A multi-stage optimization planning method for a new flexible interconnected power distribution system

Abstract

The present invention provides a multi-stage optimization planning method for a new type of flexible interconnected power distribution system, comprising: step (1): determining basic parameters, determining initial conditions for model establishment and solution; step (2): dividing the total planning period into multiple sub-planning stages according to the total planning period and the long-term forecast data of the power load; step (3): considering the total investment cost, the total operating cost and the operating cost of each scenario during the planning period, and establishing a planning-optimization double-layer model; step (4): solving the model using a GA algorithm and a cone programming hybrid algorithm to obtain a total planning scheme. The present invention provides a multi-stage optimization planning method for a new type of flexible interconnected power distribution system, dividing the distribution network planning period into multiple stages, constructing the distribution network in stages, establishing a multi-stage planning model for the distribution network that meets the load requirements, and establishing a planning-optimization double-layer model for a new type of flexible interconnected power distribution system based on distributed power sources and soft switching characteristics, while considering economic and environmental benefits.

Description

Technical Field

[0001] The present invention belongs to the technical field of distribution network optimization planning, and in particular relates to a multi-stage optimization planning method for a new type of flexible interconnected distribution system. Background Art

[0002] With the rapid scale development of distributed generation (DG), its grid-connected capacity continues to increase, greatly reducing system power loss, improving power supply reliability and reducing environmental pollution. However, it also brings problems such as bidirectional voltage over-limit, excessive harmonics in the power grid and low-voltage three-phase imbalance to the traditional distribution network with limited regulation methods. Adding soft open point (SOP) in the distribution network can achieve fast, dynamic and continuous active and reactive power flow control, balance the load flow and optimize the system voltage distribution. The application of soft switch improves the controllability of the distribution network flow, greatly improves the flexibility and reliability of the distribution network power supply, and can effectively solve the problems caused by the high proportion of distributed generation connected to the traditional distribution network. Connecting SOP in the traditional distribution network is one of the future development directions of the flexible interconnected new distribution system.

[0003] In the traditional distribution system, when the power grid is operating normally, the load point is powered by only a single power source, the output of the generator set is orderly and controllable, and the flow direction is also unidirectional. However, the access of DG will diversify the power supply of the distribution network, and the access of SOP will also change the line topology of the distribution network, which will cause the node voltage and the direction of the line flow to change, and the traditional distribution system planning method will no longer be applicable. Therefore, the planning of a new flexible interconnected distribution system with a high proportion of clean energy is an issue that needs to be studied urgently.

[0004] The medium- and long-term planning of distribution network has a large time span and needs to consider the load changes during the planning period, so the planning is often divided into several stages, which is called multi-stage planning. Multi-stage distribution network planning solves the distribution network planning problem from a long-term perspective, fully considers the dynamic growth of load and the transition plan of each stage, takes the economy during the entire planning period as the optimization goal, and can improve the overall economic and social benefits of the distribution network. It is worth noting that electricity demand has its own development laws, so there must be an optimal multi-stage division method. For medium- and long-term distribution network planning, the grid planning plan is adjusted according to load changes to seek a dynamic equipment investment plan to ensure that the planning results are optimal within the entire planning period, which is of great significance to improving the economy of distribution network planning. Summary of the invention

[0005] In order to solve the above problems, the present invention provides a multi-stage optimization planning method for a flexible interconnected new distribution system, which performs long-term planning of a flexible interconnected new distribution system under the condition that distributed power sources and soft switches are connected to the distribution network. Considering the long-term nature of distribution network planning, the distribution network planning period is divided into multiple stages, and the distribution network is constructed in stages. A multi-stage planning model for the distribution network that meets the load requirements is established. Based on the characteristics of distributed power sources and soft switches, a flexible interconnected new distribution system planning-optimization double-layer model is established while considering economic and environmental benefits.

[0006] The present invention specifically provides a multi-stage optimization planning method for a new type of flexible interconnected power distribution system, comprising the following steps:

[0007] Step (1): First, it is necessary to determine the basic parameters such as load parameters, DG parameters, SOP parameters, branch parameters and prediction data to determine the initial conditions for model establishment and solution;

[0008] Step (2): Based on the overall planning period and the long-term forecast data of power load, the overall planning period is divided into multiple sub-planning stages by using the cluster analysis method considering the annual growth rate of load;

[0009] Step (3): Considering the total investment cost, total operating cost and operating cost of each scenario during the planning period, a planning-optimization two-layer model is established;

[0010] Step (4): Use the GA algorithm and cone programming hybrid algorithm to solve the model and obtain the overall planning solution.

[0011] In step (2), the annual load growth rate is taken into consideration and the use scenarios in the planning period are clustered and divided into multiple planning stages.

