Virtual power plant transaction auxiliary calculation method and system

Abstract

The invention relates to a virtual power plant transaction auxiliary calculation method and system, and belongs to the technical field of participation of virtual power plants in power markets. According to the method, a day-ahead virtual power plant optimization model which takes the lowest total cost as an optimization target and comprises the corresponding cost of the day-ahead bidding power of a virtual power plant to an energy market, the power of a virtual power plant flexible load aggregator and the dispatching power of a virtual power plant generator set and energy storage equipment is adopted; obtaining a power track reference value of the flexible load aggregator under constraint conditions corresponding to the power flow equation, the distributed generator set, the energy storage equipment and the flexible load aggregator; calculating a power track actual value of the flexible load aggregator according to a power consumption curve of a flexible load for determining a flexible load operation plan obtained after depolymerization optimization is performed on the reference value, and substituting the power track actual value into a day-ahead virtual power plant optimization model; and obtaining the day-ahead bidding power for determining the day-ahead bidding plan and the scheduling power for determining the day-ahead scheduling decision plan under the constraint condition.

Description

Technical Field

[0001] This invention belongs to the technical field of virtual power plants participating in the electricity market, and specifically relates to a virtual power plant transaction auxiliary calculation method and system. Background Technology

[0002] Distributed generation will become an important energy source for the future power grid. The addition of a large number of active load nodes on the demand side makes the operation of the distribution network more complex and flexible. However, the randomness of distributed generation also has certain impacts on grid operation. The development of power sources and loads will lead to the future distribution network exhibiting characteristics such as distributed small capacity, diverse resource types, bidirectional power flow, and interactivity. Virtual power plant (VPP) technology effectively connects distributed resources with the power system, realizing resource integration and allocation, and balancing the dual roles of grid safety and stability operation and the electricity market. It is a feasible way for smart grids to achieve interactivity and intelligence on the energy supply side; it improves the overall stability of the power system, reduces the dispatching burden, and realizes the efficient utilization of power system resources.

[0003] In existing technologies, the main resources of virtual power plants include distributed generation resources, energy storage resources, and load resources. However, the resource allocation of virtual power plants in existing technologies often focuses primarily on distributed generation resources and energy storage resources, without considering the impact of the allocation and configuration of load resources on the operational economics of the virtual power plant. This may lead to situations where the trading and operation of virtual power plants are less economical. Summary of the Invention

[0004] The purpose of this invention is to provide a virtual power plant transaction auxiliary calculation method and system to solve the problem that the existing technology does not consider the impact of load resource pair allocation and setting on the operating economy of virtual power plants, which may lead to poor transaction operation economy of virtual power plants.

[0005] To address the aforementioned technical problems, this invention provides a virtual power plant transaction auxiliary calculation method. This method includes: using a day-ahead virtual power plant optimization model with the goal of minimizing total cost, which incorporates the day-ahead bidding power of the virtual power plant to the energy market, the power of the virtual power plant's flexible load aggregator, and the corresponding costs of the dispatch power of the virtual power plant's generator units and energy storage devices. Under constraints related to power flow equations, distributed generator units, energy storage devices, and flexible load aggregators, a reference value for the power trajectory of the flexible load aggregator is obtained. Based on the power consumption curve of the flexible load used to determine the flexible load operation plan after de-aggregation optimization of the reference value, the actual value of the power trajectory of the flexible load aggregator is calculated. This actual value is then substituted into the day-ahead virtual power plant optimization model to obtain, under constraints, the day-ahead bidding power used to determine the day-ahead bidding plan and the dispatch power used to determine the day-ahead dispatch decision plan.

[0006] Furthermore, the method includes: using a real-time virtual power plant optimization model with the goal of minimizing total cost, which includes the real-time bidding power of the virtual power plant to the energy market, the power of the virtual power plant flexible load aggregator, and the corresponding cost of the dispatch power of the virtual power plant generator sets and energy storage devices, under the constraints of power flow equations, distributed generator sets, energy storage devices, and flexible load aggregators, to obtain a reference value for the power trajectory of the flexible load aggregator; calculating the actual value of the power trajectory of the flexible load aggregator based on the power consumption curve of the flexible load used to determine the flexible load operation plan after de-aggregation optimization of the reference value, and then substituting it into the real-time virtual power plant optimization model to obtain the real-time bidding power used to determine the real-time bidding plan and the dispatch power used to determine the real-time dispatch decision plan under the constraints.

[0007] Furthermore, the optimization calculation method for the power consumption curve of the flexible load used to determine the flexible load operation plan after de-aggregation optimization of the reference value includes: de-aggregation optimization of an optimization model containing the function terms corresponding to the difference between the overall output power and the reference power trajectory corresponding to the power consumption curve of the flexible load in the flexible load aggregator and the function terms corresponding to the start and stop times of each flexible load with the objective of minimizing the value, to obtain the power consumption curve of the flexible load.

