Fair-driven power distribution system flexibility resource optimization scheduling method

Abstract

The invention discloses a fairness-driven power distribution system flexible resource optimization scheduling method which is used for comprehensively improving fairness of resource scheduling and system economy on the premise of meeting network power flow and security constraints. According to the method, a lexicographical minimum-maximum fairness model is constructed based on marginal cost and scheduling capability of a flexible resource aggregator, a cost threshold mechanism is introduced to divide scheduling cost into a plurality of intervals, and differentiated optimization objective functions are set for different cost intervals. In order to realize reasonable setting of a threshold value, an inflection point is further identified based on an economic loss change trend of a typical operation section, and an empirical cost threshold value parameter is selected. In the model building process, the scheduling problem is converted into a mixed integer linear programming model through power flow linearization and logic condition conversion, a staged sequential solving strategy is adopted, a scheduling result is optimized step by step according to the cost priority, and an optimal scheduling scheme considering fairness and economy is obtained. The method can effectively relieve the problem that a low-cost aggregator is excessively called, stimulate high-cost resources to participate in adjustment, and improve the operation efficiency and market fairness of a power distribution system.

Description

Technical Field

[0001] The present invention belongs to the fields of power market and distribution network operation, and relates to the establishment of a distribution system cost function and an optimization scheduling and clearing method for a flexibility resource market considering fairness. Specifically, it is a fairness-driven optimization scheduling method for flexibility resources in a distribution system. Background Art

[0002] With the acceleration of the global energy transition process, the penetration rate of new energy sources represented by wind power and photovoltaic power in the power system has been continuously increasing, significantly increasing the uncertainty and volatility of system operation. The traditional centralized distribution system dominated by controllable power sources is gradually transformed into an active distribution system dominated by new energy. This structural change poses higher requirements for grid flexibility. The system urgently needs to effectively coordinate and integrate various flexibility resources such as energy storage, demand-side response, and electric vehicles to achieve supply-demand balance and ensure the safe and stable operation of the system. Against this background, the power flexibility market has emerged. It stimulates different types of flexibility resources to actively participate in power balance regulation through a market mechanism, and has become an important means to improve the operation efficiency of the power system, promote the high-proportion consumption of new energy, and achieve the carbon neutrality goal.

[0003] Traditional power system scheduling is usually based on the principle of cost minimization, aiming to achieve system operation balance at the lowest economic cost. However, this scheduling method with a single economic benefit target has obvious deficiencies: on the one hand, the single scheduling target oriented by cost is likely to cause uneven interests among market players, resulting in the monopoly of the market by low-marginal-cost resources, affecting market diversity, and reducing the participation enthusiasm and market operation efficiency of other market players; on the other hand, over-deep scheduling of a certain flexibility resource will reduce the risk resistance ability of the merchant itself and at the same time significantly increase its operation, maintenance, and depreciation costs. Therefore, it is of great significance and necessity to introduce a fair scheduling mechanism. Fair scheduling can not only ensure the fair participation of various flexibility resources in market competition, avoid unbalanced resource utilization and market monopoly, and improve the overall operation efficiency of the power market; but also promote the coordinated role of various flexibility resources, improve the system's ability to respond to new energy fluctuations, and achieve the high-proportion consumption of new energy. At the same time, the fair scheduling mechanism helps to enhance the trust of market players, encourage more social capital investment, promote the reduction of the total social cost, and promote the long-term health and sustainable development of the power industry society. Summary of the Invention

[0004] The present invention provides a fairness-driven optimization scheduling method for flexibility resources in a distribution system, specifically including:

[0005] (1) The distribution network operator constructs an optimization model that takes into account the fairness among aggregators and social economy according to the basic energy consumption demand, flexibility scheduling range, and marginal cost reported by each node flexibility resource aggregator

[0006] (2) Optimize and schedule the flexibility resources of the distribution network based on the optimized model. By sequentially solving multiple scheduling objective functions in stages, gradually optimize and determine the scheduling plans and scheduling costs of each aggregator, and achieve fair optimization scheduling and market clearing of the flexibility resources in the distribution system.