[0012] For long-term planning of distribution networks with a long planning period, the planning period is usually divided into several horizontal years for planning. According to different division bases, it can be divided into stage planning and multi-stage planning. Compared with stage planning, which can ensure that each stage planning is optimal, but is not necessarily the optimal solution for the total planning period, multi-stage planning dynamically considers the load changes, takes the minimum total cost generated by the construction and operation of the distribution network during the entire planning period as the planning goal, and obtains the power supply and line investment and construction plan under the multi-stage distribution network planning.

[0013] The long-term forecast data of power load is the basic data for distribution network planning. The total power load forecast data and the power load growth forecast data in the planning area are used as the basis for the stage division of multi-stage planning. The K-means clustering analysis method is used to cluster the annual load data, so as to divide the total planning stage into multiple sub-planning stages. For each sub-planning stage, the starting year of the stage is the planning base year of the planning stage, and the distribution network load demand in the end year of the stage is the requirement that needs to be met in the planning stage.

[0014] After dividing the entire planning period into multiple sub-planning stages according to the clustering method, multi-stage planning of the new flexible interconnected distribution system considering economy can be carried out.

[0015] Step (3) considers the total investment cost, total operating cost and operating cost of each scenario during the planning period, and establishes a planning-optimization two-layer model, which is as follows:

[0016] The multi-stage optimization planning method for the new flexible interconnected distribution system proposed in this invention establishes a planning-optimization double-layer model. The upper model is a long-term planning, including the distribution network expansion planning, DG site selection and capacity determination, and SOP site selection and capacity determination. The lower model is a short-term optimization, including the distribution network operation cost and operation status under various scenarios.

[0017] Step (31): Establish an upper-level long-term planning model, taking the minimum present value of total cost during the entire planning period as the optimization goal, and realize DG site selection and capacity determination, SOP site selection and capacity determination, and distribution network expansion planning.

[0018] The objective function is to minimize the total cost present value F1 during the entire planning period. The total cost present value is the sum of the investment cost and operating cost of each sub-planning stage converted into present value by the present value coefficient. Its expression is as follows:

[0019]

[0020] Where z represents the zth sub-planning stage; Ω z is the set of sub-planning stages; μ zy is the present value coefficient; f z1 、f z2 are the investment cost and operating cost of stage u respectively. The present value coefficient can convert the average annual cost of each stage to the current, and its expression is:

[0021] μ zy =(1+γ) -[(z-1)Y+y]

[0022] In the formula, γ represents the inflation coefficient; y represents the number of years that have passed in this stage; and Y represents the total number of years in each sub-stage.

[0023] (311) Investment cost z1 Including DG investment cost, SOP investment cost and line investment cost, the expression is as follows:

[0024] f z1 =τ(f z1,DG +f z1,SOP +f z1,L )

[0025] In the formula, f z1,DGrepresents the DG investment cost in stage z; f z1,SOP represents the SOP investment cost in stage z; f z1,L represents the construction cost of the line upgrade in stage z; τ represents the capital recovery coefficient, which is expressed as:

[0026]

[0027] Where P T The useful life of the equipment.

[0028] (3111)DG investment costf z1,DG

[0029]

[0030] In the formula, The unit investment costs of wind turbines and photovoltaics are shown separately; Respectively represent the installed capacity of wind turbines and photovoltaics at node i in stage z; i represents node i in the system; Ω D Indicates the set of nodes to be installed on the DG.

[0031] (3112)SOP investment cost z1,SOP

[0032]

[0033] In the formula, It represents the unit investment cost of SOP; represents the installed capacity of the SOP at line ij in stage z; ij represents line ij in the system; Ω S Represents the set of installation lines to be selected for SOP.

[0034] (3113) Line Upgrade Construction Costf z1,L

[0035]

[0036] Where, l represents the lth line in the system; Ω L Indicates the set of lines to be newly created; represents the cost of newly building line l in stage z; x z,i L It is a 0-1 variable, indicating whether the lth line is newly created in stage z.

[0037] (312) Operation cost z2

[0038]

[0039] Where s is the scene number; D s is the number of days included in scenario s; is the DG operating cost; Cost of running the SOP; is the line network loss cost; Cost of curtailing solar and wind power; The cost of purchasing electricity from the main grid.

[0040] (3121)DG operating costs

[0041]

[0042] In the formula, They represent the operating costs per unit of electricity generated by wind turbines and photovoltaics, respectively; They represent the power generation of wind turbine and photovoltaic at time t in the z planning stage respectively.

[0043] (3122)SOP operating costs

[0044]

[0045] In the formula, It represents the operating cost of SOP; represents the active power and reactive power injected by SOP at node i at time t in the z planning stage, Ω N Represents a collection of nodes in the system.

