Snowflake type power distribution network load space transfer optimization method considering N-1 safety and reconstruction

Abstract

The invention relates to a snowflake-type power distribution network load space transfer optimization method considering N-1 safety and reconstruction. The method comprises the steps of basic data input, load transfer capability analysis and expression, reconstruction time period division, model establishment, conversion, solution and result output. According to the method, the N-1 safety criterion is considered, and the snowflake type power distribution network load transfer capacity is analyzed and expressed based on the topological incidence matrix. Then, taking the load balance degree and the minimum active power loss as targets, and establishing a snowflake type power distribution network multi-period load space transfer model considering N-1 security constraint and switch combination reconstruction; and then, converting the model through a mathematical programming method, solving the model by using commercial solvers such as CPLEX and the like, and outputting a related result. Based on the advantages of a snowflake type power distribution network frame structure, the load space transfer capacity of a feeder line cluster can be effectively utilized through sectional interconnection switch combination reconstruction, so that light and heavy load main transformers and line loads are balanced, large-range flexible transfer of the load space is achieved, and a larger safety transfer margin is provided for power grid operation.

Description

Technical Field

[0001] This invention belongs to the field of power distribution system optimization operation technology, and in particular to a snowflake-shaped power distribution network load spatial transfer optimization method that takes into account N-1 safety and reconfiguration. Background Technology

[0002] Urban power distribution networks serve a wide range of users, characterized by numerous points of operation, diverse operating modes, uneven regional development, uneven distribution of power sources and loads, and temporal and spatial mismatches in power supply and demand. Their loads naturally possess multi-dimensional potential for transfer across time and space. Due to the large number of sectionalizing switches in urban power distribution networks, their network topology is flexible and adaptable. Network reconfiguration based on these sectionalizing switches is considered a crucial means of achieving spatial load transfer. By optimizing and adjusting the opening and closing states of sectionalizing switches and interconnecting switches in a snowflake-shaped distribution network through network reconfiguration, the network topology can be altered. This allows for the transfer of some loads to interconnecting lines, enabling flexible power transfer between different lines and giving the snowflake-shaped distribution network the characteristic of flexible spatial load transfer.

[0003] With urban users increasingly demanding higher power quality, system security has become a key concern in real-time dispatching, generally referred to as the N-1 safety criterion in distribution networks. Currently, research on network reconfiguration based on segmented tie switches is insufficient. Its applications mostly involve load spatial transfer at the transformer substation or single feeder level, and rarely consider N-1 safety verification analysis for reconfiguration schemes, making it difficult to meet the safe operation requirements of new distribution systems. Furthermore, traditional N-1 safety verification methods are cumbersome, essentially requiring repeated searches for each fault scenario, necessitating multiple calculations. Especially during reconfiguration, with changes in network topology, the computational efficiency of performing N-1 safety verification element-by-element is low. This makes traditional N-1 safety verification analysis difficult to apply to load spatial transfer problems at the feeder cluster level of large-scale complex distribution networks. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a snowflake-shaped distribution network load spatial transfer optimization method that considers N-1 safety and reconfiguration. This method can effectively utilize the load spatial transfer capability of feeder clusters to balance lightly and heavily loaded main transformers and line loads, providing a greater safety margin for grid operation. Snowflake-shaped distribution networks that aggregate multiple feeders into clusters can fully utilize the complementary differences in load types within different feeders, providing a support platform for load transfer over a larger spatial range.

[0005] The technical problem solved by this invention is achieved through the following technical solution: A snowflake-shaped distribution network load spatial transfer optimization method considering N-1 safety and reconfiguration includes the following steps: Step 1: Obtain snowflake-shaped distribution network line parameters, load power, and distributed photovoltaic power output data for the area to be studied; Step 2: Based on Step 1 and using the topological correlation matrix, analytically express the load transfer capacity of the snowflake-shaped distribution network; Step 3: Perform power flow calculation on the initial state of the snowflake-shaped distribution network, and dynamically reconstruct the time period division of the load rate data of each feeder based on Fisher's optimal partitioning method; Step 4: With the goal of reducing active power loss and load balance, establish a snowflake-shaped distribution network load spatial transfer model that takes into account N-1 security and network reconfiguration based on the results of Step 3, and construct constraints based on transfer capacity. Step 5: Transform the snowflake-shaped distribution network load spatial transfer model into a mixed integer second-order cone programming model; Step 6: Solve the transformed mixed-integer second-order cone programming model; Step 7: Output the relevant results.

