An adaptive robust optimization scheduling method for main distribution coordination system

Abstract

The present invention discloses an adaptive robust optimization scheduling method for a main-distribution cooperative system, comprising: establishing an adaptive robust optimization model of the main-distribution cooperative system; obtaining a convex hull uncertainty set for photovoltaic output, combining the convex hull uncertainty set and the adaptive robust optimization model of the main-distribution cooperative system to obtain a two-stage adaptive robust optimization model of the main-distribution cooperative system; solving the two-stage adaptive robust optimization model of the main-distribution cooperative system by using a column and constraint generation algorithm to obtain an optimal scheduling strategy for the main-distribution cooperative system; and scheduling resources for the main-distribution cooperative system based on the optimal scheduling strategy. The present invention can reduce the conservatism of the robust optimization result of the main-distribution cooperative system and improve the safety and stability of the operation of the main-distribution cooperative system.

Description

Technical Field

[0001] The invention relates to an adaptive robust optimization scheduling method for a main distribution coordination system, and belongs to the field of operation and scheduling of the main distribution coordination system. Background Art

[0002] With the large-scale access of renewable energy sources such as photovoltaics and wind power, the uncertainties in the main and distribution coordinated system (transmission and distribution system) have increased, posing a huge threat to the safety and stability of the main and distribution coordinated system.

[0003] The superposition of natural factors such as global warming has led to frequent extreme weather in recent years, further amplifying the uncertainty on both sides of the source and load and the risk of power grid equipment operation, bringing difficulties to the supply guarantee of the main distribution coordination system. Taking typhoons as an example, the problems caused by extreme natural disasters such as typhoons can be attributed to the impact of forces on system components. Wind force will produce mechanical load effects on overhead wires and towers. If the force load on the component is greater than the strength of the component itself, it will cause damage to the component, causing a large-scale circuit breaker in the main distribution coordination system.

[0004] In addition, the randomness of renewable energy output is not adequately characterized. Although existing research has optimized the modeling of uncertainty factors in the main and distribution coordination system, which is mainly divided into stochastic planning, chance-constrained planning and robust optimization, there are problems with stochastic planning and chance-constrained planning, such as inaccurate estimation of the probability distribution of uncertain parameters and long model solution time. Robust optimization is relatively conservative because it does not consider any historical data information on uncertain parameters, and requires sacrificing a certain amount of economy in exchange for the safety and stability of the main and distribution coordination system operation.

[0005] Due to the diversification of industrial, commercial and residential electricity demand, the electricity consumption pattern shows obvious volatility, especially in different time periods of the day, the electricity demand varies greatly, resulting in a significant increase in the peak-to-valley difference of the main distribution coordination system. This increase in the peak-to-valley difference has brought a series of challenges to the economic and stable operation of the main distribution coordination system. First, the excessive peak-to-valley difference requires the main distribution coordination system to have sufficient peak-shaving capacity to cope with the electricity demand during peak hours, which often means that additional peak-shaving power plants or energy storage facilities are required, increasing the system operation cost. Secondly, in order to meet short-term high-load demand, the main distribution coordination system may need to enable some high-cost and low-efficiency power generation devices, which to a certain extent affects the economy and environmental friendliness of power production. In addition, the large peak-to-valley difference may also aggravate the wear of power grid equipment, reduce the service life of equipment, and increase the frequency and cost of maintenance and replacement. Therefore, introducing peak shaving and valley filling and node voltage offset constraints, and reducing the impact of peak-to-valley differences through technological innovation and demand-side management are of great significance to improving the economic benefits of the main distribution coordination system and ensuring the safety of the main distribution coordination system. Summary of the invention

[0006] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide an adaptive robust optimization scheduling method for a main distribution coordinated system, which can reduce the conservatism of the robust optimization results and improve the safety and stability of the operation of the main distribution coordinated system.

[0007] To achieve the above object, the present invention is implemented by adopting the following technical solutions:

[0008] The present invention provides an adaptive robust optimization scheduling method for a main distribution coordination system, comprising:

[0009] Establish an adaptive robust optimization model for the main and distribution coordination system;

[0010] Acquire a convex hull uncertainty set for photovoltaic output, and combine the convex hull uncertainty set with the main-distribution coordinated system adaptive robust optimization model to obtain a two-stage adaptive robust optimization model for the main-distribution coordinated system;

[0011] The column and constraint generation algorithm is used to solve the two-stage adaptive robust optimization model of the main and distribution coordination system, and the optimal scheduling strategy of the main and distribution coordination system is obtained;

[0012] Based on the optimal scheduling strategy, resources are scheduled for the main distribution coordination system.

[0013] Optionally, the adaptive robust optimization model of the main and distribution coordinated system includes an objective function and constraints; the objective function is to minimize the operating cost of the main and distribution coordinated system during the total scheduling period; the constraints include: distribution network flow constraints, transmission network flow constraints, grouped capacitor group switching constraints, energy storage charging and discharging constraints, system safety constraints, gateway power purchase constraints, flexible load demand response constraints, network reconstruction constraints, peak shaving and valley filling, and voltage offset constraints.

[0014] Optionally, the objective function is expressed as follows:

[0015] ;

[0016] in: Schedule variables for the first stage; and They are the active power and reactive power of electricity purchased at the gateway respectively; It is the state variable indicating whether the number of capacitor banks switched on and off during period t has changed; and They are the energy storage charging power and discharging power during period t respectively; is the dispatching cost of the first stage in the total dispatching period, It is the second stage dispatching fee within the total dispatching period.

[0017] Optionally, the dispatching costs of the first stage include: gateway power purchase costs, capacitor bank switching costs, and energy storage charging and discharging costs, and the formulas are as follows:

[0018] ;

[0019] ;

[0020] ;

[0021] in: Cost of purchasing electricity for the gateway; Cost of switching on and off capacitor banks; is the energy storage charging and discharging fee; i is the node number of the distribution network, and t is the time period number; is the electricity purchase price at the gate of period t; T is the total number of periods; and They are the charging cost and discharging cost per unit of electricity of the energy storage device respectively; is the number of energy storage installations in the distribution network; Cost of changing the switching status of the capacitor bank unit; is the number of capacitor banks in the distribution network; is the time variation, that is t The amount of change.