[0007] The specific process of step (1) is as follows:

[0008] 1) In this application, the objective function of the Distribution System Operator (DSO) is defined as minimizing the system flexibility resource cost SCF (System Cost Function).

[0009] Considering the need to balance fairness and economy, this application proposes a threshold model, introducing the active power cost thresholds ε1 and ε2 in the evaluation of the system flexibility resource cost, with the maximum aggregator cost C in the optimization process as a reference. When the aggregator cost is in the interval [C ＜1＞ -ε1, C ＜1＞ , denoted as interval 1, the min-max principle is used to evaluate the scheduling cost. When the aggregator cost is in the interval [0, C ＜1＞ -ε1], the total cost minimization is used to evaluate the scheduling cost. Denote [C ＜1＞ -ε1 - ε2, C ＜1＞ -ε1] as interval 2. When the aggregator cost is in the interval [0, C ＜1＞ -ε1 - ε2], denoted as interval 3, a penalty is imposed on the objective function, that is ＜1＞ -ε1 - ε2] interval (denoted as interval 3), a penalty is imposed on the objective function, that is

[0010]

[0011] In the formula, n is the number of flexibility resource aggregators; C p , respectively represent the active and reactive power flexibility resource scheduling costs, and respectively represent the i-th largest active and reactive power flexibility resource scheduling costs; and respectively represent the number of flexibility resource aggregators whose costs are in interval 1 when the flexibility resource scheduling costs are C p and C q ; and respectively represent the number of flexibility resource aggregators whose costs are in interval 2 when the flexibility resource scheduling costs are C p and C q ; and respectively represent two active cost thresholds and two reactive cost thresholds.

[0012] 2) To achieve the practical operability and flexibility of the model, the above threshold parameters ε1 and ε2 do not need to be very precise, but should reasonably reflect the cost-sensitive interval in system operation. The present invention proposes an empirical threshold setting method, constructs an economic loss curve based on typical operation sections, so as to identify the optimal balance point between economy and fairness.

[0013] (a) Construct a reference optimization model to minimize the total system cost under the condition of no cost threshold limit, and solve to obtain the reference value of the total system cost

[0014] (b) Set different combinations of ε1 and ε2 thresholds respectively, reconstruct the optimization model and solve to obtain the total system cost C(ε1, ε2)

[0015] (c) Calculate the economic loss corresponding to each group of thresholds

[0016]

[0017] (d) Plot the economic losses of different combinations as a changing curve, identify the elbow point of the curve as the balance point, and use the ε1 and ε2 corresponding to this point as the final set values.

[0018] Since the present invention is mainly used to improve scheduling fairness rather than pursuing the ultimate optimal cost, the above threshold setting can accept approximate and engineering empirical errors to meet the flexible scheduling requirements under different operation scenarios.

[0019] The specific process of step (2) is as follows:

[0020] 1) This application combines the proposed threshold model with the lexicographic minimization model, and proposes a threshold model based on lexicographic minimization. To ensure the continuity of the model, first transform Equation (9) into an expression independent of t1 and t2

[0021]

[0022] In the formula (·) + = max{0, ·}.

[0023] According to Equation (11), the k-th objective function of the lexicographic minimization model can be written as

[0024]

[0025] Furthermore

[0026]

[0027] 2) Construct a linearized power flow equation considering network losses, voltage drops, and reactive power flows

[0028] The exact power flow expression for a branch is

[0029]

[0030] where i is the sending - end node number and j is the receiving - end node number. r ij and x i+1 are the active power, reactive power, resistance, and reactance of branch ij respectively. V i and δ i are the voltage phasor, corresponding magnitude, and corresponding phase angle of node i respectively. and represent the real part and the imaginary part respectively.

[0031] Assume δ i -δ j ≈0 and V i ≈0, then

[0032]

[0033] The node injection power expression

[0034]

[0035] where is the set of all nodes in the network.

[0036] Write Equation (16) in a compact matrix form as shown in Equation (17).

[0037]

[0038] where M and N are the corresponding coefficient matrices.