[0046] (3123) Line network loss cost

[0047]

[0048] In the formula, represents the line network loss cost coefficient; Represents line network loss, obtained by power flow calculation.

[0049] (3124) Cost of curtailing solar and wind power

[0050]

[0051] In the formula, Represent the power curtailment costs of wind turbines and photovoltaics respectively; They represent the power abandonment of wind turbines and photovoltaics at time t in the z planning stage.

[0052] (3125) Cost of purchasing electricity from the main grid

[0053]

[0054] In the formula, They represent the unit electricity purchase cost of the distribution network from the upper grid respectively; They represent the active load at node i at time t in the z planning stage; Indicates the net load.

[0055] (313) The constraints include DG access capacity constraints, SOP access capacity constraints, line investment constraints and multi-stage coordination constraints.

[0056] (3131) DG access capacity constraint, which constrains the DG capacity accessed by each node to not exceed the upper limit of the capacity allowed for access, and the DG penetration rate in each sub-stage to not exceed the upper limit.

[0057] (31311) DG candidate node access capacity limit

[0058]

[0059] In the formula, Indicates the upper limit of the DG capacity allowed to be installed at the candidate node i.

[0060] (31312) DG access total capacity constraint

[0061] ∑P DG,z ≤η∑P load,z

[0062] In the formula, ∑P DG,z represents the total DG installed capacity in the planning stage; ∑P load,z represents the total load capacity of the system in the z planning stage; η represents the maximum DG penetration rate allowed by the system.

[0063] (3132) SOP access capacity limit, constraining the SOP capacity accessed between each line to not exceed the upper limit of the operating access capacity

[0064]

[0065] In the formula, It indicates the upper limit of SOP capacity allowed to be connected between lines ij in the z planning stage.

[0066] (3133) Line investment constraint, which constrains the number of newly built lines in each sub-planning stage to be equal to the number of newly built load nodes

[0067]

[0068] In the formula, Indicates the number of newly added load nodes in stage z.

[0069] (3134) Coordination constraints between multiple stages constrain the construction of wind turbines, photovoltaics, SOPs and lines between each stage. The construction of the latter stage is based on the construction of the previous stage, ensuring the connection of the construction plans between each stage.

[0070]

[0071] Step (32): Establish a lower-level short-term optimization model to meet various constraints of the distribution network while minimizing the operating cost of each scenario and optimizing the operating status of the system under the DG, SOP site selection, capacity determination and grid expansion schemes given by the upper-level planning model.

[0072] The objective function is to minimize the distribution network operation cost F2 under each scenario, including DG loss, SOP loss and line loss under each scenario. The formula is as follows:

[0073]

[0074] The constraints should take into account SOP operation constraints, DG operation constraints, branch flow constraints and safe operation constraints.

[0075] (321)SOP operation constraints

[0076]

[0077] Where: A SOP is the SOP loss coefficient; P SOP,i,t , P SOP,j,t are the active power injected by SOP at node i and node j in period t respectively; Q SOP,i,t , Q SOP,j,t are the reactive power injected by SOP at node i and node j during period t respectively; are the active power loss of SOP at node i and node j in period t respectively; S SOP,ij is the installed SOP capacity between node i and node j.

[0078] (322)DG operation constraints

[0079]

[0080] In the formula, represents the actual active power output of DG at node j, Represents the maximum output that DG can achieve at node j.

[0081] (323) Branch Flow Constraints

[0082]

[0083] Where: and φi are the child node set and parent node set of node i respectively; P ik,t and Q ik,t P is the active power and reactive power flowing from node i to node k during period t; ji,t and Q ji,t is the active power and reactive power flowing from node j to node i during period t; P i,t and Q i,t is the active power and reactive power at node i during period t; R ij and X ij is the resistance and reactance of branch ij; and is the square of the voltage amplitude at node i and node j during period t; is the square of the current amplitude of branch ij during period t; P SOP,i,t and Q SOP,i,t P is the active power and reactive power injected by SOP at node i during period t; DG,i,t and Q DG,i,t P is the active power and reactive power injected by the distributed generation at node i during period t; L,i,t and Q L,i,t is the active power and reactive power consumed by the load at node i during period t.

[0084] (324) Safety Operation Constraints

[0085]

[0086] Where: and are the upper and lower limits of the square of the node voltage amplitude respectively; P max and -P max They are the upper and lower limits of the active power of branch ij respectively.

[0087] Step (4): A hybrid algorithm of GA algorithm and second-order cone programming is used to solve the planning-optimization two-layer model of the flexible interconnected new distribution system.