[0006] Furthermore, step 2 includes the following steps: Step 2.1: Establish a feeder correlation matrix based on the interconnection relationships between feeders in a snowflake-shaped distribution network; Step 2.2: Establish intra-station interconnection and transfer matrices and inter-station interconnection and transfer matrices respectively according to the feeder correlation matrix and the different feeder load transfer methods; Step 2.3: Establish the main transformer margin matrix and the feeder margin matrix based on the main transformer capacity constraints and the feeder capacity constraints; Step 2.4: Calculate the maximum transfer capacity of each feeder using the interconnection transfer matrix and margin matrix, then compare it with the load that needs to be transferred to determine whether the N-1 criterion is met, and calculate the transfer margin.

[0007] Furthermore, the specific implementation method of step 2.1 is as follows: Among them, the power grid has a total of n The substations are numbered 1, 2, ... n The number of feeders corresponding to each substation is N 1, N 2, ..., Nn There are a total of N One feeder line, N=N 1+ N 2+…+ Nn ,matrix F In F i,j Indicates feeder F i With feeder F jThe relationships between them.

[0008] Furthermore, in step 2.2, the internal communication and supply matrix is ​​constructed. A for: Inter-station communication and transfer matrix B for: .

[0009] Furthermore, the specific implementation method of step 2.3 is as follows: calculate the principal variable capacity matrix: Among them, matrix R The i element R i Indicates the main variable R i The capacity to accommodate loads is calculated, and the feeder capacity matrix is ​​set up as follows: Among them, matrix L The i element L i Indicates feeder L i The capacity to accept load is calculated, and the main transformer load factor matrix is ​​set as follows: Among them, matrix RM The i element RM i Indicates the main variable R i The load factor is calculated, and the feeder load factor matrix is ​​set up as follows: Among them, matrix LM The i element LM i Indicates feeder L i Based on the load factor, calculate the main transformer margin matrix: Among them, matrix The i element Indicates the main variable R i The margin, i.e., the main variable R i Calculate the feeder margin matrix based on the maximum load that can still be accommodated: .

[0010] Among them, matrix The i element Indicates feeder L i The margin, i.e., the feeder L i The maximum load that can still be accepted.

[0011] Furthermore, the specific implementation method of step 2.4 is as follows: calculate the station's internal transfer capacity matrix. TAM and TBM : Calculate the maximum transfer capacity matrix: Calculate the supply margin matrix: Among them, the supply margin matrix elements in feeder L i The remaining transfer margin after load transfer, if This indicates the feeder L i The maximum power transfer capacity during a fault is greater than or equal to the load it carries, therefore the feeder... L i satisfy N -1 Safety Rule: A positive absolute value for the supply margin indicates better safety; conversely, a negative absolute value indicates poorer safety. Then the feeder L i Not satisfied N -1 is the safety criterion. The larger the absolute value of the supply margin, the less safe the system is.

[0012] Moreover, the specific implementation method of step 3 is as follows: based on the initial network topology of the snowflake-shaped distribution network, perform time-series power flow calculation to obtain the branch power flow direction and the load rate of each feeder, and use Fisher's optimal segmentation method to divide the initial load rate data of each feeder into time periods to obtain the optimal segment for dynamic reconstruction.

[0013] Furthermore, the specific implementation method of step 1 is as follows: Furthermore, the objective function of the snowflake-shaped distribution network load spatial transfer optimization model, which takes into account N-1 security and network reconfiguration, in step 4 is: in, f It is the comprehensive objective function. f 1 represents the system's active power loss. f 2 is load balancing. and These represent the weighting coefficients for power loss and load balancing, respectively. , Represents the set of branches; This represents the set of feeders downstream of the main transformer; T This indicates the total number of time periods. The time interval for calculation, Indicates connection i Ring mesh box and j Branch of ring network box No. ij The resistance value; Indicates a branch ij exist t Current amplitude at time , Indicates a branch ij exist t Transmission capacity at any given moment Indicates a branch ij Maximum transmission capacity.

[0014] Furthermore, the constraints of the snowflake-shaped distribution network load space transfer optimization model that takes into account N-1 security and network reconfiguration include N-1 security transfer margin constraints, power flow equation constraints, basic safe operation constraints, power balance constraints, network topology reconfiguration constraints, and switching action constraints.