[0039] Number the branch b(i - 1,i) with the receiving - end i, and denote the power at the receiving - end of the branch as Note that numerically, b = i. Denote the set of downstream nodes of branch i (including node i) as Denote the set of downstream branches as Then

[0040]

[0041] where and are the active power loss and reactive power loss of branch b respectively, calculated by Equation (19):

[0042]

[0043] Wherein, r b and x b are the apparent power, resistance and reactance of branch b, respectively.

[0044] Assuming that within a certain range, the node voltage does not change with the node injection power, the loss factors (LFs) can be calculated from Equation (19), and the expression is shown in Equation (20).

[0045]

[0046] Let An expression in the form of is called the load shift factor (LSF), and an expression in the form of is called the generation shift factor (GSF). The LSF is calculated from Equation (21), and GSF = LSF * (-1).

[0047]

[0048] Given The total network loss P Loss From The first-order Taylor formula approximation

[0049]

[0050] 3) Network security constraints and their linearization

[0051] The network constraints of the proposed model include:

[0052]

[0053] Wherein, V and are the upper and lower limits of the allowable voltage operation of the system, P DG,i and Q DG,i are the active power output and reactive power output of the distributed power source at the i-th node, respectively, P L,i and Q L,i are the total community load at the i-th node, respectively, is the available capacity of the distributed power source at the i-th node, is the maximum allowable capacity of branch l, and B are the sets of network nodes and branches, respectively, and the variables in the brackets are the dual variables of the corresponding constraints.

[0054] The quadratic term constraint can be linearized by the polygon inner approximation method:

[0055]

[0056] The compact forms of the above equality constraint and inequality constraint are respectively denoted as A[P T ,Q T T ≤b and G[P T ,Q T T =h, and the feasible intervals of the active power and reactive power determined thereby are respectively denoted as and

[0057] 4) Convert the efficiency threshold criterion into a mixed-integer linear expression

[0058] Introduce auxiliary variables to linearly express SCF (k) , and then convert the threshold model based on lexicographic minimization into a mixed-integer linear model. The original problem Pb (1) can be linearized into Pb (1)-L

[0059]

[0060] where n is the number of flexibility resource aggregators; and are respectively the active and reactive marginal costs of the i-th flexibility resource aggregator; and are respectively the active and reactive clearing results of the energy market; p i and q i are respectively the actual scheduling outputs of the i-th aggregator; and respectively represent the flexibility adjustment amounts of the i-th aggregator; C i is the scheduling cost of the i-th aggregator; and respectively represent the feasible operating ranges of each node under the linearized network security constraints, equipment operation constraints, and network power flow constraints; and respectively represent the adjustable ranges of flexibility under the constraints of the equipment itself within the aggregator; ε1 and ε2 are the set cost threshold criteria.

[0061]

[0062] where H (1) , κ i ,ρ,M,Δ i ​​is the linearization auxiliary variable; n is the number of flexibility resource aggregators; and are the active and reactive marginal costs of the i-th flexibility resource aggregator respectively; and are the active and reactive clearing results of the energy market respectively; p i and q i are the actual scheduling outputs of the i-th aggregator respectively; and represent the flexibility adjustment amounts of the i-th aggregator respectively; C i is the scheduling cost of the i-th aggregator; and represent the feasible operating ranges of each node under the linearized network security constraints, equipment operation constraints, and network power flow constraints respectively; and represent the adjustable ranges of flexibility under the constraints of the equipment itself within the aggregator respectively; ε1, ε2 are the set cost threshold criteria.

[0063] When k ≥ 2, the original problem Pb (k) can be linearized into Pb (k)-L .

[0064]

[0065] In the formula, n is the number of flexibility resource aggregators; and are the active and reactive marginal costs of the i-th flexibility resource aggregator respectively; and are the active and reactive clearing results of the energy market respectively; p i and q i are the actual scheduling outputs of the i-th aggregator respectively; and represent the flexibility adjustment amounts of the i-th aggregator respectively; C i is the scheduling cost of the i-th aggregator; and represent the feasible operating ranges of each node under the linearized network security constraints, equipment operation constraints, and network power flow constraints respectively; and represent the adjustable ranges of flexibility under the constraints of the equipment itself within the aggregator respectively. i k is the number of the flexibility resource aggregator with the k-th highest cost determined after solving Pb (k) ; ε1, ε2 are the set cost threshold criteria.