[0088] Compared with the prior art, the beneficial effects are:

[0089] (1) The bi-level programming-optimization model of the present invention is a large-scale mixed integer nonlinear programming model with multi-layer nesting, multi-variables and multi-constraints, and can be solved quickly and accurately by using a hybrid algorithm of genetic algorithm (GA) and second-order cone programming. The upper-level planning uses a genetic algorithm to generate a planning scheme and conduct a convergence analysis. The lower-level planning uses a cone programming algorithm to perform operation optimization calculations for each scenario under the scheme selected by the upper-level planning, and passes the optimization results to the upper-level planning for iterative optimization to obtain the optimal planning scheme.

[0090] (2) The present invention provides a method for long-term planning of a flexible interconnected new distribution system taking into account the access of distributed power sources and soft switches to the distribution network. First, the annual growth rate of the load is considered, and the use scenarios of the planning period are clustered and divided into multiple planning stages; then, an upper-level long-term planning model is established with the minimum present value of the total cost in the entire planning period as the optimization goal, and the DG site selection and capacity determination, SOP site selection and capacity determination, and distribution network grid expansion planning schemes are solved; then, under the DG, SOP site selection and capacity determination and grid expansion schemes given by the upper-level planning model, the operating cost of each scenario is minimized as the objective function, and various constraints of the distribution network are considered to establish a lower-level short-term optimization model to optimize the operating state of the system; finally, a hybrid algorithm of the GA algorithm and the second-order cone programming is used to solve the flexible interconnected new distribution system planning-optimization double-layer model. BRIEF DESCRIPTION OF THE DRAWINGS

[0091] Figure 1 The present invention is a flowchart of a multi-stage optimization planning method for a new type of flexible interconnected power distribution system.

[0092] Figure 2 It is a schematic diagram of the multi-stage division results of the present invention.

[0093] Figure 3 This is a framework diagram of the planning-optimization two-layer model proposed in the present invention. DETAILED DESCRIPTION

[0094] The specific implementation method of a multi-stage optimization planning method for a new type of flexible interconnected power distribution system of the present invention is described in detail below with reference to the accompanying drawings.

[0095] The mid- and long-term planning of distribution networks has a large time span and needs to consider the load changes during the planning period, so a multi-stage planning method is often used. The multi-stage planning of distribution networks fully considers the dynamic growth of loads and the transition plans of each stage, and takes the economy during the entire planning period as the optimization goal, which can improve the overall economic and social benefits of the distribution network. At the same time, considering the volatility of the output of distributed power sources and loads, the distribution network has a variety of operating scenarios. When solving the planning model, multiple typical scenarios should be considered.

[0096] like Figure 1 As shown, the present invention provides a multi-stage optimization planning method for a new flexible interconnected power distribution system, comprising the following steps:

[0097] Step (1): First, it is necessary to determine the basic parameters such as load parameters, DG parameters, SOP parameters, branch parameters and prediction data to determine the initial conditions for model establishment and solution;

[0098] Step (2): Based on the overall planning period and the long-term forecast data of power load, the overall planning period is divided into multiple sub-planning stages by using the cluster analysis method considering the annual growth rate of load;

[0099] Step (3): Considering the total investment cost, total operating cost and operating cost of each scenario during the planning period, a two-level planning-optimization model is established;

[0100] Step (4): Use the GA algorithm and cone programming hybrid algorithm to solve the model and obtain the overall planning solution.

[0101] In step (2), the annual load growth rate is taken into consideration and the use scenarios in the planning period are clustered and divided into multiple planning stages.

[0102] The long-term forecast data of power load is the basic data for distribution network planning. The total power load forecast data and the power load growth rate forecast data in the planning area are used as the basis for the stage division of multi-stage planning. The K-means clustering analysis method is used to cluster the annual load data, so as to divide the total planning stage into multiple sub-planning stages. For each sub-planning stage, the starting year of the stage is the planning base year of the planning stage, and the distribution network load demand in the ending year of the stage is the requirement that needs to be met in the planning stage. The schematic diagram of the multi-stage division result is shown in the figure below. Figure 2 shown.

[0103] Step (3) Considering the total investment cost, total operating cost and operating cost of each scenario during the planning period, a two-level planning-optimization model is established, specifically:

[0104] After dividing the entire planning period into multiple sub-planning stages according to the clustering method, multi-stage planning of a new flexible interconnected distribution system considering economy can be carried out.

[0105] After completing the multi-stage division, a planning-optimization two-layer model should be established. The upper model is a long-term planning, including the distribution network expansion planning, DG site selection and capacity determination, and SOP site selection and capacity determination. The lower model is a short-term optimization, including the distribution network operation cost and operation status under various scenarios. Its framework diagram is as follows: Figure 3 shown.