[0015] The advantages and positive effects of this invention are: This invention considers the N-1 safety criterion and analytically expresses the load transfer capacity of a snowflake-shaped distribution network based on the topological correlation matrix. Then, power flow calculations are performed on the initial topological state of the snowflake-shaped distribution network, and the reconfiguration period is determined by dividing the time periods using the Fisher optimal partitioning method. With the objectives of minimizing load balance and active power loss, a multi-period load spatial transfer model for the snowflake-shaped distribution network, considering N-1 safety constraints and switch combination reconfiguration, is established. Based on convex relaxation theory, the original non-convex model is transformed into a mixed-integer second-order cone programming model, which is then solved using a mature commercial solver. This invention effectively utilizes the load spatial transfer capacity of feeder clusters to balance lightly and heavily loaded transformers and line loads, providing a greater safety margin for grid operation. Snowflake-shaped distribution networks, which aggregate multiple feeders into clusters, can fully utilize the complementary differences in load types within different feeders, providing a support platform for load transfer over a larger spatial range. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the snowflake-shaped power distribution network structure of station 3 and line 6 in embodiment 3 of the present invention; Figure 2 This is a flowchart of the snowflake-shaped distribution network load spatial transfer method that takes into account N-1 security and network reconfiguration according to the present invention; Figure 3 This invention presents typical daily photovoltaic (PV) load curves for various load types. Figure 4 This is a schematic diagram of the load rate of each feeder in the snowflake distribution network before and after the load spatial transfer of the present invention. Detailed Implementation

[0017] The present invention will be further described in detail below with reference to the accompanying drawings.

[0018] Snowflake-shaped distribution network load spatial transfer optimization method considering N-1 safety and reconfiguration, such as Figure 2 As shown, it includes the following steps: Step 1: Obtain basic data such as snowflake-shaped distribution network line parameters, load power, and distributed photovoltaic output for the area to be studied.

[0019] Step 2: Based on Step 1 and the topological correlation matrix, express the load transfer capacity of the snowflake-shaped distribution network.

[0020] Step 2.1: Establish a feeder correlation matrix based on the interconnection relationships between feeders in a snowflake-shaped distribution network.

[0021] (1) Assume the distribution network has a total of n The substations are numbered 1, 2, ... n The number of feeders corresponding to each substation is N 1, N 2, ..., Nn There are a total of N One feeder, i.e. N=N 1+ N 2+…+ Nn In a snowflake-shaped power distribution network, n =3 or 4, N 1= N 2 = ... Nn= 2, N= 6 or 8. Divide the feeder association matrix into blocks according to the relationship between the substation and the feeder.

[0022] matrix F In F i,j Indicates feeder F i With feeder F j The relationships between them contain both 0 and 1 elements.F i,j A value of 1 indicates a feeder. F i With feeder F j They are related; there is a connection between them. F i,j A value of 0 indicates that the feeder... F i With feeder F j They are not related; there is no connection between them. Define a matrix. F The elements on the main diagonal are 0, that is... F i,i =0.

[0023] Step 2.2: Establish the intra-station interconnection transfer matrix and inter-station interconnection transfer matrix according to the different feeder correlation matrix and feeder load transfer methods. Construct the intra-station interconnection transfer matrix. A for: (2) Inter-station communication and transfer matrix B for: (3) Step 2.3: Establish the main transformer margin matrix and the feeder margin matrix based on the main transformer capacity constraints and the feeder capacity constraints.

[0024] Let the principal variable capacity matrix be... (4) Among them, matrix R The i element R i Indicates the main variable R i The capacity to accept loads.

[0025] Let the feeder capacity matrix be... (5) Among them, matrix L The i element L i Indicates feeder L i The capacity to accept loads.

[0026] Let the main transformer load factor matrix be... (6) Among them, matrix RM The i element RMi Indicates the main variable R i The load rate.

[0027] Let the feeder load factor matrix be... (7) Among them, matrix LM The i element LM i Indicates feeder L i The load rate.

[0028] The principal variable margin matrix is ​​then... (8) Among them, matrix The i element Indicates the main variable R i The margin, i.e., the main variable R i The maximum load that can still be accepted.

[0029] Similarly, the feeder margin matrix is (9) Among them, matrix The i element Indicates feeder L i The margin, i.e., the feeder L i The maximum load that can still be accepted.

[0030] Step 2.4: Calculate the maximum transfer capacity of each feeder using the interconnection transfer matrix and margin matrix, then compare it with the load that needs to be transferred to determine whether the N-1 criterion is met, and calculate the transfer margin.

[0031] Internal supply transfer capability matrix TAM and TBM for (10) (11) The maximum transfer capacity matrix is (12) The supply margin matrix is (13) Supply margin matrix elements in feederL i The remaining transfer margin after load transfer. If This indicates the feeder L i The maximum power transfer capacity during a fault is greater than or equal to the load it carries, therefore the feeder... L i satisfy N -1 is the safety criterion; a positive absolute value indicates a higher safety margin. Conversely, if... Then the feeder L i Not satisfied N -1 is the safety criterion. The larger the absolute value of the supply margin, the less safe the system is.