[0066]

[0067] In the formula, H(k) , κ i , ρ, M, Δ 1,i , Δ 2,i are linearization auxiliary variables; is the cost of the k-th largest flexibility resource aggregator determined for the k-th optimization; I k is the set of aggregator numbers of the undetermined scheduling situations remaining before the k-th optimization; n is the number of flexibility resource aggregators; and are the active and reactive marginal costs of the i-th flexibility resource aggregator respectively; and are the active and reactive clearing results of the energy market respectively; p i and q i are the actual scheduling outputs of the i-th aggregator respectively; and represent the flexibility adjustment amounts of the i-th aggregator respectively; C i is the scheduling cost of the i-th aggregator; and represent the feasible operating ranges of each node under the linearized network security constraints, equipment operation constraints, and network power flow constraints respectively; and represent the adjustable ranges of flexibility under the self-constraints of the equipment within the aggregator respectively; ε1, ε2 are the set cost threshold criteria.

[0068] 5) According to the results obtained by the optimization, determine the scheduling amounts and scheduling costs of all flexibility resource aggregators in sequence Description of the Drawings

[0069] Appendix Figure 1 is a schematic flow chart of a fairness-driven optimal scheduling method for flexibility resources in a distribution system in the present invention;

[0070] Appendix Figure 2 is a schematic diagram of a common application scenario in the present invention; Detailed Embodiment

[0071] As shown in Appendix Figure 1 , a fairness-driven optimal scheduling method for flexibility resources in a distribution system, the method includes the following steps:

[0072] Step 1: Propose a threshold model based on lexicographical min-max fairness, that is, the distribution network operator sets the cost threshold standard based on the flexibility scheduling range and marginal cost reported by each node flexibility resource aggregator, taking the aggregator with the highest cost as the benchmark, and redesigns the social cost function (total system cost function) accordingly, constructs an optimization objective reflecting the fairness of each aggregator, so as to achieve fair scheduling among different aggregators. This model effectively balances the overall economy of the system and the fairness among aggregators, and avoids extreme situations where individual aggregators are over-scheduled or under-scheduled;

[0073] Step 2: Use a market clearing method for fair optimization scheduling of flexibility resources, specifically including: linearize the distribution network power flow equation containing node voltage variables and network loss variables to obtain linearized network constraints; based on the above threshold fairness model, construct the optimization problem as a mixed integer linear programming model; use the linear programming method to solve the above mixed integer linear model in sequence to determine the flexibility resource scheduling volume and corresponding scheduling costs of each node aggregator, obtain a fair optimization scheduling plan for the flexibility market of the distribution system, and finally form the market clearing result.

[0074] Unless otherwise specifically stated, the relative steps, numerical expressions and values of the components and steps described in these embodiments do not limit the scope of the present invention.

[0075] The flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions and operations according to the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a program segment or a part of code, and the part of the module, the program segment or the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the accompanying drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0076] Finally, it should be noted that the above-described embodiments are only specific embodiments of the present invention, used to illustrate the technical solutions of the present invention, rather than limiting it. The protection scope of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that any technician familiar with the technical field of the present invention can still modify the technical solutions recorded in the foregoing embodiments, or can easily think of changes, or perform equivalent replacements on some of the technical features; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A fairness-driven optimal scheduling method for flexible resources in a distribution system, characterized in that (1) The distribution network operator constructs an optimization model that takes into account the fairness among aggregators and social economy based on the basic energy consumption demand, flexible scheduling range, and marginal cost reported by each node flexible resource aggregator; (2) Based on the optimization model, perform optimal scheduling of flexible resources in the distribution network. By sequentially solving multiple scheduling objective functions in stages, gradually optimize and determine the scheduling plans and scheduling costs of each aggregator, so as to achieve fair optimal scheduling and market clearing of flexible resources in the distribution system.