[0106] Step (31): Establish an upper-level long-term planning model, taking the minimum present value of total cost during the entire planning period as the optimization goal, and realize DG site selection and capacity determination, SOP site selection and capacity determination, and distribution network expansion planning.

[0107] The objective function is to minimize the total investment cost and operating cost during the entire planning period. The constraints considered include DG capacity constraints, SOP capacity constraints, line investment constraints, and multi-stage coordination constraints. The decision variables are the DG site selection and capacity determination plan, the SOP site selection and capacity determination plan, and the new line plan.

[0108] Considering the long planning period, the cost of each sub-planning stage needs to be converted to present value. The objective function is to minimize the total cost present value F1 during the entire planning period. The total cost present value is the sum of the investment cost and operating cost of each sub-planning stage converted into present value by the present value coefficient. Its expression is as follows:

[0109]

[0110] Where z represents the zth sub-planning stage; Ω z is the set of sub-planning stages; μ zy is the present value coefficient; f z1 、f z2 are the investment cost and operating cost of stage u respectively. The present value coefficient can convert the average annual cost of each stage to the current period, and its expression is: zy =(1+γ) -[(z-1)Y+y]

[0111] In the formula, γ represents the inflation coefficient; y represents the number of years that have passed in this stage; and Y represents the total number of years in each sub-stage.

[0112] (311) Investment cost z1 Including DG investment cost, SOP investment cost and line investment cost, the expression is as follows:

[0113] f z1 =τ(f z1,DG +f z1,SOP +f z1,L )

[0114] In the formula, f z1,DG represents the DG investment cost in stage z; f z1,SOP represents the SOP investment cost in stage z; f z1,L represents the construction cost of the line upgrade in stage z; τ represents the capital recovery coefficient, which is expressed as:

[0115]

[0116] Where P T The useful life of the equipment.

[0117] (3111)DG investment costf z1,DG

[0118]

[0119] In the formula, The unit investment costs of wind turbines and photovoltaics are shown separately; Respectively represent the installed capacity of wind turbines and photovoltaics at node i in stage z; i represents node i in the system; Ω D Indicates the set of nodes to be installed on the DG.

[0120] (3112)SOP investment cost z1,SOP

[0121]

[0122] In the formula, It represents the unit investment cost of SOP; represents the installed capacity of the SOP at line ij in stage z; ij represents line ij in the system; Ω S Represents the set of installation lines to be selected for SOP.

[0123] (3113) Line Upgrade Construction Costf z1,L

[0124]

[0125] Where, l represents the lth line in the system; Ω L Indicates the set of lines to be newly created; C z l represents the cost of newly building line l in stage z; x z,i L It is a 0-1 variable, indicating whether the lth line is newly created in stage z.

[0126] (312) Operation cost z2

[0127]

[0128] Where s is the scene number; D s is the number of days included in scenario s; is the DG operating cost; Cost of running the SOP; is the line network loss cost; Cost of curtailing solar and wind power; The cost of purchasing electricity from the main grid.

[0129] (3121)DG operating costs

[0130]

[0131] In the formula, They represent the operating costs per unit of electricity generated by wind turbines and photovoltaics, respectively; They represent the power generation of wind turbine and photovoltaic at time t in the z planning stage respectively.

[0132] (3122)SOP operating costs

[0133]

[0134] In the formula, It represents the operating cost of SOP; represents the active power and reactive power injected by SOP at node i at time t in the z planning stage, Ω N Represents a collection of nodes in the system.

[0135] (3123) Line network loss cost

[0136]

[0137] In the formula, represents the line network loss cost coefficient; Represents line network loss, obtained by power flow calculation.

[0138] (3124) Cost of curtailing solar and wind power

[0139]

[0140] In the formula, Represent the power curtailment costs of wind turbines and photovoltaics respectively; They represent the power abandonment of wind turbines and photovoltaics at time t in the z planning stage.

[0141] (3125) Cost of purchasing electricity from the main grid

[0142]

[0143] In the formula, They represent the unit electricity purchase cost of the distribution network from the upper grid respectively; They represent the active load at node i at time t in the z planning stage; Indicates the net load.

[0144] (313) The constraints of the upper-level planning model include DG access capacity constraints, SOP access capacity constraints, line investment constraints, and multi-stage coordination constraints.

[0145] (3131) DG access capacity constraint, which constrains the DG capacity accessed by each node to not exceed the upper limit of the capacity allowed for access, and the DG penetration rate in each sub-stage to not exceed the upper limit.

[0146] (31311) DG candidate node access capacity limit

[0147]

[0148] In the formula, Indicates the upper limit of the DG capacity allowed to be installed at the candidate node i.