[0032] Step 3: Perform power flow calculation on the initial state of the snowflake-shaped distribution network, and dynamically reconstruct the time period division of the load rate data of each feeder based on Fisher's optimal partitioning method.

[0033] Based on the initial network topology of the snowflake-shaped distribution network, time-series power flow calculations are performed to obtain the branch power flow direction and the load rate of each feeder. Fisher's optimal segmentation method is used to divide the initial load rate data of each feeder into time periods to obtain the optimal segment for dynamic reconfiguration.

[0034] Step 4: With the goal of reducing active power loss and load balance, establish a snowflake-shaped distribution network load spatial transfer model that takes into account N-1 security and network reconfiguration based on the results of Step 3, and construct constraints based on transfer capacity.

[0035] The objective function of the snowflake-shaped distribution network load spatial transfer optimization model considering N-1 security and network reconfiguration is: (14) (15) (16) in, f It is the comprehensive objective function. f 1 represents the system's active power loss. f 2 is load balancing. and These represent the weighting coefficients for power loss and load balancing, respectively. . Represents the set of branches; This represents the set of feeders downstream of the main transformer (the set of branches directly connected to the main transformer). T This indicates the total number of time periods for calculation; in this invention, 24 hours is used. The calculation time interval is set to 1 hour. Indicates connection i Ring mesh box and j Branch of ring network box No. ij The resistance value; Indicates a branch ij exist t The current amplitude at a given time. Indicates a branch ij exist t Transmission capacity at any given moment Indicates a branch ij Maximum transmission capacity.

[0036] The constraints include N-1 safety margin for power transfer, power flow equation constraints, basic safe operation constraints, power balance constraints, network topology reconfiguration constraints, and switching action constraints.

[0037] N-1 Safety Transfer Margin Constraint (17) Power flow equation constraints (18) (19) (20) (twenty one) in, , These respectively represent the snowflake net with j The No. 1 ring network box is a combination of the end ring network box and the beginning ring network box; , They represent t From time to time j The flow direction of the No. 1 ring mesh cage k The active and reactive power of the ring network box; , They represent t time j Active and reactive power injection of ring network box No. 1; Indicates connection i Ring mesh box and j The wiring of the No. 1 ring network box ij The magnitude of reactance; 0-1 variable This indicates the open / closed state of the switch on the branch line; a value of 1 indicates... t Time Branch ij The switch on is in the ON state; a value of 0 indicates that the switch is ON. t Time Branch ij The switch on is in the on position.

[0038] Basic safety operation constraints (twenty two) (twenty three) (twenty four) Equation (22) represents the power constraint at the feeder outlet. and express t Time of the first i The active and reactive power at the feeder outlet, and They represent t Time of the first i The upper and lower limits of the active power output of each feeder; and They represent t Time of the first i The upper and lower limits of reactive power at the feeder outlet; Equation (23) represents the set of transformer nodes in a snowflake-shaped distribution network; Equation (24) represents the branch capacity constraint; Equation (25) represents the ring network box node voltage constraint.

[0039] Power balance constraints (25) in, , They represent t time j The active and reactive power outputs of the distributed power sources connected to the ring network box; , They represent t time j The active and reactive power of the load connected to the ring network box.

[0040] Network topology reconfiguration constraints (26) (27) (28) (29) In equation (26), and Represent the number of nodes and transformers in the network, respectively; 0-1 variables. express t Time-based ring network box node i and ring network box nodes j Relationship, A value of 1 indicates a ring network box node. j It is a ring network box node i The parent node, if 0, indicates a ring network box node.j Not a ring network box node i The parent node. Equation (28) indicates that the transformer source node in a snowflake-type distribution network has no parent node, and the transformer source node must be the parent node of other nodes. Equation (29) indicates that all nodes except the transformer source node can only have one parent node. Network topology constraints ensure that the system operates radially during the reconfiguration process.

[0041] Switch action constraints (30) (31) (32) Equation (30) represents the switch position change information. and All are 0-1 variables, when When, it means t Time Branch ij The switch on the top changes from open to closed. When, it means t Time Branch ij The switch on the device changes from closed to open; equation (31) is used to limit the time. t The switch can only change position once; Equation (32) constrains the number of switch actions within the scheduling cycle. This indicates the maximum number of times the switch can be activated. During dynamic reconfiguration, it is necessary to avoid frequent changes in the switch state as much as possible to ensure the lifespan of the switch and reduce switching costs.

[0042] The original model is transformed into a mixed-integer second-order cone model based on cone relaxation techniques and the big M method.