[0022] Optionally, the second stage dispatching fee includes: load shedding penalty fee, the formula is as follows:

[0023] ;

[0024] Where: N is the number of nodes in the distribution network; is the load shedding penalty coefficient during period t; is the load shedding power during period t.

[0025] Optionally, the power flow constraint of the distribution network is expressed as follows:

[0026] ;

[0027] ;

[0028] ;

[0029] ;

[0030] in: , are the active power and reactive power injected into node i during period t, respectively; ; are the active power and reactive power flowing through branch ij in period t respectively; , are the active power and reactive power flowing through branch ji in period t respectively; is the set of end nodes of the branch with node i as the first endpoint; is the set of first-segment nodes of the branch with node i as the terminal node; is the current amplitude of branch ij during period t; is the current amplitude of branch ji during period t; , are the resistance and reactance of branch ij respectively, , are the resistance and reactance of branch ji respectively; is the voltage amplitude of node i during period t, is the voltage amplitude of node j during period t;

[0031] The power flow constraints of the transmission network are expressed as follows:

[0032] ;

[0033] ;

[0034] in: , are the active power and reactive power flowing through the transmission network branch IJ during period t, respectively; , are the active power and reactive power flowing through the transmission network branch JI in period t respectively; is the set of end nodes of the branch of the transmission network with node I as the first end point; is the first section node set of the branch of the transmission network with node I as the terminal node; and are respectively the active output and reactive output of the generator at node I in the transmission network during period t; and are the active and reactive loads of node I in period t respectively; is the voltage amplitude of the transmission network node I during period t, is the voltage amplitude of the transmission network node J at period t; , are the resistance and reactance of the transmission network branch IJ respectively; is the reference node voltage amplitude, is an integer variable, i.e., the line state variable of the transmission network branch IJ, which is 0 for open and 1 for closed, and K is a constant;

[0035] The switching constraints of grouped capacitor banks are expressed as follows:

[0036] ;

[0037] ;

[0038] ;

[0039] ;

[0040] ;

[0041] ;

[0042] in: is the number of switched capacitor groups at node i during period t; is the maximum switching quantity of the capacitor bank at node i; , They are state variables of whether the number of capacitor groups switched in period t increases or decreases; is the daily maximum switching times of the capacitor bank at node i; The rated injection power per unit capacitor bank; is the voltage amplitude of the capacitor bank at node i during period t, Inject reactive power into the capacitor bank at node i during period t;

[0043] The energy storage charging and discharging constraints are expressed as follows:

[0044] ;

[0045] ;

[0046] ;

[0047] ;

[0048] ;

[0049] ;

[0050] ;

[0051] in: , are the upper limit of energy storage charging power and the upper limit of discharging power of node i in period t respectively; , are the energy storage charging state variable and discharging state of node i in period t respectively; The upper limit of the number of daily charging and discharging times of energy storage; represents the energy storage capacity of node i during period t; Indicates the first time period node i of energy storage capacity; express T Time period node i of energy storage capacity; , are the energy storage charging efficiency and discharging efficiency respectively; , are the upper and lower limits of the energy storage capacity of node i respectively;

[0052] The system safety constraints are expressed as follows:

[0053] ;

[0054] ;

[0055] ;

[0056] ;

[0057] in: , are the minimum and maximum values ​​of the voltage at the distribution network nodes, respectively; , are the minimum and maximum values ​​of the current flowing through the distribution network branch ij respectively; , are the minimum and maximum values ​​of the voltage at the transmission network nodes, respectively; , are the minimum and maximum values ​​of the current flowing through the transmission network branch IJ respectively; for t Transmission network branch IJ The current;

[0058] The power constraint of the power purchased at the gateway is expressed as follows:

[0059] ;

[0060] ;

[0061] in: , are the minimum and maximum values ​​of the active power of the electricity purchased at the gateway; , are the minimum and maximum values ​​of reactive power of electricity purchased at the gateway, respectively;

[0062] The flexible load demand response constraint is expressed as follows:

[0063] ;

[0064] ;

[0065] ;

[0066] ;

[0067] ;

[0068] ;

[0069] ;

[0070] ;

[0071] in: and are the increase and decrease of the transferable load of node i in period t, respectively, which are controlled by the user; and are the response states of the increase and decrease of the transferable load of node i in period t, respectively, which are integer variables; is the fixed load power of node i during period t; is the ratio of the transferable load that can respond to the total load; is the charging power of the electric vehicle at node i during period t; is the maximum charging power; and The starting charging time and the ending charging time are respectively; For charging efficiency; The power of the electric vehicle at node i during period t; and They are the minimum and maximum capacity of electric vehicles respectively;

[0072] The network reconstruction constraints are expressed as follows:

[0073] ;

[0074] ;

[0075] Where: S is an integer; is the on-off state of branch ij, that is For switch closure, Disconnect the branch circuit;

[0076] The peak shaving and valley filling and voltage deviation constraints are expressed as follows:

[0077] ;

[0078] ;

[0079] in: , are the power fluctuation coefficient and the voltage deviation coefficient respectively; is the rated voltage amplitude of node i, is the number of nodes; , They are the average power fluctuation and the average voltage offset respectively.

[0080] Optionally, obtaining a convex hull uncertainty set for photovoltaic output includes:

[0081] Based on the historical data of photovoltaic output, the uncertainty of photovoltaic output is modeled using the convex hull, and the convex hull uncertainty set for photovoltaic output is obtained, which is expressed as follows:

[0082] ;

[0083] ;

[0084] ;

[0085] ;

[0086] Where: CU is the convex hull uncertainty set for photovoltaic output; z is the convex hull pole; is the set of convex hull extreme points; is the convex combination coefficient, is the vector composed of photovoltaic output in each period; is the convex hull point z The corresponding photovoltaic output.

