Urban power grid partition coordinated recovery method and system considering source-network-load interaction

Abstract

The invention relates to an urban power grid partition coordinated recovery method and system considering source-network-load interaction, and the method comprises the following steps: building a topological structure of an urban power distribution network, the topological structure representing a connection relation between a load node and a power node; based on the topological structure, dividing the urban power distribution network by adopting a shortest path algorithm to obtain a plurality of partitions; establishing a power distribution network recovery mathematical model considering partitions by taking the maximum cumulative weighted load recovery amount as a target; and based on the plurality of partitions, solving the power distribution network recovery mathematical model so as to recover each partition in parallel. Compared with the prior art, the method has the advantages that the recovery speed is high, and important loads can be recovered as much as possible.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid restoration, and in particular to a method and system for coordinated restoration of urban power grid zones taking into account source-grid-load interaction. Background Art

[0002] The safety of the power system must be ensured during the recovery process from a blackout. Since the power grid recovery process involves the switching operations of multiple devices, the system's active and reactive power demands change frequently, posing challenges to the stability and control of the system's voltage and frequency.

[0003] In terms of frequency constraints during the power system recovery process, the main task of frequency control during the recovery process is to offset the frequency changes caused by the gradual recovery of loads. The paper "Real-time optimized load recovery considering frequency constraints" analyzes the impact of the location and size of the connected load on the system frequency during the system recovery process, and points out that the static and dynamic characteristics of the load will affect the selection of the connected power source during the recovery process. The paper "Incorporating demand response with spinning reserve to realize an adaptive frequency restoration plan for system contingencies" studies the role of load demand response in the system frequency recovery process after the event. Different from using low-frequency load reduction as the last line of defense for system defense, the paper proposes a frequency recovery strategy in which load demand response and spinning reserve actively participate in frequency regulation in emergency events.

[0004] In terms of power supply constraints during the power system restoration process, the overvoltage of light-load lines in the initial stage of black start is the main influencing factor in the system restoration process. The document "Milanicz. Energizing high and extra-high voltage lines during restoration" proposes an approximate assessment method for transient overvoltage, which can be used to guide the optimization of line and load access during the grid restoration process. The document "A method for sustained overvoltage control during power system restoration" proposes a grid control strategy in the initial stage of restoration based on sensitivity analysis, and constructs voltage constraints in the initial stage of grid restoration by coordinating generators, transformer taps, reactors and loads. The document "Microgrids for service restoration to critical load in a resilient distribution system" takes into account the influence of controllable distributed generation (DG) on frequency and transient voltage safety constraints during the restoration process, and simplifies the dynamic model of DG into a synchronous generation model for modeling.

[0005] At present, the existing research results on black start partitioning methods are mainly divided into three categories: agglomerative algorithms, divisive algorithms, and methods based on optimization modeling. The Girvan-Newman (GM) divisive algorithm is a widely used divisive algorithm. This algorithm can make up for some of the shortcomings of the above-mentioned agglomerative methods, but it also has defects such as the inability to determine the number of community structures in advance and the large amount of calculation. In addition, for partition recovery, there is still room for improvement in recovery speed and recovery quantity. Summary of the invention

[0006] The purpose of the present invention is to provide a method and system for coordinated restoration of urban power grid zones taking into account the interaction between sources, grids and loads, which can improve the restoration speed and the restoration amount at the same time.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] A method for coordinated restoration of urban power grid zones considering source-grid-load interaction includes the following steps:

[0009] Establishing a topological structure of a city distribution network, wherein the topological structure represents a connection relationship between load nodes and power supply nodes;

[0010] Based on the topological structure, the urban distribution network is divided into multiple partitions using the shortest path algorithm;

[0011] A distribution network restoration mathematical model considering partitions is established with the goal of maximizing the cumulative weighted load restoration.

[0012] Based on the multiple partitions, the distribution network restoration mathematical model is solved to restore each partition in parallel.

