved by the vertex enumeration method. Under the premise of considering the network constraints and distributed resource constraints of the virtual power plant, the power adjustable range of each distributed resource in the virtual power plant at each time is mapped to the point of common coupling (PCC) between the virtual power plant and the distribution network, so that the power adjustable range at each time at the PCC is used to represent the feasible domain of the virtual power plant. The process is as follows: (1) The objective function for enumerating the vertices of the feasible region of the virtual power plant is: In the formula, The unit direction vector for enumerating new vertices, is the power factor angle at time t, W a,t =[Q a,t ,P a,t ] is a point in the feasible region at time t, is the feasible domain of the ath VPP at time t. (2) The constraint conditions for enumerating the vertices of the feasible domain of the virtual power plant include the distributed resource model and the network model in the virtual power plant of claim 2, and in addition, to determine the specific power factor angle For the constraints of the feasible domain, set the constraints as follows: Then determine the optimization problem F0 for solving the feasible domain of the virtual power plant. (3) Solving for different power factor angles The optimization problem F0 is solved to obtain all boundary vertices of the feasible domain of the virtual power plant, and finally the feasible domain of the virtual power plant is obtained by forming a convex hull based on each vertex.

5. A method for distributed resource aggregation and coordinated scheduling of a virtual power plant according to claim 1, characterized in that: The process of establishing the active distribution network multi-virtual power plant collaborative optimization model in step S3 is: S31. Aggregate the active power costs of each distributed resource within the virtual power plant to obtain a piecewise quadratic function aggregation cost model for the active power output of the virtual power plant: In the formula, is the active cost of the ath VPP at time t, is the cost of CDG in the ath VPP at time t, is the cost of the RDG in the ath VPP at time t, is the cost of the ES in the ath VPP at time t, is the quadratic cost coefficient of CDG in the ath VPP, is the linear cost coefficient of CDG in the ath VPP, is the constant cost coefficient of CDG in the ath VPP, is the first-order cost coefficient of RDG in the ath VPP, is the constant cost coefficient of the RDG in the ath VPP, is the first-order cost coefficient of ES in the ath VPP, is the constant cost coefficient of ES in the ath VPP, The ath VPP output is active at N B The quadratic cost coefficient for the segment, The ath VPP output is active at N B The first-order cost coefficient of the segment, The ath VPP output is active at N B The constant cost coefficient for the segment, The ath VPP output is active at N B Boundary value of the segment; S32. The reactive power cost of a virtual power plant is the cost of active power generation lost by the virtual power plant due to providing reactive services, and its expression is: In the formula, is the reactive power cost of the ath VPP at time t, is the reactive power cost coefficient of the ath VPP, is the maximum active power of the ath VPP, is the maximum apparent power of the ath VPP; S33. The goal of collaborative optimization of multiple virtual power plants in active distribution networks is to minimize the total cost of the system, which is expressed as: Where C is the total cost of the system, C a,t is the cost of the ath VPP at time t, N is the number of virtual power plants included in the distribution network, and Γ is the scheduling period; S34. Apply the constraints for collaborative optimization of multiple virtual power plants in active distribution networks as follows: Where P t D is the total active power demand of the system at time t, is the total reactive power demand of the system at time t, P t loss is the total active power loss of the system at time t, is the total reactive power loss of the system at time t, u i,t is the voltage of node i at time t, is the maximum value of the node voltage, u is the minimum value of the node voltage, and Θ is the node set of the distribution network.

6. A method for distributed resource aggregation and coordinated scheduling of a virtual power plant according to claim 1, characterized in that: The step S4 adopts the ADMM algorithm to solve the active distribution network multi-virtual power plant collaborative optimization model, and the process is: S41. Decouple the coupling constraints by introducing auxiliary variables, expressed as: In the formula, and P a,t and Q a,t The corresponding auxiliary variables; S42. The augmented Lagrangian equation for establishing the collaborative optimization model of multiple virtual power plants in active distribution networks is: Where L is the augmented Lagrangian equation, λ a,t For Constraint The corresponding Lagrange multiplier, μ a,t For Constraint The corresponding Lagrange multiplier, ρ is the penalty factor; S43. Use the ADMM algorithm to perform distributed iterative optimization on the optimization scheduling model.

7. A method for distributed resource aggregation and coordinated scheduling of a virtual power plant according to claim 6, characterized in that: The process of using the ADMM algorithm to perform distributed iterative optimization on the optimization scheduling model in step S43 is as follows: S430. Each virtual power plant independently and in parallel solves its own autonomous optimization sub-problem, which is expressed as: In the formula, is the output active power value of the ath VPP at the PCC at time t of the k+1th iteration, is the output reactive power value of the ath VPP at the PCC at time t of the k+1th iteration, Auxiliary variables The value of the kth iteration, Auxiliary variables The value of the kth iteration, is the Lagrange multiplier λ a,t The value of the kth iteration, is the Lagrange multiplier μ a,t The value of the kth iteration. S431. The upper distribution system operator (DSO) updates the auxiliary variables, which are expressed as: In the formula, Auxiliary variables The value of the k+1th iteration, Auxiliary variables The value of the k+1th iteration. S432. Update the Lagrange multiplier, expressed as: In the formula, is the Lagrange multiplier λ a,t The value of the k+1th iteration, is the Lagrange multiplier μ a,t The value of the k+1th iteration. S433. Determine whether the optimization result meets the convergence condition. If the original residual and the dual residual of the optimization result are less than the set threshold, it is determined that the optimal scheduling strategy is achieved at this time, and the optimization result is output; If the optimization result does not meet the convergence condition, the number of iterations is increased by one and the above process is repeated. The convergence condition is: In the formula, r k and d k are the original residual and dual residual of the kth iteration, ∈ pri and ∈ dual are the convergence criteria of the primal residual and the dual residual, respectively. Auxiliary variables The value of the k-1th iteration, Auxiliary variables The value of the k-1th iteration.

Citation Information

Cited By

  • Multi-virtual power plant-containing power system collaborative optimization method based on feasible region equivalent projection theory

    CN120150101A

  • Method for evaluating feasible region of mine virtual power plant under incomplete information

    CN120181614A

  • Feasible region evaluation method for mine virtual power plant under incomplete information

    CN120181614B

  • Virtual power plant resource flexibility aggregation method and device based on linear programming projection

    CN121457092A

  • Virtual power plant aggregation resource regulation and control method and system

    CN122620427A