[0008] Furthermore, the optimization model aimed at minimizing the difference between the overall output power and the reference power trajectory corresponding to the power consumption curve of the flexible loads in the flexible load aggregator, and the function term corresponding to the number of start-stop cycles for each flexible load, can be optimized through the following methods:

[0009]

[0010] In the formula, N L,k,m N represents the number of flexible loads inside aggregator k at node m; AGG,m The number of aggregators at node m; M is a set constant; F1 DIS () represents the first optimization objective used to make the overall output power of the aggregator follow the reference power trajectory of the pre-optimization stage; F2 DIS () represents the second optimization objective for reducing the number of start-stop cycles per flexibility load; P L,n,t The actual power consumption of the nth flexible load at time t; T is the time interval; Let be the reference power consumption of aggregator k at node m at time t.

[0011] Furthermore, methods for calculating the actual power of the flexible load aggregator based on the power consumption curve of the flexible load include:

[0012]

[0013] In the formula, Let be the actual power consumption of aggregator k at node m at time t; n is the nth flexibility load at node m.

[0014] Furthermore, the current virtual power plant optimization model includes:

[0015]

[0016] In the formula, T represents the total number of operating periods of the virtual power plant; The cost corresponding to the day-ahead bid power of the virtual power plant in the energy market; Let t be the day-ahead forecast electricity price in the energy market. Let t be the power output of the virtual power plant bidding in the energy market. Let i be the active power output of generator unit i at time t; Let i be the start-stop state of generator set i at time t; and These are the active power input and active power output of energy storage device j at time t, respectively. F represents the power consumption of the flexible polymerizer k at node m at time t. i DG (·) and These represent the corresponding costs of the dispatch power of virtual power plant generator unit i, virtual power plant energy storage device j, and the corresponding costs of the power of virtual power plant flexible load aggregator k, respectively; N DG N represents the number of generator sets. ESS N represents the number of energy storage devices. AGG This represents the number of flexible load aggregators.

[0017] Furthermore, the real-time virtual power plant optimization model includes:

[0018]

[0019] In the formula, T represents the total number of operating periods of the virtual power plant; The cost corresponding to the real-time bidding power of a virtual power plant in the energy market; Let t be the current energy market electricity price. The active power of the virtual power plant interacting with the energy market at time t; The reference active power of the virtual power plant and the energy market at time t; Let i be the active power output of generator unit i at time t; and These represent the active power output and active power input of energy storage device j at time t, respectively. Let be the power consumption of node m in the flexible aggregator k at time t; F is the deviation penalty factor.i DG (·) and These represent the corresponding costs of the dispatch power of virtual power plant generator unit i, virtual power plant energy storage device j, and the corresponding costs of the power of flexible load aggregator k, respectively; N DG N represents the number of generator sets. ESS N represents the number of energy storage devices. AGG This represents the number of flexible load aggregators.

[0020] Furthermore, the constraints corresponding to the power flow equations include:

[0021]

[0022]

[0023] In the formula, N B r represents the total number of nodes in the virtual power plant. pq and x pq These represent the resistance and reactance between line p and line q, respectively; θ m V m P m and Q m These represent the phase, voltage magnitude, injected active power, and injected reactive power at node m, respectively. Let be the power consumption of aggregator k at node m at time t; and These represent the active and reactive loads at node m at time t, respectively; P pq and Q pq These represent the active power flow and reactive power flow between line p and line q, respectively. Let t be the power output of the virtual power plant bidding in the energy market. Let i be the active power output of generator unit i at time t; and These represent the active power input and active power output of energy storage device j at time t, respectively; k pq1 k is the first impedance parameter. pq2 θ is the second impedance parameter; p θ is the phase angle at node p; q V is the phase angle at node q; p V is the voltage at node p; q Let q be the voltage at node q; This refers to the reactive power provided by generator set i at time t.

[0024]

[0025] V m,min ≤V m ≤Vm,max

[0026] In the formula, S pq,N V is the rated capacity of the transmission line between line p and line q; m,max and V m,min These are the upper and lower limits of the node voltage amplitude, respectively.

[0027] The constraints for distributed generator sets include:

[0028]

[0029]

[0030] In the formula, and These are the upper and lower limits of the output of generator set i, respectively; The maximum climbing rate of generator set i; UT represents the rated capacity of generator set i; i DG and DT i DG These are the minimum power-on time and the minimum power-off time, respectively; y i,t and z i,t Integer variables introduced; Let i be the start-stop state of generator set i at time t; This refers to the reactive power provided by generator set i at time t.

[0031] The constraints for energy storage devices include:

[0032]

[0033] In the formula, and and are the maximum input / output power and minimum input / output power of energy storage device j, respectively; and These are the upper and lower limits of energy storage for energy storage device j, respectively; η in and η out These are the power input and output energy conversion coefficients, respectively; λ is the energy storage dissipation coefficient; ΔT is the energy storage operation time interval; Let j be the state of charge of the energy storage device at time t; Let J represent the initial state of charge of the energy storage device J.