[0087] Optionally, the two-stage adaptive robust optimization model of the main distribution coordination system is expressed as follows:

[0088] ;

[0089] ;

[0090] ;

[0091] ;

[0092] ;

[0093] ;

[0094] in: is the coefficient matrix of the first-stage dispatching variables, including the cost of purchasing electricity at the gateway, the cost of charging and discharging energy storage, and the cost of switching on and off capacitor banks. is the coefficient matrix of the second-stage dispatching variables, including the load shedding penalty fee; A and a are the coefficient matrix and right-hand side vector of the inequality constraints of the gateway power purchase, capacitor bank switching, and energy storage charging and discharging rewritten in matrix-vector form; E, H, G, M, and b are the coefficient matrix and right-hand side vector of the first-stage dispatching variables, the second-stage dispatching variables, the photovoltaic output, and the state variables of the main-distribution coordinated system after all the first- and second-stage coupled inequality constraints are rewritten in matrix-vector form; μ is the state variable of the main-distribution coordinated system; , , are the minimum values ​​of x, y, and μ respectively; , , are the maximum values ​​of x, y, and μ respectively.

[0095] Optionally, the column and constraint generation algorithm is used to solve the two-stage adaptive robust optimization model of the main distribution coordination system to obtain the optimal scheduling strategy of the main distribution coordination system, including:

[0096] The two-stage adaptive robust optimization model of the main distribution coordination system is decomposed into a main problem and sub-problems, and the sub-problems with double-level optimization problems are transformed into single-level optimization sub-problems.

[0097] The main problem is solved to obtain the optimal dispatching variables of the first stage under the constraints of limited severe scenarios, and the lower bound of the two-stage adaptive robust optimization model of the main and distribution coordination system is updated;

[0098] Based on the worst scenario corresponding to the given first-stage optimal dispatching variables, the single-layer optimization subproblem is solved to obtain the second-stage optimal dispatching variables corresponding to the scenario, and the upper bound of the two-stage adaptive robust optimization model of the main and distribution coordination system is updated;

[0099] The upper bound of the two-stage adaptive robust optimization model of the main and distribution coordination system is provided to the main problem for iteration;

[0100] Based on the increase in the number of iterations, it converges to the optimal solution of the two-stage adaptive robust optimization model of the main and distribution cooperative system, that is, the optimal scheduling strategy of the main and distribution cooperative system.

[0101] Optionally, the two-stage adaptive robust optimization model of the main and distribution coordination system is decomposed into a main problem and sub-problems, and the sub-problems of the double-layer optimization problem are converted into single-layer optimization sub-problems, including:

[0102] The main question is as follows:

[0103] ;

[0104] ;

[0105] ;

[0106] ;

[0107] ;

[0108] ;

[0109] ;

[0110] in: is an auxiliary variable; Schedule variables for the second phase of the kth iteration; is the photovoltaic output of the kth iteration, is the state variable of the master-distributor coordination system in the kth iteration; m is the total number of iterations;

[0111] The sub-questions are as follows:

[0112] ;

[0113] ;

[0114] ;

[0115] ;

[0116] in: is the optimal scheduling variable in the first stage;

[0117] Based on the second-order cone duality rule and dual transformation, the sub-problems of the two-level optimization problem are transformed into single-level optimization sub-problems, including:

[0118] The sub-problem is equivalently transformed into:

[0119] ;

[0120] ;

[0121] ;

[0122] ;

[0123] in: , , All are linear constraint dual variables; , , , , , U is the unit diagonal matrix.

[0124] The above equivalent converted subproblem is transformed into a single-layer optimization subproblem, which is expressed as follows:

[0125] ;

[0126] ;

[0127] ;

[0128] in: is the transpose of *.

[0129] Compared with the prior art, the present invention has the following beneficial effects:

[0130] The adaptive robust optimization scheduling method for the main-distribution coordinated system proposed in the present invention considers the convex hull uncertainty set of photovoltaic output, constructs a two-stage adaptive robust optimization model of the main-distribution coordinated system, and solves it to obtain the optimal scheduling strategy, thereby further reducing the conservatism of the robust optimization result (optimal scheduling strategy) and improving the safety and stability of the operation of the main-distribution coordinated system.

[0131] In order to improve the peak-to-valley difference of the system and compensate for the degree of voltage deviation, the present invention adds peak shaving and valley filling and voltage deviation constraints, reasonably configures energy storage to improve the peak-shaving capacity of the main and distribution cooperative system, smoothes the power demand curve, reduces power consumption during peak hours, and increases power consumption during valley hours, thereby reducing the maximum load of the main and distribution cooperative system, alleviating the burden of the main and distribution cooperative system during peak hours, preventing overload operation, reducing the risk of accidents, improving the stability and reliability of the main and distribution cooperative system, maintaining the voltage fluctuation of the main and distribution cooperative system within a reasonable range, ensuring the efficiency of power transmission, reducing the power loss caused by voltage abnormality, and improving the energy utilization efficiency of the main and distribution cooperative system.

[0132] The present invention effectively reduces the risk of line overload caused by extreme weather through optimizing operation measures such as power flow adjustment (adjusting the switching of capacitor banks and configuring energy storage), network reconstruction (adjusting the line start), and demand response (flexible load regulation), ensures the safe operation of the main and distribution coordinated system, improves the redundancy level of the main and distribution coordinated system, enhances the fault tolerance capability to extreme weather events, and reduces the risk of large-scale power outages. BRIEF DESCRIPTION OF THE DRAWINGS

[0133] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative labor, among which:

[0134] Figure 1 The flowchart of the adaptive robust optimization scheduling method of the main and distribution coordination system in the embodiment of the present invention is shown;

[0135] Figure 2 The schematic diagram of the box-type uncertain set and the convex hull uncertain set in the embodiment of the present invention is shown;

[0136] Figure 3 The topology diagram of the improved IEEE 30&3x33 node power transmission and distribution system in the embodiment of the present invention is shown;

[0137] Figure 4 A line graph showing the change in charging and discharging power of the energy storage device in an embodiment of the present invention is shown;

[0138] Figure 5 A bar graph showing changes in charge and discharge capacity of an energy storage device in an embodiment of the present invention;

[0139] Figure 6 The transferable load variation curve diagram in the embodiment of the present invention is shown;

[0140] Figure 7 Shown is a topological diagram of a power transmission and distribution system after network reconstruction in an embodiment of the present invention. DETAILED DESCRIPTION

[0141] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0142] The application principle of the present invention is described in detail below in conjunction with the accompanying drawings.