[0149] (31312) DG access total capacity constraint

[0150] ∑P DG,z ≤η∑P load,z

[0151] In the formula, ∑P DG,z represents the total DG installed capacity in the planning stage; ∑P load,z represents the total load capacity of the system in the z planning stage; η represents the maximum DG penetration rate allowed by the system.

[0152] (3132) SOP access capacity limit, constraining the SOP capacity accessed between each line to not exceed the upper limit of the operating access capacity

[0153]

[0154] In the formula, It indicates the upper limit of SOP capacity allowed to be connected between lines ij in the z planning stage.

[0155] (3133) Line investment constraint, which constrains the number of newly built lines in each sub-planning stage to be equal to the number of newly built load nodes

[0156]

[0157] In the formula, Indicates the number of newly added load nodes in stage z.

[0158] (3134) Coordination constraints between multiple stages constrain the construction of wind turbines, photovoltaics, SOPs and lines between each stage. The construction of the latter stage is based on the construction of the previous stage, ensuring the connection of the construction plans between each stage.

[0159]

[0160] Step (32): Establish a lower-level short-term optimization model based on the distribution network planning and construction plan given by the upper-level planning model, consider various scenarios in each sub-planning stage, take the lowest operating cost of each scenario as the objective function, meet the SOP operation constraints, DG operation constraints, branch flow constraints and safety operation constraints, solve the decision variables such as line flow, SOP transmission power, node voltage and amplitude, and realize the optimization of distribution system operation.

[0161] Under the DG, SOP site selection, capacity determination and grid expansion schemes given by the upper-level planning model, while meeting various constraints of the distribution network, the operating cost of each scenario is minimized and the operating status of the system is optimized.

[0162] The objective function is to minimize the distribution network operation cost F2 under each scenario, including DG loss, SOP loss and line loss under each scenario. The formula is as follows:

[0163]

[0164] The constraints should take into account SOP operation constraints, DG operation constraints, branch flow constraints and safe operation constraints.

[0165] (321)SOP operation constraints

[0166]

[0167] Where: A SOP is the SOP loss coefficient P SOP,i,t , P SOP,j,t are the active power injected by SOP at node i and node j in period t respectively; Q SOP,i,t , Q SOP,j,t are the reactive power injected by SOP at node i and node j during period t respectively; are the active power loss of SOP at node i and node j in period t respectively; S SOP,ij is the installed SOP capacity between node i and node j.

[0168] (322)DG operation constraints

[0169]

[0170] In the formula, represents the actual active power output of DG at node j, Represents the maximum output that DG can achieve at node j.

[0171] (323) Branch Flow Constraints

[0172]

[0173] Where: and φ i are the child node set and parent node set of node i respectively; P ik,t and Q ik,t P is the active power and reactive power flowing from node i to node k during period t; ji,t and Q ji,t is the active power and reactive power flowing from node j to node i during period t; P i,t and Q i,t is the active power and reactive power at node i during period t; R ij and X ij is the resistance and reactance of branch ij; and is the square of the voltage amplitude at node i and node j during period t; is the square of the current amplitude of branch ij during period t; P SOP,i,t and Q SOP,i,t P is the active power and reactive power injected by SOP at node i during period t; DG,i,t and Q DG,i,t P is the active power and reactive power injected by the distributed generation at node i during period t; L,i,t and Q L,i,t is the active power and reactive power consumed by the load at node i during period t.

[0174] (324) Safety Operation Constraints

[0175]

[0176] -P max ≤P ij,t ≤P max

[0177] Where: and are the upper and lower limits of the square of the node voltage amplitude respectively; P max and -P max They are the upper and lower limits of the active power of branch ij respectively.

[0178] Step (4): A hybrid algorithm of GA algorithm and second-order cone programming is used to solve the planning-operation two-layer model of the flexible interconnected new distribution system.

[0179] For the above-mentioned planning-operation double-layer model, it can be seen that the upper-level planning passes the DG site selection and capacity determination, SOP site selection and capacity determination, and distribution network grid expansion plan to the lower-level operation optimization model. On this basis, the lower layer optimizes the distribution system operation status of each scenario in each sub-planning period, and returns the optimization results to the upper layer. The upper-level planning model calculates the objective function value of the current planning scheme using the results uploaded by the lower-level optimization model. This planning problem is a large-scale mixed integer nonlinear planning problem with multi-layer nesting, multi-variables, and multi-constraints. It is solved quickly and accurately using a hybrid algorithm of genetic algorithm and second-order cone programming. The upper-level planning uses genetic algorithm to generate planning schemes and conduct convergence analysis. The lower-level planning uses cone programming algorithm to perform operation optimization calculations for each scenario under the scheme selected by the upper-level planning, and passes the optimization structure to the upper-level planning, and finally solves the optimal distribution network planning scheme.