[0034] The constraints for the flexible load aggregator include:

[0035]

[0036] In the formula, and These are the upper and lower limits of the power demand for aggregator k at node m, respectively; Let be the power consumption of aggregator k at node m at time t; and These are the start and end demand times for aggregator k at node m, respectively. and These are the upper and lower limits of the energy demand for aggregator k at node m, respectively. Let be the reference energy state of aggregator k at node m at time t.

[0037] Furthermore, the flexible load aggregator aggregates flexible loads by combining historical data of the demand parameters of the flexible loads with the NJW spectrum clustering algorithm. The demand parameters of the flexible loads include the lower and upper limits of the power demand of the flexible loads, the lower and upper limits of the energy demand of the flexible loads, and the start and end times of the demand of the flexible loads.

[0038] To address the aforementioned technical problems, the present invention also provides a virtual power plant transaction auxiliary calculation system, including a processor, which executes a computer program to implement the steps of the virtual power plant transaction auxiliary calculation method.

[0039] The beneficial effects of the above technical solution are as follows: This invention provides a new auxiliary calculation method for virtual power plant transactions. This method considers the impact of load resource allocation and setting on the operational economy of the virtual power plant. Based on the power trajectory reference value of the flexible load aggregator obtained by the virtual power plant optimization model under corresponding constraints, the actual power trajectory value of the flexible load aggregator is obtained after de-aggregation optimization of the power trajectory reference value. The actual power trajectory value of the flexible load aggregator is then substituted into the day-ahead virtual power plant optimization model to obtain the day-ahead bidding power and the day-ahead dispatch power of distributed generator sets and energy storage equipment. This is equivalent to first treating the day-ahead bidding power of the virtual power plant to the energy market, the power of the virtual power plant's flexible load aggregator, and the dispatch power of the virtual power plant's generator sets and energy storage equipment as unknowns to optimize the day-ahead virtual power plant's total cost. Based on the preliminary solution of the power trajectory reference value of the flexible load aggregator that minimizes the day-ahead virtual power plant's total cost, further optimization is performed. After de-aggregation optimization, the power consumption curve of the flexible load is obtained, which can be used to formulate flexible load operation plans and to calculate the actual value of the power trajectory of the flexible load aggregator. Using this actual value as a known quantity, the day-ahead bidding power of the virtual power plant to the energy market and the dispatch power of the virtual power plant's generators and energy storage devices are used as unknown quantities to re-optimize to minimize the total cost of the virtual power plant on the day-ahead. The corresponding power obtained in the end (i.e., the day-ahead bidding power of the virtual power plant to the energy market and the day-ahead dispatch power of the distributed generators and energy storage devices) can be used to formulate the guaranteed day-ahead bidding plan and day-ahead dispatch plan of the virtual power plant. Thus, the power information used to formulate the flexible load operation plan, the day-ahead bidding plan, and the day-ahead dispatch plan that can guarantee the minimum operating cost of the virtual power plant can be obtained. This solves the problem of poor economic efficiency of virtual power plant trading operations caused by the failure to consider the impact of load resource pair allocation and setting on the operating economy of the virtual power plant in the existing technology. Attached Figure Description

[0040] Figure 1 A flowchart illustrating the virtual power plant transaction auxiliary calculation method of this invention;

[0041] Figure 2 This is a schematic diagram illustrating the principle of the virtual power plant transaction auxiliary calculation method according to an embodiment of the present invention. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0043] Example of a virtual power plant transaction auxiliary calculation method:

[0044] This embodiment of a virtual power plant transaction auxiliary calculation method considers the situation of flexible loads in the virtual power plant. It de-aggregates the power trajectory reference values ​​of the flexible load aggregator that minimizes the total cost of the virtual power plant on the day-ahead to obtain the power consumption curve of the flexible load. Based on the power consumption curve of the flexible load, a flexible load operation plan is formulated. After determining the flexible load operation plan, the power of the virtual power plant's day-ahead bidding plan and day-ahead dispatch plan is further determined, thereby minimizing the total cost of the virtual power plant on the day-ahead. This solves the problem in the prior art that the impact of the allocation and setting of load resource pairs on the operating economy of the virtual power plant is not considered, which may lead to poor economic efficiency in the transaction operation of the virtual power plant.

[0045] Reference Figure 1 The specific calculation methods for virtual power plant transactions include:

[0046] A day-ahead virtual power plant optimization model, with the goal of minimizing total cost, incorporates the day-ahead bidding power of the virtual power plant to the energy market, the power of the virtual power plant's flexible load aggregator, and the corresponding costs of the dispatch power of the virtual power plant's generator sets and energy storage devices. Under the constraints of power flow equations, distributed generator sets, energy storage devices, and flexible load aggregators, a reference value for the power trajectory of the flexible load aggregator is obtained. Based on the power consumption curve of the flexible load used to determine the flexible load operation plan after de-aggregation optimization of the reference value, the actual value of the power trajectory of the flexible load aggregator is calculated and substituted into the day-ahead virtual power plant optimization model. Under the constraints, the day-ahead bidding power used to determine the day-ahead bidding plan and the dispatch power used to determine the day-ahead dispatch decision plan are obtained.