[0143] Example 1

[0144] like Figure 1 As shown, an embodiment of the present invention provides an adaptive robust optimization scheduling method for a main distribution coordination system, comprising the following steps:

[0145] S01: Establish an adaptive robust optimization model for the main distribution coordination system;

[0146] S02: obtaining a convex hull uncertainty set for photovoltaic output, combining the convex hull uncertainty set with an adaptive robust optimization model of a main-distribution coordination system, and obtaining a two-stage adaptive robust optimization model of a main-distribution coordination system;

[0147] S03: Solve the two-stage adaptive robust optimization model of the main and distribution coordination system using the column and constraint generation algorithm to obtain the optimal scheduling strategy of the main and distribution coordination system;

[0148] S04: Based on the optimal scheduling strategy, resources are scheduled for the main distribution coordination system.

[0149] The adaptive robust optimization scheduling method for the main-distribution coordinated system proposed in the present invention considers the convex hull uncertainty set of photovoltaic output, constructs a two-stage adaptive robust optimization model of the main-distribution coordinated system, and solves it to obtain the optimal scheduling strategy, thereby further reducing the conservatism of the robust optimization result (optimal scheduling strategy) and improving the safety and stability of the operation of the main-distribution coordinated system.

[0150] In this embodiment, in step S01, an adaptive robust optimization model of the main distribution coordination system is established, and the specific steps are as follows:

[0151] S011: Constructing the objective function

[0152] The objective function is to minimize the operating cost of the main and distribution coordination system during the total dispatch period, which is expressed as follows:

[0153] ;

[0154] in: Schedule variables for the first stage; and They are the active power and reactive power of electricity purchased at the gateway respectively; It is the state variable indicating whether the number of capacitor banks switched on and off during period t has changed; and are the energy storage charging power and discharging power respectively; is the dispatching cost of the first stage in the total dispatching period, It is the second stage dispatching fee within the total dispatching period;

[0155] Specifically, the dispatching costs in the first phase include: power purchase costs at the gateway, capacitor bank switching costs, and energy storage charging and discharging costs. The formulas are as follows:

[0156] ;

[0157] ;

[0158] ;

[0159] in: Cost of purchasing electricity for the gateway; Cost of switching on and off capacitor banks; is the energy storage charging and discharging fee; i is the node number of the distribution network, and t is the time period number; is the electricity purchase price at the gate of period t; T is the total number of periods; and They are the charging cost and discharging cost per unit of electricity of the energy storage device respectively; is the number of energy storage installations in the distribution network; Cost of changing the switching status of the capacitor bank unit; is the number of capacitor banks in the distribution network; is the time variation, that is t The amount of change.

[0160] Specifically, the second stage dispatching fee includes: load shedding penalty fee, the formula is as follows:

[0161] ;

[0162] Where: N is the number of nodes in the distribution network; is the load shedding penalty coefficient during period t; is the load shedding power during period t.

[0163] S012: Construction Constraints

[0164] The constraints include: distribution network flow constraints, transmission network flow constraints, grouped capacitor bank switching constraints, energy storage charging and discharging constraints, system safety constraints, gateway power purchase constraints, flexible load demand response constraints, network reconstruction constraints, peak shaving and valley filling, and voltage offset constraints;

[0165] ;

[0166] ;

[0167] ;

[0168] ;

[0169] in: , are the active power and reactive power injected into node i during period t, respectively; , are the active power and reactive power flowing through branch ij in period t respectively; , are the active power and reactive power flowing through branch ji in period t respectively; is the set of end nodes of the branch with node i as the first endpoint; is the set of first-segment nodes of the branch with node i as the terminal node; is the current amplitude of branch ij during period t; is the current amplitude of branch ji during period t; , are the resistance and reactance of branch ij respectively, , are the resistance and reactance of branch ji respectively; is the voltage amplitude of node i during period t, is the voltage amplitude of node j during period t;

[0170] The power flow constraints of the transmission network are expressed as follows:

[0171] ;

[0172] ;

[0173] in: , are the active power and reactive power flowing through the transmission network branch IJ during period t, respectively; , are the active power and reactive power flowing through the transmission network branch JI in period t respectively; is the set of end nodes of the branch of the transmission network with node I as the first end point; is the first section node set of the branch of the transmission network with node I as the terminal node; and are respectively the active output and reactive output of the generator at node I in the transmission network during period t; and are the active and reactive loads of node I in period t respectively; is the voltage amplitude of the transmission network node I during period t, is the voltage amplitude of the transmission network node J at period t; , are the resistance and reactance of the transmission network branch IJ respectively; is the reference node voltage amplitude, is an integer variable and K is a constant.