[0043] Introducing auxiliary variables as shown in equation (33) Load balancing f The expression for 2 is transformed into a linear objective function as shown in equation (34). Equation (33) is then relaxed to obtain the second-order cone constraint as shown in equation (35).

[0044] (33) (34) (35) Employing second-order cone relaxation techniques combined with large M The method transforms the power flow equations into those containing voltage square terms. and the square term of the current The mixed integer second-order cone programming model is shown in equations (36)-(39).

[0045] (36) (37) (38) (39) In equations (36) and (38), , and It is a sufficiently large positive number.

[0046] Similarly, the constraint condition (21) is relaxed to the rotational cone constraint shown in the following equation (40): (40) For the active power loss objective function f 1. Perform linearization processing, as shown in equation (41).

[0047] (41) Step 5: Transform the snowflake-shaped distribution network load spatial transfer model into a mixed integer second-order cone programming model; Step 6: Use a mature commercial solver such as CPLEX to solve the transformed second-order cone programming model; Step 7: Output relevant results, including network reconfiguration scheme, feeder load rate, security indicators, etc.

[0048] Based on the above-mentioned snowflake-shaped distribution network load spatial transfer optimization method that takes into account N-1 safety and reconfiguration, the effectiveness of the present invention is verified by conducting tests.

[0049] Step 1: Basic Data Acquisition. Acquire basic data such as snowflake-shaped distribution network line parameters, load power, and distributed photovoltaic output for the area to be studied.

[0050] The example uses a three-station, six-wire snowflake-shaped distribution network, such as... Figure 1 As shown. Substation A supplies residential load, i.e., the load from ring network node 7 to ring network node 22 is residential load. Substation B supplies commercial load, i.e., the load from ring network node 23 to ring network node 36 is commercial load. Substation C supplies office load, i.e., the load from ring network node 37 to ring network node 49 is office load. Branch lines B44, B45, and B46 represent intra-station connections, and branch lines B47, B48, and B49 represent inter-station connections. The main transformer parameters are shown in Table 1, the feeder capacity parameters are shown in Table 2, and the distributed photovoltaic parameters are shown in Table 3. Typical daily curves for each type of load and the distributed photovoltaic output curves are shown in... Figure 3 As shown. This invention sets the maximum number of times a switch can be activated within a day. The weight coefficient is 4, and the weight coefficient is obtained using the analytic hierarchy process (AHP). =0.846, =0.154.

[0051] Table 1 Main Transformer Parameters

[0052] Table 2 Feeder Capacity Parameters

[0053] Table 3 Distributed Photovoltaic Parameters

[0054] Step 2: Analytical Expression of Load Transfer Capacity. The load transfer capacity analysis model for a snowflake-shaped distribution network is expressed analytically based on the correlation matrix. A feeder correlation matrix is ​​established according to the interconnection relationships between feeders in the snowflake-shaped distribution network; intra-station interconnection transfer matrices and inter-station interconnection transfer matrices are established according to different feeder load transfer methods; a main transformer margin matrix and a feeder margin matrix are established based on the main transformer capacity constraints and feeder capacity constraints; the maximum transfer capacity of each feeder is calculated using the interconnection transfer matrix and margin matrix, and then compared with the load to be transferred to determine whether the N-1 criterion is met, and the transfer margin is calculated.

[0055] Its feeder correlation matrix F for: (42) Internal contact and transfer matrix A As shown in equation (43), the inter-station communication transfer matrix B As shown in equation (43).

[0056] (43) (44) Based on the connection transfer matrix and the margin matrix, the transfer capacity is calculated, and the station transfer capacity matrix is ​​as follows: (45) The inter-station transfer capacity matrix is ​​as follows: (46) The maximum transfer capacity matrix is: (47) The supply margin matrix is ​​as follows: (48) Step 3: Reconstruct the optimal time period segmentation. Based on the initial network topology of the snowflake-shaped distribution network, time-series power flow calculations are performed to obtain the branch power flow direction and the load rate of each feeder. The Fisher optimal segmentation method is used to divide the initial load rate data of each feeder into time periods, obtaining the optimal segments for dynamic reconstruction, which are the following three time periods: 1:00-8:00, 9:00-18:00, and 19:00-24:00.

[0057] Step 4: Model Construction. Establish a snowflake-shaped distribution network load spatial transfer optimization model that takes into account N-1 security and network reconfiguration, with the goal of reducing system balance and active power loss. In addition to satisfying general power flow constraints, operation constraints, and topology reconfiguration constraints, it is also necessary to satisfy the N-1 security transfer margin constraint.