[0047] This embodiment considers the impact of load resource allocation and configuration on the operational economy of the virtual power plant. Based on the power trajectory reference value of the flexible load aggregator obtained from the virtual power plant optimization model under corresponding constraints, the actual power trajectory value of the flexible load aggregator is obtained after de-aggregation optimization. This actual power trajectory value is then substituted into the day-ahead virtual power plant optimization model to obtain the day-ahead bidding power and the day-ahead dispatch power of distributed generator sets and energy storage devices. This is equivalent to first treating the day-ahead bidding power of the virtual power plant to the energy market, the power of the virtual power plant's flexible load aggregator, and the dispatch power of the virtual power plant's generator sets and energy storage devices as unknowns to optimize for minimizing the total day-ahead cost of the virtual power plant. Based on the initially solved power trajectory reference value of the flexible load aggregator that minimizes the total day-ahead cost of the virtual power plant, further de-aggregation optimization is performed to obtain the power consumption curve of the flexible load. The line can be used to formulate flexible load operation plans and to calculate the actual value of the power trajectory of flexible load aggregators. Using this actual value as a known quantity, the day-ahead bidding power of the virtual power plant to the energy market and the dispatch power of the virtual power plant's generators and energy storage devices are used as unknown quantities to re-optimize to minimize the total cost of the virtual power plant on the day-ahead. The corresponding power obtained in the end (i.e., the day-ahead bidding power of the virtual power plant to the energy market and the day-ahead dispatch power of the distributed generators and energy storage devices) can be used to formulate guaranteed day-ahead bidding plans and day-ahead dispatch plans for the virtual power plant. Thus, the power information used to formulate flexible load operation plans, day-ahead bidding plans, and day-ahead dispatch plans that can guarantee the minimum operating cost of the virtual power plant can be obtained. This solves the problem of poor economic efficiency of virtual power plant trading operations caused by the failure to consider the impact of load resource pair allocation and setting on the operating economy of virtual power plants in the existing technology.

[0048] In this embodiment, the flexible load aggregator aggregates the flexible loads by combining the NJW spectrum clustering algorithm with historical data of the flexible load demand parameters. The flexible load demand parameters include the lower and upper limits of the flexible load power demand, the lower and upper limits of the flexible load energy demand, and the start and end times of the flexible load demand.

[0049] Specifically, the flexible load aggregation method based on the Ng-Jordan-Weiss (NJW) spectral clustering algorithm clusters massive flexible loads with similar characteristics into the same group, forming an aggregator to participate in virtual power plant scheduling and operation. The demand parameters of the flexible loads can be obtained based on historical data.

[0050] For each flexible load, its parameter (T) start.L,h ,T end,L,h P L,h,max ,P L,h,min E L,h,max-deE L,h,min-de The general model of flexible loads, used to reflect their demand information, can be represented as follows:

[0051] P L,h,min ≤P L,h,t ≤P L,h,max ,T start,L,h ≤t≤T end,L,h

[0052] P L,h,t =0, t<T start,L,h or t>T end,L,h

[0053] E L,h,t =E L,h,t-1 +ΔT·P L,h,t

[0054]

[0055] In the formula, P L,h,min and P L,h,max These represent the lower and upper limits of the power demand for the flexible load h, respectively; E L,h,min-de and E L,h,max-de These represent the lower and upper limits of the energy demand for the flexible load h, respectively. start,L,h and T end,L,h ΔT represents the start and end times of the demand for the flexible load h, respectively; ΔT is the scheduling time interval.

[0056] Step 1). Based on the input flexible load parameters, utilize the number of aggregators (N) AGG,i Use this to initialize the desired number of clusters.

[0057] Step 2). Normalize the extreme values ​​of each flexible load parameter. The normalization method is as follows:

[0058]

[0059] Where: N L,m x represents the number of flexible loads at node m; L,h The original demand parameters for the flexible load h; x′ L,h The normalized flexible load h requirement parameter.

[0060] Step 3). Use Euclidean distance to measure the differences between different flexible loads, and obtain the difference matrix D between flexible loads as follows:

[0061] D(x,y)=α1·||T′ start,L,x -T′ start,L,y ||2+α2·||T′ end,L,x -T′ end,L,y||2+α3·||E′ L,x,max -E′ L,x,max ||2+α4·||E′ L,x,min -E′ L,x,min ||2+α5·||P′ L,x,max -P′ L,y,max ||2+α6·||P′ L,x,min -P′ L,y,min ||2

[0062] Step 4). Use the Gaussian kernel function to transform the difference matrix D into the adjacency matrix K as follows:

[0063]

[0064] In the formula: These are the parameters of the Gaussian kernel function.

[0065] Step 5). Normalize the adjacency matrix K and calculate its top N. AGG,m The largest eigenvalues ​​and their corresponding eigenvectors are used to construct the eigenvector space matrix.