[0013] Furthermore, the power supply node includes a black start power supply node and a non-black start power supply node.

[0014] Furthermore, the step of obtaining multiple partitions includes:

[0015] Modeling the topology as an undirected weighted graph;

[0016] Determining the number of partitions and black start power supply nodes in each partition based on the undirected weighted graph;

[0017] Based on the undirected weighted graph, the shortest path algorithm is used to solve the shortest distance from each node to be restored to the black start power supply node, so as to divide the node to be restored into the partition where the start power supply node with the shortest distance is located, and obtain an initial partition result;

[0018] Based on the initial partitioning result, a power balance check is performed within the partition to obtain a final partitioning result.

[0019] Furthermore, the weight of the edge in the undirected weighted graph is adjusted according to the electrical distance of the line, and the expression of the weight of the edge is:

[0020] ω ij =Z line l ij

[0021] In the formula, ω ij is the weight of edge (i, j), Z line Indicates the impedance value corresponding to the line per unit distance, l ij Indicates the length of line (i, j).

[0022] Furthermore, the expression for determining the number of partitions is:

[0023]

[0024] m=min{N black , N load}

[0025] In the formula, is the start-stop state of the dispatchable DG without black start function, μ g，t is the start and stop status of the unit, m is the maximum number of partitions, N black 、N load They represent the number of black start nodes and load nodes in the system respectively.

[0026] Furthermore, the step of performing intra-partition power balance verification includes:

[0027] Determine whether the load power connected to each initial partition exceeds the total power that each initial partition can provide.

[0028] If so, delete the unimportant load nodes on the edge of each initial partition to meet the power requirements.

[0029] If not, do nothing.

[0030] Furthermore, the constraint expression of the power balance check is:

[0031]

[0032] In the formula, a ij Indicates whether node j is in partition i, n is the total number of nodes, is the load input state of node j, is the active load connected at node j, represents the active power of the power source connected to node i, Indicates whether the power supply of node j is connected, and m is the number of partitions.

[0033] Furthermore, the distribution network restoration mathematical model includes an objective function and constraints, and the objective function is:

[0034]

[0035] In the formula, f is the objective function, n is the number of nodes, m is the number of partitions, T is the number of time periods, ΔT is the time taken for each time step, and a is the ij Indicates whether node j is in partition i, represents the weight of the load at node j, is the active power demand of the load at node j at time t, and L is the load set;

[0036] The constraints include:

[0037] (1) Power flow equation constraints

[0038]

[0039] -M·μ jk，t ≤P jk，t ≤M·μ jk，t , (j, k)∈A

[0040] -M·μ jk，t ≤Q jk，t ≤M·μ jk，t (j,k)∈4

[0041] V j，t -V k，t ≤2(r jk ·P jk，t +x jk Q jk，t )+M(1-μ jk，t ), j, k = 1, 2, ..., n

[0042] V j，t -V k，t ≥2(r jk ·P jk，t +x jk Q jk，t )-M(1-μ jk，t ), j, k = 1, 2, ..., n

[0043] In the formula, is the reactive power demand of the load at node j at time t, and are the active and reactive outputs of the distributed generation at node j at time t, M is a constant, μ jk，t is a binary variable representing the recovery state of line (j, k) in time period t, μ jk，t =1 indicates that line (j, k) is restored in time period t, P jk，t , Q jk，t 、r jk and x jk are respectively the active power, reactive power, resistance and reactance on line (j, k) in time period t, V j，t is the square of the voltage amplitude of the first node j of line (j, k) in time period t, A is the set of lines, V k，t is the square of the voltage amplitude at the end node k of line (j, k) in time period t;

[0044] (2) Distributed power output constraints

[0045]