2. The fairness-driven flexible resource optimal scheduling method for a distribution system according to claim 1, wherein, The optimization model described in step (1) is specifically a cost threshold model based on lexicographic min-max fairness, which specifically includes: The distribution network operator sets cost threshold criteria ε1 and ε2 according to the basic energy consumption demand, flexible scheduling range, and marginal cost reported by the aggregator. Based on the cost threshold criteria, redesign the total system cost function, construct an optimization objective considering fairness, and avoid excessive or insufficient scheduling of individual aggregators, so as to effectively balance the overall economy of the system and the fairness among aggregators.

3. A fairness-driven flexible resource optimal scheduling method for a distribution system according to claim 2, characterized in that, For the threshold model based on lexicographic min-max fairness, the cost threshold criteria are specifically divided into three cost intervals and corresponding cost evaluation methods are implemented: Remember C ＜1＞ is the maximum fee determined during the optimization process, when the aggregator fee is between [C ＜1＞ -ε1,C ＜1＞ ] interval, the lexicographic minimum-maximum principle is used for cost evaluation; when the aggregator cost is in [C ＜1＞ -ε1-ε2,C ＜1＞ -ε1] interval, the total cost minimization principle is used for cost evaluation; when the aggregator cost is in the range of [0,C ＜1＞ -ε1-ε2] interval, an additional penalty is imposed on the excess cost in the optimization objective function; The specific expression of the objective function SCF of the optimization problem is: Where n is the number of flexibility resource aggregators; represents the flexibility resource scheduling cost, C = [C ＜1＞ ,..., C ＜n＞ T , C ＜i＞ represents the i-th largest flexibility resource scheduling cost determined in the optimization process; t1(C) and t2(C) respectively represent the number of flexibility resource aggregators when the flexibility resource scheduling cost is C and the cost is in the intervals [C ＜1＞ -ε1, C ＜1＞ and [C ＜1＞ -ε1 - ε2, C ＜1＞ -ε1]; ε1 and ε2 respectively represent the cost thresholds of the two intervals.​ 4. A fairness-driven flexible resource optimal scheduling method for a distribution system according to claim 3, characterized in that The specific selection method for the cost thresholds ε1 and ε2 is as follows: Based on the same operation scenario, the distribution network operator uses the objective function of minimizing the total system cost to calculate the benchmark value of the total system cost The distribution network operator uses the threshold model to calculate the total system cost corresponding to different threshold combinations and the economic loss corresponding to different threshold combinations According to the curve of the above economic loss amount changing with the threshold combination, determine the elbow point of the economic loss amount curve, and select the thresholds corresponding to the elbow point as the finally adopted values of ε1 and ε2.

5. A fairness-driven flexible resource optimal scheduling method for a distribution system according to claim 1, characterized in that The optimal scheduling of flexible resources in the distribution network described in step (2) specifically includes: linearize the distribution network power flow equation containing node voltage variables, network loss variables, and reactive power variables to obtain linearized network constraint conditions; combine the optimization model reflecting fairness, construct the optimization problem as a mixed-integer linear programming model, and use the method of mixed-integer linear programming to sequentially solve the above model to determine the flexible resource scheduling amounts and corresponding scheduling costs of each aggregator, and finally achieve fair optimal scheduling and clearing of the flexible market in the distribution network.