[0066] Step 6). Use the K-means algorithm to cluster the eigenvector space matrix U.

[0067] Based on the flexible load clustering results, for each flexible load aggregator, the demand parameter information of the aggregator can be obtained according to all the flexible load parameters within it, as follows:

[0068]

[0069] Where: N L,k,m Let m be the number of flexible loads inside aggregator k at node m. and These are the start and end demand times for aggregator k at node m, respectively. and These are the upper and lower limits of the power demand for aggregator k at node m, respectively; and These are the upper and lower limits of the energy demand for aggregator k at node m, respectively.

[0070] In this embodiment, the current-day virtual power plant optimization model includes:

[0071]

[0072] In the formula, T represents the total number of operating periods of the virtual power plant; The cost corresponding to the day-ahead bid power of the virtual power plant in the energy market; Let t be the day-ahead forecast electricity price in the energy market. Let t be the power output of the virtual power plant bidding in the energy market. Let i be the active power output of generator unit i at time t; Let i be the start-stop state of generator set i at time t; and These are the active power input and active power output of energy storage device j at time t, respectively. F represents the power consumption of the flexible polymerizer k at node m at time t. i DG (·) and These represent the corresponding costs of the dispatch power of virtual power plant generator unit i, virtual power plant energy storage device j, and the corresponding costs of the power of virtual power plant flexible load aggregator k, respectively; N DG N represents the number of generator sets. ESS N represents the number of energy storage devices. AGG This represents the number of flexible load aggregators.

[0073] Specifically, F i DG (·) and F i ESS (·) represents the operating costs of generator set i and energy storage device i, respectively, which can be expressed as follows:

[0074]

[0075] In the formula, a i b i and c i These are the operating cost coefficients for generator set i; AP i This represents the operating cost coefficient for energy storage equipment.

[0076] In this embodiment, the constraints corresponding to the power flow equations include:

[0077]

[0078] In the formula, N B r represents the total number of nodes in the virtual power plant. pq and x pq These represent the resistance and reactance between line p and line q, respectively; θ m V m P m and Q m These represent the phase, voltage magnitude, injected active power, and injected reactive power at node m, respectively. Let be the power consumption of aggregator k at node m at time t; and These represent the active and reactive loads at node m at time t, respectively; P pq and Q pq These represent the active power flow and reactive power flow between line p and line q, respectively. Let t be the power output of the virtual power plant bidding in the energy market. Let i be the active power output of generator unit i at time t; and These represent the active power input and active power output of energy storage device j at time t, respectively; k pq1 k is the first impedance parameter. pq2 θ is the second impedance parameter; p θ is the phase angle at node p; q V is the phase angle at node q; p V is the voltage at node p; q Let q be the voltage at node q; This refers to the reactive power provided by generator set i at time t.

[0079]

[0080] V m,min ≤V m ≤V m,max

[0081] In the formula, P pq and Q pq These represent the active power flow and reactive power flow between line p and line q, respectively; S pq,N V is the rated capacity of the transmission line between line p and line q; m,max and V m,min These are the upper and lower limits of the node voltage amplitude, respectively.

[0082] The constraints for distributed generator sets include:

[0083]

[0084] In the formula, and These are the upper and lower limits of the output of generator set i, respectively; The maximum climbing rate of generator set i; UT represents the rated capacity of generator set i; i DG and DT i DG These are the minimum power-on time and the minimum power-off time, respectively; y i,t and z i,t Integer variables introduced; Let i be the start-stop state of generator set i at time t; This refers to the reactive power provided by generator set i at time t.

[0085] The constraints for energy storage devices include:

[0086]

[0087] In the formula, and and are the maximum input / output power and minimum input / output power of energy storage device j, respectively; and These are the upper and lower limits of energy storage for energy storage device j, respectively; η in and η out These are the power input and output energy conversion coefficients, respectively; λ is the energy storage dissipation coefficient; ΔT is the energy storage operation time interval; Let j be the state of charge of the energy storage device at time t; Let J represent the state of charge of energy storage device J at the initial moment.

[0088] The constraints for the flexible load aggregator include:

[0089]

[0090] In the formula, and These are the upper and lower limits of the power demand for aggregator k at node m, respectively; Let be the power consumption of aggregator k at node m at time t; and These are the start and end demand times for aggregator k at node m, respectively. and These are the upper and lower limits of the energy demand for aggregator k at node m, respectively. Let be the reference energy state of aggregator k at node m at time t.

[0091] In this embodiment, the optimization calculation method for the power consumption curve of the flexible load used to determine the flexible load operation plan after de-aggregation optimization of the reference value includes: de-aggregation optimization of an optimization model containing the function terms corresponding to the difference between the overall output power and the reference power trajectory corresponding to the power consumption curve of the flexible load in the flexible load aggregator and the function terms corresponding to the start and stop times of each flexible load with the goal of minimizing the value, to obtain the power consumption curve of the flexible load.