[0058] The objective function expression is: (49) (50) (51) in, f It is the comprehensive objective function. f 1 represents the system's active power loss. f 2 is load balancing. and These represent the weighting coefficients for power loss and load balancing, respectively. . Represents the set of branches; This represents the set of feeders downstream of the main transformer (the set of branches directly connected to the main transformer). T This represents the total number of time periods for calculation; in this invention, it is taken as 24 hours. The calculation time interval is set to 1 hour. Indicates connection i Ring mesh box and j Branch of ring network box No. ij The resistance value; Indicates a branch ij exist t The current amplitude at a given time. Indicates a branch ij exist t Transmission capacity at any given moment Indicates a branch ij Maximum transmission capacity.

[0059] The constraints are as follows: N-1 Safety Transfer Margin Constraint (52) Power flow equation constraints (53) (54) (55) (56) in, , These respectively represent the snowflake net with j The No. 1 ring network box is a combination of the end ring network box and the beginning ring network box; , They represent t From time to time j The flow direction of the No. 1 ring mesh cage k The active and reactive power of the ring network box; , They represent t time j Active and reactive power injection of ring network box No. 1; Indicates connection i Ring mesh box and j The wiring of the No. 1 ring network box ij The magnitude of reactance; 0-1 variable This indicates the open / closed state of the switch on the branch line; a value of 1 indicates... t Time Branch ij The switch on is in the ON state; a value of 0 indicates that the switch is ON. t Time Branch ij The switch on is in the on position.

[0060] Basic safety operation constraints (57) (58) (59) Equation (57) represents the power constraint at the feeder outlet. and express t Time of the first i The active and reactive power at the feeder outlet, and They represent t Time of the first i The upper and lower limits of the active power output of each feeder; and They represent t Time of the first i The upper and lower limits of reactive power at the feeder outlet; Equation (58) represents the set of transformer nodes in a snowflake-shaped distribution network; Equation (59) represents the branch capacity constraint; Equation (59) represents the ring network box node voltage constraint.

[0061] Power balance constraints (60) in, , They represent t time j The active and reactive power outputs of the distributed power sources connected to the ring network box; , They represent t time j The active and reactive power of the load connected to the ring network box.

[0062] Network topology reconfiguration constraints (61) (62) (63) (64) In equation (61), and Represent the number of nodes and transformers in the network, respectively; 0-1 variables. express t Time-based ring network box node i and ring network box nodes j Relationship, A value of 1 indicates a ring network box node. j It is a ring network box node i The parent node, if 0, indicates a ring network box node. j Not a ring network box node i The parent node. Equation (63) indicates that in a snowflake-type distribution network, the transformer source node has no parent node, and the transformer source node must be the parent node of other nodes. Equation (64) indicates that all nodes except the transformer source node can only have one parent node. Network topology constraints ensure that the system operates radially during the reconfiguration process.

[0063] Switch action constraints (65) (66) (67) Equation (65) represents the switch position change information. and All are 0-1 variables, when When, it meanst Time Branch ij The switch on the top changes from open to closed. When, it means t Time Branch ij The switch on the device changes from closed to open; equation (66) is used to limit the time. t The switch can only change position once; Equation (67) constrains the number of switch actions within the scheduling cycle. This indicates the maximum number of times the switch can be activated. During dynamic reconfiguration, it is necessary to avoid frequent changes in the switch state as much as possible to ensure the lifespan of the switch and reduce switching costs.

[0064] Step 5: Model Transformation. Based on cone relaxation techniques and the Big M method, the original model is transformed into a mixed-integer second-order cone model.

[0065] Introducing auxiliary variables as shown in equation (68) Load balancing f The expression for 2 is transformed into a linear objective function as shown in equation (69). Equation (68) is then relaxed to obtain the second-order cone constraint as shown in equation (70).

[0066] (68) (69) (70) Employing second-order cone relaxation techniques combined with large M The method transforms the power flow equations into those containing voltage square terms. and the square term of the current The mixed integer second-order cone programming model is shown in equations (71)-(74).

[0067] (71) (72) (73) (74) In equations (71) and (73), , and It is a sufficiently large positive number.

[0068] Similarly, the constraint condition (56) is relaxed to the rotating cone constraint shown in the following equation (75): (75) For the active power loss objective function f 1. Perform linearization processing, as shown in equation (76).

[0069] (76) Step Six: Solve the Model. Use a mature commercial solver such as CPLEX to solve the transformed second-order cone programming model.

[0070] For this embodiment, the following two scenarios are selected for comparative analysis: Scenario 1 is the initial scenario; Scenario 2 is the scenario after optimization based on the method provided by this invention.

[0071] Step 7: Output Results. Output relevant results, such as dynamic reconfiguration scheme, load space transfer strategy, feeder load rate, and safety indicators.