[0174] The switching constraints of grouped capacitor banks are expressed as follows:

[0175] ;

[0176] ;

[0177] ;

[0178] ;

[0179] ;

[0180] ;

[0181] in: is the number of switched capacitor groups at node i during period t; is the maximum switching quantity of the capacitor bank at node i; , They are state variables of whether the number of capacitor groups switched in period t increases or decreases; is the daily maximum switching times of the capacitor bank at node i; The rated injection power per unit capacitor bank; is the voltage amplitude of the capacitor bank at node i during period t, Inject reactive power into the capacitor bank at node i during period t;

[0182] The energy storage charging and discharging constraints are expressed as follows:

[0183] ;

[0184] ;

[0185] ;

[0186] ;

[0187] ;

[0188] ;

[0189] ;

[0190] in: , are the upper limit of energy storage charging power and the upper limit of discharging power of node i in period t respectively; , are the energy storage charging state variable and discharging state of node i in period t respectively; The upper limit of the number of daily charging and discharging times of energy storage; represents the energy storage capacity of node i during period t; Indicates the first time period node i of energy storage capacity; express T Time period node i of energy storage capacity; , are the energy storage charging efficiency and discharging efficiency respectively; , are the upper and lower limits of the energy storage capacity of node i respectively;

[0191] The system safety constraints are expressed as follows:

[0192] ;

[0193] ;

[0194] ;

[0195] ;

[0196] in: , are the minimum and maximum values ​​of the voltage at the distribution network nodes, respectively; , are the minimum and maximum values ​​of the current flowing through the distribution network branch ij respectively; , are the minimum and maximum values ​​of the voltage at the transmission network nodes, respectively; , are the minimum and maximum values ​​of the current flowing through the transmission network branch IJ respectively; for t Transmission network branch IJ The current;

[0197] The power constraint of the power purchased at the gateway is expressed as follows:

[0198] ;

[0199] ;

[0200] in: , are the minimum and maximum values ​​of the active power of the electricity purchased at the gateway; , are the minimum and maximum values ​​of reactive power of electricity purchased at the gateway, respectively;

[0201] The flexible load demand response constraint is expressed as follows:

[0202] ;

[0203] ;

[0204] ;

[0205] ;

[0206] ;

[0207] ;

[0208] ;

[0209] ;

[0210] in: and are the increase and decrease of the transferable load of node i in period t, respectively, which are controlled by the user; and are the response states of the increase and decrease of the transferable load of node i in period t, respectively, which are integer variables; is the fixed load power of node i during period t; is the ratio of the transferable load that can respond to the total load; is the charging power of the electric vehicle at node i during period t; is the maximum charging power; and The starting charging time and the ending charging time are respectively; For charging efficiency; The power of the electric vehicle at node i during period t; and They are the minimum and maximum capacity of electric vehicles respectively;

[0211] The network reconstruction constraints are expressed as follows:

[0212] ;

[0213] ;

[0214] Where: S is an integer; is the on-off state of branch ij, that is For switch closure, Disconnect the branch circuit;

[0215] The peak shaving and valley filling and voltage deviation constraints are expressed as follows:

[0216] ;

[0217] ;

[0218] in: , They are power fluctuation coefficient and voltage deviation coefficient respectively. is the rated voltage amplitude of node i, N is the number of nodes; , They are the average power fluctuation and the average voltage offset respectively.

[0219] In this embodiment, in step S02, a convex hull uncertainty set for photovoltaic output is obtained, and a two-stage adaptive robust optimization model of the main and distribution coordinated system is obtained by combining the convex hull uncertainty set and the main and distribution coordinated system adaptive robust optimization model, which specifically includes:

[0220] S021: Obtain the convex hull uncertainty set for photovoltaic output;

[0221] Based on the historical data of photovoltaic output, the uncertainty of photovoltaic output is modeled using the convex hull, and the convex hull uncertainty set for photovoltaic output is obtained, which is expressed as follows:

[0222] ;

[0223] ;

[0224] ;

[0225] ;

[0226] Where: CU is the convex hull uncertainty set for photovoltaic output; z is the convex hull pole; is the set of convex hull extreme points; is the convex combination coefficient, is the vector composed of photovoltaic output in each period; is the convex hull point z The corresponding photovoltaic output.

[0227] Specifically, the box-type uncertain set of the two-dimensional plane and the uncertain set based on the convex hull are as follows: Figure 2 As shown in the figure, it can be seen that the box-type uncertain set contains redundant parts, such as Figure 2 In the upper right and lower left corners, the model is more conservative; the convex hull uncertainty set is a subset of the box uncertainty set, and the model conservatism is reduced.

[0228] S022: Obtain a two-stage adaptive robust optimization model of the main distribution coordination system;

[0229] The two-stage adaptive robust optimization model of the main distribution coordination system is expressed as follows:

[0230] ;

[0231] ;

[0232] ;

[0233] ;

[0234] ;

[0235] ;

[0236] in: is the coefficient matrix of the first-stage dispatching variables, including the cost of purchasing electricity at the gateway, the cost of charging and discharging energy storage, and the cost of switching on and off capacitor banks. is the coefficient matrix of the second-stage dispatching variables, including the load shedding penalty fee; A and a are the coefficient matrix and right-hand side vector of the inequality constraints of the first-stage dispatching variables (gateway power purchase, capacitor bank switching, and energy storage charging and discharging constraints) rewritten in matrix-vector form; E, H, G, M, and b are the coefficient matrix and right-hand side vector of all the first- and second-stage coupled inequality constraints (inequality constraints of the first-stage dispatching variables and distribution network flow constraints, transmission network flow constraints, system security constraints, flexible load demand response constraints, network reconstruction constraints, peak shaving and valley filling, and voltage offset constraints) rewritten in matrix-vector form, the first-stage dispatching variables, photovoltaic output, and the state variables of the main distribution coordination system; μ is the state variable of the main distribution coordination system; , , are the minimum values ​​of x, y, and μ respectively; , , are the maximum values ​​of x, y, and μ respectively.

[0237] In this embodiment, in step S03, the column and constraint generation algorithm is used to solve the two-stage adaptive robust optimization model of the main distribution coordination system to obtain the optimal scheduling strategy of the main distribution coordination system, which specifically includes:

[0238] S031: Decompose the two-stage adaptive robust optimization model of the main distribution coordination system into a main problem and sub-problems;

[0239] The main question is as follows:

[0240] ;

[0241] ;

[0242] ;

[0243] ;

[0244] ;

[0245] ;

[0246] ;

[0247] in: is an auxiliary variable; Schedule variables for the second phase of the kth iteration; is the photovoltaic output of the kth iteration, is the state variable of the master-distributor coordination system in the kth iteration; m is the total number of iterations;

[0248] The sub-questions are as follows:

[0249] ;

[0250] ;

[0251] ;

[0252] ;

[0253] in: is the optimal scheduling variable in the first stage.