[0092] The optimization model aimed at minimizing the power consumption curves of flexible loads in a flexible load aggregator includes the following methods for de-aggregating and optimizing the function terms corresponding to the difference between the overall output power and the reference power trajectory, and the function terms corresponding to the start-stop counts of each flexible load:

[0093]

[0094] In the formula, N L,k,m N represents the number of flexible loads inside aggregator k at node m; AGG,m The number of aggregators at node m; M is a set constant; F1 DIS () represents the first optimization objective used to make the overall output power of the aggregator follow the reference power trajectory of the pre-optimization stage; F2 DIS () represents the second optimization objective for reducing the number of start-stop cycles per flexibility load; P L,n,t The actual power consumption of the nth flexible load at time t; T is the time interval; Let be the reference power consumption of aggregator k at node m at time t.

[0095] Where M is a constant obtained from the actual model.

[0096] Methods for calculating the actual power of a flexible load aggregator based on the power consumption curve of the flexible load include:

[0097]

[0098] In the formula, Let be the actual power consumption of aggregator k at node m at time t; n is the nth flexibility load at node m.

[0099] In this embodiment, a real-time virtual power plant optimization model, which aims to minimize total cost and includes the corresponding costs of the virtual power plant's real-time bidding power in the energy market, the power of the virtual power plant's flexible load aggregator, and the scheduling power of the virtual power plant's generator sets and energy storage devices, is used. Under the constraints of power flow equations, distributed generator sets, energy storage devices, and flexible load aggregators, a reference value for the power trajectory of the flexible load aggregator is obtained. Based on the power consumption curve of the flexible load used to determine the flexible load operation plan after de-aggregation optimization of the reference value, the actual value of the power trajectory of the flexible load aggregator is calculated and substituted into the real-time virtual power plant optimization model. Under the constraints, the real-time bidding power used to determine the real-time bidding plan and the scheduling power used to determine the real-time scheduling decision plan are obtained.

[0100] Specifically, in real-time operation, for a specific point in time τ, flexible loads can be divided into two categories: one is loads that have not yet arrived, and the other is loads that have already arrived. For the former, the method for obtaining flexible load demand information is still based on historical data; for the latter, its demand information can be obtained directly through collection, and some demand information (including the start time T of flexible load h) is also available. start,L,h The lower and upper limits of energy demand E for flexible load h L,h,max-de / E L,h,min-deτ should be updated promptly at every moment:

[0101] T start,L,h =τ

[0102] E L,h,min-de =E L,h,min-de -E L,h,τ-1

[0103] E L,h,max-de =E L,h,max-de -E L,h,τ-1

[0104] In the formula, E L,h,τ-1 This represents the energy demand of the flexible load at time τ-1.

[0105] Based on the flexible load demand information at each time point, the clustering method described earlier can be used to obtain the parameters of the flexible resource aggregator. It is worth noting that because the demand parameters of flexible loads are constantly changing, virtual power plant cluster aggregation should be performed again at each scheduling time. Therefore, in real-time scheduling of virtual power plants, the cluster aggregation of flexible loads is a time-sharing dynamic process.

[0106] Real-time virtual power plant optimization models include:

[0107]

[0108] In the formula, T represents the total number of operating periods of the virtual power plant; The cost corresponding to the real-time bidding power of a virtual power plant in the energy market; Let t be the current energy market electricity price. The active power of the virtual power plant interacting with the energy market at time t; The reference active power of the virtual power plant and the energy market at time t; Let i be the active power output of generator unit i at time t; and These represent the active power output and active power input of energy storage device j at time t, respectively. Let be the power consumption of node m in the flexible aggregator k at time t; F is the deviation penalty factor. i DG (·) and These represent the corresponding costs of the dispatch power of virtual power plant generator unit i, virtual power plant energy storage device j, and the corresponding costs of the power of flexible load aggregator k, respectively; N DG N represents the number of generator sets. ESS N represents the number of energy storage devices. AGG This represents the number of flexible load aggregators.

[0109] Specifically, the constraints in the real-time virtual power plant optimization model are the same as those in the day-ahead virtual power plant optimization model.

[0110] Unlike the previous virtual power plant optimization model, the real-time virtual power plant optimization model includes constraints on the power bid by the virtual power plant in the energy market at time t. The active power of the virtual power plant and the energy market at time t should be respectively considered as real-time interactive data. Alternatives; furthermore, optimizing the current energy market electricity price at time t. It can be obtained directly based on the real-time market clearing results, and the future energy market electricity price... Based on real-time electricity price forecasts. In real-time dispatching, the start-up and shutdown status of generator unit i at time t. It has been confirmed.

[0111] The de-aggregation process in the real-time virtual power plant optimization model is exactly the same as that in the day-ahead virtual power plant optimization model. Through de-aggregation, the actual values ​​of the flexible load power trajectories of each node can be obtained.