[0180] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not intended to limit the present invention. A person skilled in the art should understand that the specific implementation of the present invention can be modified or replaced by equivalents, but these modifications or changes are within the scope of protection of the pending claims.

Claims

1. A multi-stage optimization planning method for a new flexible interconnected power distribution system, characterized in that: The following steps are involved: Step (1): First, it is necessary to determine the load parameters, DG parameters, SOP parameters, branch parameters and basic parameters of the prediction data to determine the initial conditions for model establishment and solution; Step (2): Based on the overall planning period and the long-term forecast data of power load, the overall planning period is divided into multiple sub-planning stages by using the cluster analysis method considering the annual growth rate of load; Step (3): Considering the total investment cost, total operating cost and operating cost of each scenario during the planning period, a planning-optimization two-layer model is established; Step (4): Use the GA algorithm and cone programming hybrid algorithm to solve the model and obtain the overall planning solution.

2. According to claim 1, a multi-stage optimization planning method for a new flexible interconnection power distribution system is characterized in that: In step (2), the annual growth rate of load is taken into consideration and the use scenarios in the planning period are clustered and divided into multiple planning stages, specifically: The total power load forecast data and the power load growth rate forecast data in the planning area are used as the basis for the stage division of multi-stage planning. The K-means clustering analysis method is used to cluster the annual load data, so as to divide the total planning stage into multiple sub-planning stages.

3. A multi-stage optimization planning method for a new flexible interconnected power distribution system according to claim 2, characterized in that: Step (3) considers the total investment cost, total operating cost and operating cost of each scenario during the planning period, and establishes a planning-optimization two-layer model, which specifically includes: Step (31): Establish an upper-level long-term planning model, taking the minimum present value of total cost during the entire planning period as the optimization goal, and realize DG site selection and capacity determination, SOP site selection and capacity determination, and distribution network expansion planning; Step (32): Establish a lower-level short-term optimization model to meet various constraints of the distribution network while minimizing the operating cost of each scenario and optimizing the operating status of the system under the DG, SOP site selection, capacity determination and grid expansion schemes given by the upper-level planning model.

4. A multi-stage optimization planning method for a new flexible interconnected power distribution system according to claim 3, characterized in that: The objective function of the upper-level long-term planning model is to minimize the total cost present value F1 during the entire planning period; the total cost present value is obtained by converting the investment cost and operating cost of each sub-planning stage into present value through the present value coefficient and then accumulating them, and its expression is as follows: Where z represents the zth sub-planning stage; Ω z is the set of sub-planning stages; μ zy is the present value coefficient; f z1 、f z2 are the investment cost and operating cost of stage u respectively; the present value coefficient converts the average annual cost of each stage to the current, and its expression is: m zy =(1+c) -[(z-1)Y+y] In the formula, γ represents the inflation coefficient; y represents the number of years that have passed in this stage; and Y represents the total number of years in each sub-stage.

5. A multi-stage optimization planning method for a new flexible interconnected power distribution system according to claim 4, characterized in that: The investment cost f z1 Including DG investment cost, SOP investment cost and line investment cost, the expression is as follows: f z1 =τ(f z1,DG +f z1,SOP +f z1,L ) In the formula, f z1,DG represents the DG investment cost in stage z; f z1,SOP represents the SOP investment cost in stage z; f z1,L represents the construction cost of the line upgrade in stage z; τ represents the capital recovery coefficient, which is expressed as: Where P T The service life of the equipment; (3111)DG investment costf z1,DG In the formula, The unit investment costs of wind turbines and photovoltaics are shown separately; Respectively represent the installed capacity of wind turbines and photovoltaics at node i in stage z; i represents node i in the system; Ω D Indicates the set of nodes to be installed on DG; (3112)SOP investment cost z1,SOP In the formula, It represents the unit investment cost of SOP; represents the installed capacity of the SOP at line ij in stage z; ij represents the line ij in the system; Ω S Represents the set of installation lines to be selected for SOP; (3113) Line Upgrade Construction Costf z1,L In the formula, l represents the lth line in the system; Ω L Indicates the set of lines to be newly created; C z l represents the cost of newly building line l in stage z; x z,i L It is a 0-1 variable, indicating whether the lth line is newly created in stage z.