6. The fairness-driven flexible resource optimal scheduling method for a distribution system according to claim 5, characterized in that The construction process of the mixed-integer linear programming model specifically includes the following steps (1) To ensure the continuity of the objective function, first transform Equation (1) into an expression independent of t1 and t2 where (·) + = max{0, ·}; n is the number of flexibility resource aggregators; C ＜i＞ represents the i-th largest flexibility resource scheduling cost determined during the optimization process; ε1 and ε2 are the set cost threshold criteria; (2) According to Equation (3), the k-th objective function of the lexicographic minimization model can be written as where (·) + = max{0, ·}; n is the number of flexibility resource aggregators; C ＜i＞ represents the i-th largest flexibility resource scheduling cost determined during the optimization process; ε1 and ε2 are the set cost threshold criteria; (3) Introduce auxiliary variables to linearly represent the SCF (k) and then transform it into a mixed-integer linear model based on the lexicographically minimized threshold model. The original problem Pb (1) can be linearized into Pb (1)-L Where n is the number of flexibility resource aggregators; and are the active and reactive marginal costs of the i-th flexibility resource aggregator respectively; and are the active and reactive clearing results of the energy market respectively; p i and q i are the actual scheduling outputs of the i-th aggregator respectively; and represent the flexibility adjustment amounts of the i-th aggregator respectively; C i is the scheduling cost of the i-th aggregator; and represent the feasible operating ranges of each node under linearized network security constraints, equipment operation constraints, and network power flow constraints respectively; and represent the adjustable ranges of flexibility under the constraints of the equipment itself within the aggregator respectively; ε1 and ε2 are the set cost threshold criteria. Where, H (1) , κ i , ρ, M, Δ i are linearization auxiliary variables; n is the number of flexibility resource aggregators; and are the active and reactive marginal costs of the i-th flexibility resource aggregator respectively; and are the active and reactive clearing results of the energy market respectively; p i and q i are the actual scheduling outputs of the i-th aggregator respectively; and represent the flexibility regulation amounts of the i-th aggregator respectively; C i is the scheduling cost of the i-th aggregator; and represent the feasible operating ranges of each node under the linearized network security constraints, equipment operation constraints, and network power flow constraints respectively; and represent the adjustable ranges of flexibility under the constraints of the equipment itself within the aggregator respectively; ε1, ε2 are the set cost threshold criteria; When k ≥ 2, the original problem Pb (k) can be linearly transformed into Pb (k)-L , Where n is the number of flexibility resource aggregators; and are the active and reactive marginal costs of the i-th flexibility resource aggregator respectively; and are the active and reactive clearing results of the energy market respectively; p i and q i are the actual scheduling outputs of the i-th aggregator respectively; and represent the flexibility adjustment amounts of the i-th aggregator respectively; C i is the scheduling cost of the i-th aggregator; and represent the feasible operating ranges of each node under linearized network security constraints, equipment operation constraints, and network power flow constraints respectively; and represent the adjustable ranges of flexibility under the constraints of the equipment within the aggregator respectively, i k is the number of the flexibility resource aggregator with the k-th highest cost determined after solving Pb (k) ; ε1 and ε2 are the set cost threshold criteria; where H (k) , κ i , ρ, M, Δ 1,i , Δ 2,i are linearization auxiliary variables; is the cost of the k-th largest flexibility resource aggregator determined by the k-th optimization; I k is the set of aggregator numbers of the undetermined scheduling situations remaining before the k-th optimization; n is the number of flexibility resource aggregators; and are the active and reactive marginal costs of the i-th flexibility resource aggregator respectively; and are the active and reactive clearing results of the energy market respectively; p i and q i are the actual scheduling outputs of the i-th aggregator respectively; and represent the flexibility adjustment amounts of the i-th aggregator respectively; C i is the scheduling cost of the i-th aggregator; and represent the feasible operating ranges of each node under linearized network security constraints, equipment operation constraints, and network power flow constraints respectively; and represent the adjustable ranges of flexibility under the constraints of the equipment itself within the aggregator respectively; ε1 and ε2 are the set cost threshold criteria.

7. A fairness-driven flexible resource optimal scheduling method for a distribution system according to claim 6, characterized in that The model adopts a strategy of sequential solution in stages, and the sequential solution strategy specifically includes the following steps: 1) Solve for Pb (1)-L , and obtain the scheduling volumes and scheduling costs of all aggregators; 2) Fix the scheduling cost and scheduling volume of the aggregator with the highest cost in step 1), and solve for Pb (2)-L , and obtain the scheduling volume and scheduling cost of the remaining aggregators; 3) Fix the scheduling cost and scheduling volume of the second-highest-cost aggregator in step 2), and solve for Pb (3)-L And so on, gradually optimize the remaining aggregators; 4) The above process is carried out sequentially until the scheduling results of all aggregators are determined, and finally an optimal scheduling plan for flexible resources in the distribution network that takes into account fairness and economy is obtained.

Images