[0112] Specifically, both the day-ahead virtual power plant optimization model and the real-time virtual power plant optimization model send forecast results to the electricity market to assist in its calculations. In the day-ahead phase, the day-ahead virtual power plant optimization model is used to formulate day-ahead bidding plans and day-ahead dispatch decision plans, which are then submitted to the electricity market. In the real-time phase, the real-time virtual power plant optimization model is used to formulate real-time bidding plans and real-time dispatch decision plans, which are also submitted to the electricity market. The data submission frequency in the real-time phase is higher than that in the day-ahead phase. For example... Figure 2 As shown.

[0113] By studying the integrated energy service operation model of distributed power virtual power plants, active users, and distribution networks in the context of the electricity market, this study aims to achieve complementary advantages of energy resources and improve energy utilization efficiency by effectively regulating the behavior of distributed power sources and responsive loads within virtual power plants to participate in the electricity market balancing and integrated energy ancillary service market.

[0114] Example of a virtual power plant trading auxiliary calculation system:

[0115] This invention also provides a virtual power plant transaction auxiliary calculation system, including a processor for executing a computer program to implement the steps of the virtual power plant transaction auxiliary calculation method. The specific processes, principles, and effects of the virtual power plant transaction auxiliary calculation system have been described in detail in the embodiments of the virtual power plant transaction auxiliary calculation method, and will not be repeated here.

[0116] Specific implementation methods have been given above, but the present invention is not limited to the described implementation methods. The basic idea of ​​the present invention lies in the above basic scheme. For those skilled in the art, designing various modified models, formulas, and parameters based on the teachings of the present invention does not require creative effort. Changes, modifications, substitutions, and variations made to the implementation methods without departing from the principles and spirit of the present invention still fall within the protection scope of the present invention.

Claims

1. A virtual power plant transaction auxiliary calculation method, characterized in that, The method includes: using a day-ahead virtual power plant optimization model with the goal of minimizing total cost, which includes the day-ahead bidding power of the virtual power plant to the energy market, the power of the virtual power plant flexible load aggregator, and the corresponding cost of the dispatch power of the virtual power plant generator sets and energy storage devices, under the constraints of power flow equations, distributed generator sets, energy storage devices, and flexible load aggregators, obtaining a reference value for the power trajectory of the flexible load aggregator; calculating the actual value of the power trajectory of the flexible load aggregator based on the power consumption curve of the flexible load used to determine the flexible load operation plan after de-aggregation optimization of the reference value, and then substituting it into the day-ahead virtual power plant optimization model to obtain the day-ahead bidding power used to determine the day-ahead bidding plan and the dispatch power used to determine the day-ahead dispatch decision plan under the constraints.

2. The virtual power plant transaction auxiliary calculation method according to claim 1, characterized in that, The method includes: using a real-time virtual power plant optimization model with the goal of minimizing total cost, which includes the real-time bidding power of the virtual power plant to the energy market, the power of the virtual power plant flexible load aggregator, and the corresponding cost of the dispatch power of the virtual power plant generator sets and energy storage devices, under the constraints of power flow equations, distributed generator sets, energy storage devices, and flexible load aggregators, obtaining a reference value for the power trajectory of the flexible load aggregator; calculating the actual value of the power trajectory of the flexible load aggregator based on the power consumption curve of the flexible load used to determine the flexible load operation plan after de-aggregation optimization of the reference value, and substituting it into the real-time virtual power plant optimization model, to obtain the real-time bidding power used to determine the real-time bidding plan and the dispatch power used to determine the real-time dispatch decision plan under the constraints.

3. The virtual power plant transaction auxiliary calculation method according to claim 1 or 2, characterized in that, The optimization calculation method for the power consumption curve of the flexible load used to determine the flexible load operation plan after de-aggregation optimization of the reference value includes: de-aggregation optimization of an optimization model containing the function terms corresponding to the difference between the overall output power of the power consumption curve of the flexible load in the flexible load aggregator and the reference power trajectory, and the function terms corresponding to the number of start-stop times of each flexible load, with the goal of minimizing the value, to obtain the power consumption curve of the flexible load.

4. The virtual power plant transaction auxiliary calculation method according to claim 3, characterized in that, The optimization model, which aims to minimize the difference between the overall output power corresponding to the power consumption curve of the flexible loads in the flexible load aggregator and the reference power trajectory, and the function term corresponding to the number of start-stop cycles of each flexible load, is optimized through deaggregation. The methods include: In the formula, N L,k,m N represents the number of flexible loads inside aggregator k at node m; AGG,m M is the number of aggregators at node m; M is a set constant. The first optimization objective is to make the overall output power of the aggregator follow the reference power trajectory of the pre-optimization stage; The second optimization objective is to reduce the number of start-stop cycles per flexibility load; P L,n,t The actual power consumption of the nth flexible load at time t; T is the time interval; Let be the reference power consumption of aggregator k at node m at time t.

5. The virtual power plant transaction auxiliary calculation method according to claim 4, characterized in that, Methods for calculating the actual power of a flexible load aggregator based on the power consumption curve of the flexible load include: In the formula, Let be the actual power consumption of aggregator k at node m at time t; n is the nth flexibility load at node m.