[0072] Table 4 shows the optimization results of active power loss and load balance in snowflake-shaped distribution networks under two scenarios. Compared with scenario 1, both active power loss and load balance are reduced in scenario 2.

[0073] Table 4 Results of the Objective Function

[0074] The reconfiguration scheme for Scenario 2 is shown in Table 5. Heavy-load feeder loads are transferred to adjacent feeders through intra-station or inter-station connections, thereby achieving system load balancing.

[0075] Table 5 Comparison of Branch Disconnection Schemes

[0076] Load rates in two scenarios Figure 4 As shown in the diagram. Scenario 1 represents the initial state, with significant differences in load rates among the feeders, and a clear coexistence of lightly and heavily loaded lines. From 1:00 to 8:00, feeder F1 had the highest load rate, reaching 69.23% at 8:00, while the load rates of the other feeders were relatively low. From 9:00 to 18:00, the load rates of feeders F3, F5, and F6 gradually increased, while the load rate of feeder F1 gradually decreased. Feeder F3 reached its peak load rate of approximately 96.81% at 12:00 noon, indicating severe overload. From 19:00 to 24:00, the load rate of feeder F3 gradually decreased, while the load rate of feeder F1 showed a trend of first increasing and then decreasing, reaching 77.78% at 21:00. In Scenario 2, the overload phenomenon of feeders was alleviated. At 21:00, the load rate of feeder F1 decreased from 77.78% to 33.02%, with a peak load of 57.72%; the peak load of feeder F3 decreased from 96.81% to 57.68%; and the peak load of feeder F6 decreased from 77.29% to 55.71%. Based on the advantages of multiple interconnections within and between stations in the snowflake-shaped distribution network feeder cluster, load was flexibly transferred spatially, shifting the load from the feeder side with a high load rate to the feeder side with a low load rate, thereby eliminating branch overload and ensuring a balanced load across the entire network.

[0077] The safety indicators are compared in Table 6. Safety margin matrix in Scenario 1. The presence of 6 negative numbers indicates that the system fails to meet the feeder N-1 safety criterion at 6 specific moments. In Scenario 2, the system consistently meets the feeder N-1 safety criterion in terms of the safety transfer margin matrix. Table 7 shows a comparison of the safety transfer margin values ​​for feeder F4 during the 9:00-18:00 period. Compared to Scenario 1, the safety transfer margin value for feeder F4 in Scenario 2 is significantly improved, all being greater than 0. By dynamically reconfiguring the network topology, the spatial dimension of feeder load can be transferred, thereby enhancing system safety.

[0078] Table 6 Comparison of Safety Indicators

[0079] Table 7 Safety Transfer Margin for Feeder F4 / MVA

[0080] It should be emphasized that the embodiments described in this invention are illustrative and not limiting. Therefore, this invention includes, but is not limited to, the embodiments described in the specific implementation. Any other implementation methods derived by those skilled in the art based on the technical solutions of this invention also fall within the scope of protection of this invention.

Claims

1. A snowflake-shaped distribution network load spatial transfer optimization method considering N-1 safety and reconfiguration, characterized in that: Includes the following steps: Step 1: Obtain snowflake-shaped distribution network line parameters, load power, and distributed photovoltaic power output data for the area to be studied; Step 2: Based on Step 1 and using the topological correlation matrix, analytically express the load transfer capacity of the snowflake-shaped distribution network; Step 3: Perform power flow calculation on the initial state of the snowflake-shaped distribution network, and dynamically reconstruct the time period division of the load rate data of each feeder based on Fisher's optimal partitioning method; Step 4: With the goal of reducing active power loss and load balance, establish a snowflake-shaped distribution network load spatial transfer model that takes into account N-1 security and network reconfiguration based on the results of Step 3, and construct constraints based on transfer capacity. Step 5: Transform the snowflake-shaped distribution network load spatial transfer model into a mixed integer second-order cone programming model; Step 6: Solve the transformed mixed-integer second-order cone programming model; Step 7: Output the relevant results.

2. The snowflake-shaped distribution network load spatial transfer optimization method considering N-1 security and reconfiguration as described in claim 1, characterized in that: Step 2 includes the following steps: Step 2.1: Establish a feeder correlation matrix based on the interconnection relationships between feeders in a snowflake-shaped distribution network; Step 2.2: Establish intra-station interconnection and transfer matrices and inter-station interconnection and transfer matrices respectively according to the feeder correlation matrix and the different feeder load transfer methods; Step 2.3: Establish the main transformer margin matrix and the feeder margin matrix based on the main transformer capacity constraints and the feeder capacity constraints; Step 2.4: Calculate the maximum transfer capacity of each feeder using the interconnection transfer matrix and margin matrix, then compare it with the load that needs to be transferred to determine whether the N-1 criterion is met, and calculate the transfer margin.