6. A multi-stage optimization planning method for a new flexible interconnected power distribution system according to claim 5, characterized in that: The running cost f z2 The calculation method is: Where s is the scene number; D s is the number of days included in scenario s; is the DG operating cost; Cost of running the SOP; is the line network loss cost; Cost of curtailing solar and wind power; Cost of purchasing electricity from the main grid; (3121)DG operating costs In the formula, They represent the operating costs per unit of electricity generated by wind turbines and photovoltaics, respectively; They represent the power generation of wind turbine and photovoltaic at time t in the z planning stage respectively; (3122)SOP operating costs In the formula, Indicates the operating cost of the SOP; represents the active power and reactive power injected by SOP at node i at time t in the z planning stage, Ω N Represents the set of nodes in the system; (3123) Line network loss cost In the formula, represents the line network loss cost coefficient; Represents line network loss, obtained by power flow calculation; (3124) Cost of curtailing solar and wind power In the formula, Represent the power curtailment costs of wind turbines and photovoltaics respectively; They represent the abandoned power of wind turbines and photovoltaics at time t in the z planning stage; (3125) Cost of purchasing electricity from the main grid In the formula, They represent the unit electricity purchase cost of the distribution network from the upper grid respectively; They represent the active load at node i at time t in the z planning stage; Indicates the net load.

7. A multi-stage optimization planning method for a new flexible interconnected power distribution system according to claim 6, characterized in that: The constraints include DG access capacity constraints, SOP access capacity constraints, line investment constraints and multi-stage coordination constraints; (3131) DG access capacity constraint, which constrains the DG capacity accessed by each node to not exceed the upper limit of the capacity allowed for access, and the DG penetration rate in each sub-stage to not exceed the upper limit; (31311) DG candidate node access capacity limit In the formula, Indicates the upper limit of the DG capacity allowed to be installed at the candidate node i; (31312) DG access total capacity constraint ∑P DG,z ≤η∑P load,z In the formula, ∑P DG,z represents the total DG installed capacity in the planning stage; ∑P load,z represents the total load capacity of the system in the z planning stage; η represents the maximum DG penetration rate allowed by the system; (3132) SOP access capacity limit, constraining the SOP capacity accessed between each line to not exceed the upper limit of the operating access capacity In the formula, It represents the upper limit of the SOP capacity allowed to be connected between lines ij in the z planning stage; (3133) Line investment constraint, which constrains the number of newly built lines in each sub-planning stage to be equal to the number of newly built load nodes In the formula, Indicates the number of newly added load nodes in stage z; (3134) Coordination constraints between multiple stages, constraining the construction of wind turbines, photovoltaics, SOPs and lines between each stage. The construction of the latter stage is based on the construction of the previous stage, ensuring the connection of the construction plans between each stage:

8. A multi-stage optimization planning method for a new flexible interconnected power distribution system according to claim 7, characterized in that: Step (32): Establish a lower-level short-term optimization model. Under the given DG, SOP site selection, capacity determination and grid expansion scheme of the upper-level planning model, meet various constraints of the distribution network, minimize the operating cost of each scenario, and optimize the operating status of the system. Specifically: The objective function is to minimize the distribution network operation cost F2 under each scenario, including DG loss, SOP loss and line loss under each scenario. The formula is as follows: The constraints should take into account SOP operation constraints, DG operation constraints, branch power flow constraints and safe operation constraints; (321)SOP operation constraints Where: A SOP is the SOP loss coefficient; P SOP,i,t , P SOP,j,t are the active power injected by SOP at node i and node j in period t respectively; Q SOP,i,t , Q SOP,j,t are the reactive power injected by SOP at node i and node j during period t respectively; are the active power loss of SOP at node i and node j in period t respectively; S SOP,ij is the installed SOP capacity between node i and node j; (322)DG operation constraints In the formula, represents the actual active power output of DG at node j, represents the maximum output that the DG can achieve at node j; (323) Branch Flow Constraints Where: and φ i are the child node set and parent node set of node i respectively; P ik,t and Q ik,t is the active power and reactive power flowing from node i to node k during period t; P ji,t and Q ji,t is the active power and reactive power flowing from node j to node i during period t; P i,t and Q i,t is the active power and reactive power at node i during period t; R ij and X ij is the resistance and reactance of branch ij; and is the square of the voltage amplitude at node i and node j during period t; is the square of the current amplitude of branch ij during period t; P SOP,i,t and Q SOP,i,t P is the active power and reactive power injected by SOP at node i during period t; DG,i,t and Q DG,i,t P is the active power and reactive power injected by the distributed generation at node i during period t; L,i,t and Q L,i,t is the active power and reactive power consumed by the load at node i during period t; (324) Safety Operation Constraints -P max ≤P ij,t ≤P max Where: and are the upper and lower limits of the square of the node voltage amplitude respectively; P max and -P max They are the upper and lower limits of the active power of branch ij respectively.

9. A multi-stage optimization planning method for a new flexible interconnected power distribution system according to claim 8, characterized in that: Step (4) specifically adopts a hybrid algorithm of GA algorithm and second-order cone programming to solve the planning-optimization two-layer model of the flexible interconnected new distribution system.

Images