6. The virtual power plant transaction auxiliary calculation method according to claim 1, characterized in that, The aforementioned day-ahead virtual power plant optimization model includes: In the formula, T represents the total number of operating periods of the virtual power plant; The cost corresponding to the day-ahead bid power of the virtual power plant in the energy market; Let t be the day-ahead forecast electricity price in the energy market. Let t be the power output of the virtual power plant bidding in the energy market. Let i be the active power output of generator unit i at time t; Let i be the start-stop state of generator set i at time t; and These are the active power input and active power output of energy storage device j at time t, respectively. Let be the power consumption of node m in the flexible aggregator k at time t; and These represent the corresponding costs of the dispatch power of virtual power plant generator unit i, virtual power plant energy storage device j, and the corresponding costs of the power of virtual power plant flexible load aggregator k, respectively; N DG N represents the number of generator sets. ESS N represents the number of energy storage devices. AGG This represents the number of flexible load aggregators.

7. The virtual power plant transaction auxiliary calculation method according to claim 2, characterized in that, The real-time virtual power plant optimization model includes: In the formula, T represents the total number of operating periods of the virtual power plant; The cost corresponding to the real-time bidding power of a virtual power plant in the energy market; Let t be the current energy market electricity price. The active power of the virtual power plant interacting with the energy market at time t; The reference active power of the virtual power plant and the energy market at time t; Let i be the active power output of generator unit i at time t; and These represent the active power output and active power input of energy storage device j at time t, respectively. Let be the power consumption of node m in the flexible aggregator k at time t; As a deviation penalty factor, and These represent the corresponding costs of the dispatch power of virtual power plant generator unit i, virtual power plant energy storage device j, and the corresponding costs of the power of flexible load aggregator k, respectively; N DG N represents the number of generator sets. ESS N represents the number of energy storage devices. AGG This represents the number of flexible load aggregators.

8. The virtual power plant transaction auxiliary calculation method according to claim 1 or 2, characterized in that, The constraints corresponding to the power flow equations include: In the formula, N B r represents the total number of nodes in the virtual power plant. pq and x pq These represent the resistance and reactance between line p and line q, respectively; θ m V m P m and Q m These represent the phase, voltage magnitude, injected active power, and injected reactive power at node m, respectively. Let be the power consumption of aggregator k at node m at time t; and These represent the active and reactive loads at node m at time t, respectively; P pq and Q pq These represent the active power flow and reactive power flow between line p and line q, respectively. Let t be the power output of the virtual power plant bidding in the energy market. Let i be the active power output of generator unit i at time t; and These represent the active power input and active power output of energy storage device j at time t, respectively; k pq1 k is the first impedance parameter. pq2 θ is the second impedance parameter; p θ is the phase angle at node p; q V is the phase angle at node q; p V is the voltage at node p; q Let q be the voltage at node q; The reactive power provided by generator set i at time t; In m,min ≤V m ≤V m,max In the formula, S pq,N V is the rated capacity of the transmission line between line p and line q; m,max and V m,min These are the upper and lower limits of the node voltage amplitude, respectively. The constraints corresponding to the distributed generator set include: In the formula, and These are the upper and lower limits of the output of generator set i, respectively; The maximum climbing rate of generator set i; The rated capacity of generator set i; and These are the minimum power-on time and the minimum power-off time, respectively; y i,t and z i,t Integer variables introduced; Let i be the start-stop state of generator set i at time t; The reactive power provided by generator set i at time t; The constraints corresponding to the energy storage device include: In the formula, and and are the maximum input / output power and minimum input / output power of energy storage device j, respectively; and These are the upper and lower limits of energy storage for energy storage device j, respectively; η in and η out These are the power input and output energy conversion coefficients, respectively; λ is the energy storage dissipation coefficient; ΔT is the energy storage operation time interval; Let j be the state of charge of the energy storage device at time t; The initial state of charge of energy storage device j; The constraints corresponding to the flexible load aggregator include: In the formula, and These are the upper and lower limits of the power demand for aggregator k at node m, respectively; Let be the power consumption of aggregator k at node m at time t; and These are the start and end demand times for aggregator k at node m, respectively. and These are the upper and lower limits of the energy demand for aggregator k at node m, respectively. Let be the reference energy state of aggregator k at node m at time t.

9. The virtual power plant transaction auxiliary calculation method according to claim 1, characterized in that, The flexible load aggregator aggregates flexible loads using the NJW spectrum clustering algorithm combined with historical data of the demand parameters of the flexible loads. The demand parameters of the flexible loads include the lower and upper limits of the power demand, the lower and upper limits of the energy demand, and the start and end times of the demand.

10. A virtual power plant trading auxiliary computing system, comprising a processor, characterized in that, The processor is used to execute a computer program to implement the steps of the virtual power plant transaction auxiliary calculation method as described in any one of claims 1 to 9.

Images