[0046] In the formula, represents the startup state of the kth distributed generation in time period t, and They represent the lower and upper limits of the active output of the kth distributed generation, and They represent the lower and upper limits of reactive power output of the kth distributed generation, respectively. G is the set of distributed generation. is the active power of the DG connected to node k in partition s at time t, is the reactive power of the DG connected to node k in partition s at time t;

[0047] (3) Load constraints

[0048]

[0049] In the formula, Indicates whether the jth load is put into operation in period t. If yes, it takes 1, otherwise it takes 0. j，max , Q j，max Respectively represent the active and reactive demands of the load;

[0050] (4) Distributed power supply ramp constraints

[0051]

[0052] In the formula, is the maximum ramp rate of the kth unit, is the active power of the DG connected to node k at time t;

[0053] (5) Node voltage constraints

[0054]

[0055] Where V i min and V i max Respectively represent the maximum and minimum node voltage, μ i，t Indicates whether node i is engaged in time period t, if yes, it takes 1, otherwise it takes 0;

[0056] (6) Line power constraints

[0057]

[0058] In the formula, is the maximum transmission capacity of branch (j, k), P ij , t is the active transmission capacity of branch (i, j) at time t, W ij , t is the reactive transmission capacity of branch (i, j) at time t, S jk is the radius of the polygon involved in the polygon-based linearization method to approximate the quadratic term, n is the number of sides of the polygon used for linearization;

[0059] (7) Restoring order constraints

[0060] 1) A non-black start power node can only be restored after the node it is on is restored:

[0061]

[0062] 2) The line can be powered only when both end nodes are powered:

[0063] μ jk，t ≤μ j，t , (j, k)∈A

[0064] μ jk，t ≤μ k，t , (j, k)∈A

[0065] 3) If the node is connected to a black start power node or a substation node, it will be powered on:

[0066]

[0067] 4) The load can only be powered when connected to a powered node:

[0068]

[0069] 5) Constraints on the number of lines to be closed:

[0070]

[0071] 6) Additional constraints to ensure that distribution network components will not be cut off after power is applied:

[0072]

[0073] In the formula, μ j , t is the input state of node j at time t, G N is the set of units without black start capability, G Y is the black start power node, N lim To allow the number of lines to be put into operation simultaneously;

[0074] (8) Network topology constraints

[0075]

[0076] Where N D is the set of all nodes, μ ij , t is the input state of branch (i, j) at time t;

[0077] (9) Initial condition constraints

[0078] All lines are opened in the first step:

[0079] μ ij， 1=0,(i,j)∈A

[0080] The black start power node is powered on in the first step:

[0081]

[0082] In the formula, μ ij , 1 is the initial state of branch (i, j), is the input state of the black start power supply on node i at the initial moment.

[0083] Furthermore, ILOG CPLEX is used for solving.

[0084] The present invention also provides a city power grid zone coordinated restoration system considering source-grid-load interaction, comprising:

[0085] Topology building module: used to establish the topology of the urban distribution network, wherein the topology represents the connection relationship between the load nodes and the power supply nodes;

[0086] Partition division module: used for dividing the urban distribution network based on the topological structure by using the shortest path algorithm to obtain multiple partitions;

[0087] Model building module: used to establish a distribution network restoration mathematical model taking into account partitions with the goal of maximizing the cumulative weighted load restoration amount;

[0088] Restoration module: used for solving the distribution network restoration mathematical model based on the multiple partitions to restore each partition in parallel.

[0089] Compared with the prior art, the present invention has the following beneficial effects:

[0090] (1) The present invention uses the shortest path algorithm to partition the topological structure of the distribution network. By solving the shortest path, the recovery speed of each partition can be improved. At the same time, regional parallel recovery is performed with the maximum cumulative weighted load recovery amount as the goal, achieving the maximum recovery amount at a higher recovery speed.

[0091] (2) With regard to the black start zoning method, the present invention converts the distribution network into an undirected weighted graph, and confirms in advance the number of partitions and the black start power supply that needs to be allocated to each partition. By solving the shortest distance from the node to be restored to the black start power supply node, accurate regional division for parallel restoration of the distribution network is achieved.