[0254] S032: Convert the sub-problems of the two-level optimization problem into single-level optimization sub-problems;

[0255] The above sub-problem is a two-level planning problem, and the inner and outer optimization variables are coupled and cannot be solved directly.

[0256] The present invention solves the problem through the second-order cone duality rule and CCG algorithm. The duality transformation realizes the decoupling of the double-layer problem by performing duality changes on the inner-layer optimization problem, and transforms the original double-layer problem into a single-layer problem for solution.

[0257] Specifically, we first transform the sub-problem model into the following equivalent transformation: , , , , , U is the unit diagonal matrix.

[0258] The above sub-problems are equivalent to:

[0259] ;

[0260] ;

[0261] ;

[0262] ;

[0263] in: , , Linearly constrain the dual variables separately.

[0264] The above equivalent sub-problem is transformed into a single-layer optimization sub-problem:

[0265] ;

[0266] ;

[0267] ;

[0268] S033: Next, we solve the main and sub-problems (main problem and single-layer optimization sub-problem) respectively.

[0269] The main problem is solved to obtain the optimal dispatching variables of the first stage under the constraints of limited severe scenarios, and the lower bound of the two-stage adaptive robust optimization model of the main and distribution coordination system is updated;

[0270] Based on the worst scenario corresponding to the given first-stage optimal dispatching variables, the subproblems are solved to obtain the second-stage optimal dispatching variables corresponding to the scenario, and the upper bound of the two-stage adaptive robust optimization model of the main and distribution coordination system is updated;

[0271] The upper bound of the two-stage adaptive robust optimization model of the main and distribution coordination system is provided to the main problem for iteration;

[0272] Based on the increase in the number of iterations, it converges to the optimal solution of the two-stage adaptive robust optimization model of the main and distribution cooperative system, that is, the optimal scheduling strategy of the main and distribution cooperative system.

[0273] Example 2

[0274] Case Analysis

[0275] The present invention adopts Figure 3 The example simulation is carried out based on the improved IEEE 30&3x33 node transmission and distribution system (main-distribution coordinated system) shown in the figure. The main grid (transmission network) area is the IEEE 30-node transmission system, in which nodes 1, 2, 22, 27, 23, and 13 are connected to generators respectively; the distribution network (distribution network) area is the three IEEE 33-node distribution systems DS1-DS3, which are respectively connected to nodes 24, 26, and 29 of the main grid system; the load of distribution network DS1 is 8.28MW+j6.85MVar, the load of DS2 is 3.18MW+j2.43MVar, and the load of DS3 is 1.93MW+j1.16Mvar. The voltage upper limit of the main grid is 1.1 pu, and the voltage lower limit is 0.9 pu; the voltage square term upper limit of the distribution network is 1.21 pu, the voltage square term lower limit is 0.81 pu, and the reference node voltage amplitude is fixed at 0.99pu. Nodes 47, 62, 77 and 83 are connected to photovoltaic systems and energy storage devices with rated capacities of 800kW, 1000kW, 500kW and 300kW respectively.

[0276] like Figure 4As shown in FIG. 1 , a line graph of the charging and discharging power variation of the energy storage device is shown. In the graph, Pess1-Pess4 represent the charging and discharging power of the energy storage devices with rated capacities of 800kW, 1000kW, 500kW and 300kW connected to nodes 47, 62, 77 and 83, respectively. In the graph, a line greater than 0 indicates charging, and a line less than 0 indicates discharging. The corresponding capacity variation is shown in FIG. Figure 5 As shown in the figure, Eess1-Eess4 represent the storage capacity of the above four energy storages respectively; in summary, the charging and discharging of the energy storage device roughly presents a process of "discharging, charging, discharging, charging, and discharging". The initial power of the energy storage device is large. As the load increases, the energy storage is discharged and the power decreases. At noon, due to the sufficient photovoltaic output, the power of the energy storage device is replenished, but because the load power at noon is also large, the energy storage device needs to continue to discharge to maintain the voltage stability of the transmission and distribution system. Overall, due to the large capacity of the energy storage devices connected to nodes 47 and 62, the power changes of charging and discharging are large, and the load borne is large.

[0277] like Figure 6 The transferable load change curve is shown in the figure. In the figure, the load increment is greater than 0, and the load reduction is less than 0. As can be seen from the figure, the load increment is mainly concentrated in time periods 2-4 and 10-13. This is because the energy storage power in time periods 2-4 is sufficient, and the photovoltaic output in time periods 10-13 is large. The transferable load is transferred to these time periods as much as possible, and the system can provide sufficient output to meet the load and reduce costs.

[0278] like Figure 7 The figure shows the topology of the power transmission and distribution system after network reconstruction. Due to the inclusion of network reconstruction, there are no more power outage islands in the system. This is because the investment of the tie line provides an additional power supply path for the load, greatly reducing the risk of large-scale power outages. The investment of the tie line forms a large microgrid with multiple distributed resources for coordinated power supply, which reduces the system's requirements for backup power capacity and significantly reduces the risk of large-scale power outages in the system under extreme weather conditions, thereby significantly reducing investment pressure and operating costs.