3. The snowflake-shaped distribution network load spatial transfer optimization method considering N-1 security and reconfiguration as described in claim 2, characterized in that: The specific implementation method of step 2.1 is as follows: ； Among them, the power grid has a total of n The substations are numbered 1, 2, ... n The number of feeders corresponding to each substation is N 1, N 2, ..., Nn There are a total of N One feeder line, N=N 1+ N 2+…+ Nn ,matrix F In F i,j Indicates feeder F i With feeder F j The relationships between them.

4. The snowflake-shaped distribution network load spatial transfer optimization method considering N-1 security and reconfiguration as described in claim 3, characterized in that: In step 2.2, an internal communication and supply matrix is ​​constructed. A for: ； Inter-station communication and transfer matrix B for: 。 5. The snowflake-shaped distribution network load spatial transfer optimization method considering N-1 security and reconfiguration as described in claim 4, characterized in that: The specific implementation method of step 2.3 is as follows: Calculate the principal variable capacity matrix: ； Among them, matrix R The i element R i Indicates the main variable R i The capacity to accommodate loads is calculated, and the feeder capacity matrix is ​​set up as follows: ； Among them, matrix L The i element L i Indicates feeder L i The capacity to accept load is calculated, and the main transformer load factor matrix is ​​set as follows: ； Among them, matrix RM The i element RM i Indicates the main variable R i The load factor is calculated, and the feeder load factor matrix is ​​set up as follows: ； Among them, matrix LM The i element LM i Indicates feeder L i Based on the load factor, calculate the main transformer margin matrix: ； Among them, matrix The i element Indicates the main variable R i The margin, i.e., the main variable R i Calculate the feeder margin matrix based on the maximum load that can still be accommodated: ； Among them, matrix The i element Indicates feeder L i The margin, i.e., the feeder L i The maximum load that can still be accepted.

6. The snowflake-shaped distribution network load spatial transfer optimization method considering N-1 security and reconfiguration as described in claim 5, characterized in that: The specific implementation method of step 2.4 is as follows: calculate the station's internal transfer capacity matrix. TAM and TBM : ； ； Calculate the maximum transfer capacity matrix: ； Calculate the supply margin matrix: ； Among them, the supply margin matrix elements in feeder L i The remaining transfer margin after load transfer, if This indicates the feeder L i The maximum power transfer capacity during a fault is greater than or equal to the load it carries, therefore the feeder... L i satisfy N -1 Safety Rule: A positive absolute value for the supply margin indicates better safety; conversely, a negative absolute value indicates poorer safety. Then the feeder L i Not satisfied N -1 is the safety criterion. The larger the absolute value of the supply margin, the less safe the system is.

7. The snowflake-shaped distribution network load spatial transfer optimization method considering N-1 security and reconfiguration as described in claim 1, characterized in that: The specific implementation method of step 3 is as follows: based on the initial network topology of the snowflake-shaped distribution network, perform time-series power flow calculation to obtain the branch power flow direction and the load rate of each feeder, and use Fisher's optimal segmentation method to divide the initial load rate data of each feeder into time periods to obtain the optimal segment for dynamic reconstruction.

8. The snowflake-shaped distribution network load spatial transfer optimization method considering N-1 security and reconfiguration as described in claim 1, characterized in that: The objective function of the snowflake-shaped distribution network load spatial transfer optimization model that takes into account N-1 security and network reconfiguration in step 4 is: ； ； ； in, f It is the comprehensive objective function. f 1 represents the system's active power loss. f 2 is load balancing. and These represent the weighting coefficients for power loss and load balancing, respectively. , Represents the set of branches; This represents the set of feeders downstream of the main transformer; T This indicates the total number of time periods. The time interval for calculation, Indicates connection i Ring mesh box and j Branch of ring network box No. ij The resistance value; Indicates a branch ij exist t Current amplitude at time , Indicates a branch ij exist t Transmission capacity at any given moment Indicates a branch ij Maximum transmission capacity.

9. The snowflake-shaped distribution network load spatial transfer optimization method considering N-1 security and reconfiguration according to claim 9, characterized in that: The constraints of the snowflake-shaped distribution network load spatial transfer optimization model that takes into account N-1 security and network reconfiguration include N-1 security transfer margin constraints, power flow equation constraints, basic safe operation constraints, power balance constraints, network topology reconfiguration constraints, and switching action constraints.

Images