[0279] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. An adaptive robust optimization scheduling method for a main distribution coordination system, characterized in that: include: Establish an adaptive robust optimization model for the main and distribution coordination system; Acquire a convex hull uncertainty set for photovoltaic output, and combine the convex hull uncertainty set with the main-distribution coordinated system adaptive robust optimization model to obtain a two-stage adaptive robust optimization model for the main-distribution coordinated system; The column and constraint generation algorithm is used to solve the two-stage adaptive robust optimization model of the main and distribution coordination system, and the optimal scheduling strategy of the main and distribution coordination system is obtained; Based on the optimal scheduling strategy, resources are scheduled for the main distribution coordination system; The main distribution coordination system adaptive robust optimization model includes an objective function and constraint conditions; The objective function is to minimize the operating cost of the main and distribution coordination system in the total dispatch period; The constraints include: distribution network flow constraints, transmission network flow constraints, grouped capacitor bank switching constraints, energy storage charging and discharging constraints, system safety constraints, gateway power purchase constraints, flexible load demand response constraints, network reconstruction constraints, peak shaving and valley filling, and voltage offset constraints; The objective function is expressed as follows: min{f1(x)+max(minf2)}; Where: x = {P Mt ,Q Mt ,u CB,t ,P ch,t ,P dis,t } is the first stage scheduling variable; P Mt and Q Mt are the active power and reactive power of electricity purchased at the gateway respectively; u CB,t P is the state variable indicating whether the number of capacitor banks switched on and off during period t has changed; ch,t and P dis,t are the energy storage charging power and discharging power in period t respectively; f1(x) is the dispatching cost of the first stage in the total dispatching period, and f2 is the dispatching cost of the second stage in the total dispatching period; The dispatching costs in the first stage include: power purchase costs at the gateway, capacitor bank switching costs, and energy storage charging and discharging costs. The formulas are as follows: Where: f 11 The cost of purchasing electricity at the checkpoint; 12 is the switching cost of the capacitor bank; f 13 is the energy storage charging and discharging fee; i is the node number of the distribution network, t is the time period number; C M,t is the electricity purchase price at the threshold of period t; T is the total number of periods; C ch and C dis N are the charging cost and discharging cost per unit of energy storage device respectively; Es is the number of energy storage installations in the distribution network; C CB The cost of changing the switching state of the capacitor bank; N CB is the number of capacitor banks in the distribution network; Δt is the time variation, i.e. the variation of t; The second stage dispatching fee includes: load shedding penalty fee, the formula is as follows: Where: N is the number of nodes in the distribution network; η d,t is the load shedding penalty coefficient during period t; Ldi,t is the load shedding power during period t; The distribution network flow constraint is expressed as follows: Where: P i,t , Q i,t are the active power and reactive power injected into node i during period t; P ij,t , Q ij,t are the active power and reactive power flowing through branch ij during period t respectively; P ji,t , Q ji,t are respectively the active power and reactive power flowing through branch ji in period t; F(i) is the set of end nodes of the branch with node i as the first end point; T(i) is the set of first nodes of the branch with node i as the end node; l ij,t is the current amplitude of branch ij during period t; l ji,t is the current amplitude of branch ji during period t; r ij , x ij are the resistance and reactance of branch ij, r ji , x ji are the resistance and reactance of branch ji respectively; V i,t is the voltage amplitude of node i during period t, V j,t is the voltage amplitude of node j during period t; The power flow constraints of the transmission network are expressed as follows: Where: P′ IJ,t , Q′ IJ,t are the active power and reactive power flowing through the transmission network branch IJ during period t; P′ JI,t , Q' JI,t are respectively the active power and reactive power flowing through the transmission network branch JI in period t; F'(I) is the set of terminal nodes of the branch with node I as the first end point of the transmission network; T'(I) is the set of first section nodes of the branch with node I as the end node of the transmission network; and are respectively the active output and reactive output of the generator at node I in the transmission network during period t; and are the active and reactive loads of node I in period t respectively; V′ I,t is the voltage amplitude of the transmission network node I during period t, V′ J,t is the voltage amplitude of the transmission network node J during period t; r′ IJ , x′ IJ are the resistance and reactance of the transmission network branch IJ respectively; V o is the reference node voltage amplitude, Z IJ is an integer variable, i.e., the line state variable of the transmission network branch IJ, which is 0 for open and 1 for closed, and K is a constant; The switching constraints of grouped capacitor banks are expressed as follows: Where: C CB,t,i is the number of switched capacitor groups at node i during period t; is the maximum switching quantity of the capacitor bank at node i; They are state variables of whether the number of capacitor groups switched in period t increases or decreases; is the maximum daily switching times of the capacitor bank at node i; S CBN is the rated injection power per unit capacitor bank; U CB,t,i is the voltage amplitude of the capacitor bank at node i during period t, Inject reactive power into the capacitor bank at node i during period t; The energy storage charging and discharging constraints are expressed as follows: in ch,t,i +in dis,t,i ≤1; ES 1,i =EST ,i ; in: are the upper limit of energy storage charging power and the upper limit of discharging power of node i in period t respectively; u ch,t,i ,u dis,t,i are the energy storage charging state variable and discharging state of node i in period t respectively; The upper limit of the number of times of energy storage charging and discharging per day; ES t,i represents the energy storage capacity of node i during period t; ES 1,i represents the energy storage capacity of node i in the first period; ES T,i represents the energy storage capacity of node i during period T; η ch , η dis are the energy storage charging efficiency and discharging efficiency respectively; are the upper and lower limits of the energy storage capacity of node i respectively; The system safety constraints are expressed as follows: In min ≤V i,t ≤V max ; l min ≤l ij,t ≤l max ； In min ≤V′ I,t ≤V′ max ; l' min ≤l′ IJ,t ≤l' max ； Where: V min 、V max are the minimum and maximum values ​​of the node voltage in the distribution network respectively; l min , l max are the minimum and maximum values ​​of the current flowing through the distribution network branch ij; V′ min , V′ max are the minimum and maximum values ​​of the node voltage of the transmission network respectively; l' min , l' max are the minimum and maximum values ​​of the current flowing through the transmission network branch IJ; l′ IJ,t is the current flowing through the transmission network branch IJ during period t; The power constraint of the power purchased at the gateway is expressed as follows: in: are the minimum and maximum values ​​of the active power of the electricity purchased at the gateway; are the minimum and maximum values ​​of reactive power of electricity purchased at the gateway, respectively; The flexible load demand response constraint is expressed as follows: in: and are the increase and decrease of the transferable load of node i in period t, respectively, which are controlled by the user; and are the response states of the increase and decrease of the transferable load of node i in period t, respectively, which are integer variables; is the fixed load power of node i in period t; θ is the ratio coefficient of the transferable load that can respond to the total load; is the charging power of the electric vehicle at node i during period t; is the maximum charging power; t st and t end and the end charging time respectively; γ is the charging efficiency; The power of the electric vehicle at node i during period t; and They are the minimum and maximum capacity of electric vehicles respectively; The network reconstruction constraints are expressed as follows: 0≤l ij,t ≤y ij,t l max ； Where: S is an integer; y ij,t is the on-off state of branch ij, that is, y ij,t =1 means the switch is closed, y ij,t =0 means the branch is disconnected; The peak shaving and valley filling and voltage deviation constraints are expressed as follows: Where: K P , K V are power fluctuation coefficient and voltage deviation coefficient respectively; V i,N is the rated voltage amplitude of node i, N is the number of nodes; ΔP0 and ΔV0 are the average power fluctuation and average voltage offset respectively.

2. The adaptive robust optimization scheduling method for the main distribution coordination system according to claim 1 is characterized in that: The obtaining of the convex hull uncertainty set for photovoltaic output includes: Based on the historical data of photovoltaic output, the uncertainty of photovoltaic output is modeled using the convex hull, and the convex hull uncertainty set for photovoltaic output is obtained, which is expressed as follows: 0≤λ z ≤1; P vg =(P vg,1 ,...P vg,t ,...P vg,T ); Where: CU is the convex hull uncertainty set for photovoltaic output; z is the convex hull pole; Conv z is the convex hull extreme point set; z is the convex combination coefficient, P vg is the vector composed of photovoltaic output in each period; P vg,z is the photovoltaic output corresponding to the convex hull pole z.

3. The adaptive robust optimization scheduling method for the main distribution coordination system according to claim 2 is characterized in that: The two-stage adaptive robust optimization model of the main distribution coordination system is expressed as follows: min{c T x+max(ming T y)}; Ax ≥ a; Ex+Hy+GP vg +Mμ≥b; x min ≤x≤x max ; and min ≤y≤y max ; m min ≤μ≤μ max ; Where: c T is the coefficient matrix of the first-stage dispatch variables, including the cost of purchasing electricity at the gateway, the cost of charging and discharging energy storage, and the cost of switching on and off capacitor banks, g T is the coefficient matrix of the second-stage dispatching variables, including the load shedding penalty fee; A and a are the coefficient matrix and right-hand side vector of the inequality constraints of the gateway power purchase, capacitor bank switching, and energy storage charging and discharging rewritten in matrix-vector form; E, H, G, M, and b are the coefficient matrix and right-hand side vector of the first-stage dispatching variables, the second-stage dispatching variables, the photovoltaic output, and the state variables of the main-distribution coordination system after all the first- and second-stage coupling inequality constraints are rewritten in matrix-vector form; μ is the state variable of the main-distribution coordination system; x min ,y min , μ min are the minimum values ​​of x, y, and μ respectively; x max ,y max , μ max are the maximum values ​​of x, y, and μ respectively.

4. The method for adaptive robust optimization scheduling of the main distribution coordination system according to claim 3 is characterized in that: The column and constraint generation algorithm is used to solve the two-stage adaptive robust optimization model of the main distribution coordination system to obtain the optimal scheduling strategy of the main distribution coordination system, including: The two-stage adaptive robust optimization model of the main distribution coordination system is decomposed into a main problem and sub-problems, and the sub-problems with double-level optimization problems are transformed into single-level optimization sub-problems. The main problem is solved to obtain the optimal dispatching variables of the first stage under the constraints of limited severe scenarios, and the lower bound of the two-stage adaptive robust optimization model of the main and distribution coordination system is updated; Based on the worst scenario corresponding to the given first-stage optimal dispatching variables, the single-layer optimization subproblem is solved to obtain the second-stage optimal dispatching variables corresponding to the scenario, and the upper bound of the two-stage adaptive robust optimization model of the main and distribution coordination system is updated; The upper bound of the two-stage adaptive robust optimization model of the main and distribution coordination system is provided to the main problem for iteration; Based on the increase in the number of iterations, it converges to the optimal solution of the two-stage adaptive robust optimization model of the main and distribution cooperative system, that is, the optimal scheduling strategy of the main and distribution cooperative system.

5. The method for adaptive robust optimization scheduling of the main distribution coordination system according to claim 4 is characterized in that: The two-stage adaptive robust optimization model of the main distribution coordination system is decomposed into a main problem and sub-problems, and the sub-problems of the double-layer optimization problem are transformed into single-layer optimization sub-problems, including: The main question is as follows: min{c T x+β}; b≥g T m (k) ; Ax ≥ a; x min ≤x≤x max ; and min ≤y (k) ≤y max ; m min ≤μ (k) ≤μ max ; Among them: β is an auxiliary variable; y (k) Schedule variables for the second phase of the kth iteration; is the photovoltaic output of the kth iteration, μ (k) is the state variable of the master-distributor coordination system in the kth iteration; m is the total number of iterations; The sub-questions are as follows: max(min g T y); Ex * +Hy+GP vg +Mμ≥b; and min ≤y≤y max ; m min ≤μ≤μ max ; Where: x * is the optimal scheduling variable in the first stage; Based on the second-order cone duality rule and dual transformation, the sub-problems of the two-level optimization problem are transformed into single-level optimization sub-problems, including: The sub-problem is equivalently transformed into: max(minC T Y); DY≥(b-GP vg -Ex * )(π); U·Y≤Y max (p); U·Y≥Y min (s); Among them: π, ρ, σ are all linear constraint dual variables; D = [HM], U is the unit diagonal matrix; The above equivalent transformed sub-problem is transformed into a single-layer optimization sub-problem, which is expressed as follows: max{π T (b-GP vg -Ex * )+r T Y max +s T Y min }; p T D+r T U+s T U=C; π≥0,σ≥0,ρ≤0; in:(*) T is the transpose of *.

Images