[0092] (3) The distribution network restoration mathematical model of the present invention not only takes into account the load restoration amount, but also takes into account the influence of the load importance. By adding the load weight to the objective function to reflect the load importance, the important load can be restored as quickly and as much as possible during restoration. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0094] Figure 2 It is a two-step schematic diagram of power grid partition recovery according to the present invention.

[0095] Figure 3 It is the partition result diagram of the present invention;

[0096] Figure 4 Recover steps for each region of the present invention;

[0097] Figure 5 It is a network diagram of a 10kV distribution network node system of the present invention;

[0098] Figure 6 It is a schematic diagram of the node system area division of the present invention;

[0099] Figure 7 The partition one recovery effect of the present invention;

[0100] Figure 8 The second partition recovery effect of the present invention;

[0101] Fig. 9 The third partition recovery effect of the present invention;

[0102] Fig.10 The fourth partition recovery effect of the present invention;

[0103] Fig.11 This is the recovery effect of partition five of the present invention. DETAILED DESCRIPTION

[0104] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0105] Example 1

[0106] This embodiment provides a method for coordinated restoration of urban power grid zones taking into account the interaction between sources, grids and loads. Figure 1 As shown, the method comprises the following steps:

[0107] Step 1: Partitioning methods and principles.

[0108] This example adopts a partitioning method based on the shortest path algorithm. The specific steps are to first simplify the urban distribution network into an undirected weighted graph, and then solve the shortest distance from each node to be restored to the black start node based on the shortest path algorithm. After comparison, the node to be restored is divided into the partition where the black start power source with the shortest distance is located, thereby obtaining the initial partitioning result. Finally, based on the preliminary partitioning result, the power balance within the partition is verified, and the node loads at the edge and with low importance are eliminated to obtain the final partitioning result.

[0109] The following principles should be followed as much as possible during the partitioning process:

[0110] (1) Each partition contains at least one black start unit.

[0111] (2) The internal network should be as compact as possible and the electrical connections between subsystems should be as few as possible.

[0112] (3) The sizes of the partitions should be roughly the same.

[0113] (4) Whenever possible, use the black start power supply as the electrical center.

[0114] Step 2: Consider the mathematical model of distribution network restoration with partitions.

[0115] For urban distribution networks, a two-step strategy of partitioning and then restoring is adopted. Figure 2 The first step is to divide the distribution network into multiple partitions according to the shortest path algorithm, and the second step is to restore each partition in parallel with the goal of maximizing the cumulative weighted load restoration.

[0116] Before partitioning, the number of partitions and the black start units in each partition should be determined. First, according to the document "Two-step strategy for black start partitioning of power system after blackout", the maximum number of restored sub-areas m is determined, and its mathematical expression is:

[0117]

[0118] m=min{N black , N load}m (2)

[0119] Where: N black 、N load Represent the number of black start units and load nodes in the system respectively. Then, according to the general rule, it can be determined that m = N black .

[0120] The urban distribution network is usually a ring network structure, running radially. Its topology is a connected graph, in which each transmission line can be regarded as an edge of the connected graph. The weight of the edge is generally set according to the electrical distance of the line. The specific setting formula is as follows:

[0121] ω ij =Z line l jk (3)

[0122] Where: Z line Indicates the impedance value corresponding to the line per unit distance; l jk Indicates the length of line (j, k).

[0123] After abstracting the power network into a sparsely connected graph, its partitioning problem is similar to the graph partitioning problem. However, the difference from the general graph partitioning problem is that black start partitioning needs to be combined with the characteristics of the power system itself.

[0124] Here, the shortest path algorithm is used to determine the shortest path length from the source node to any other node, thereby realizing the regional division of parallel restoration of the distribution network.

[0125] After dividing each node into different subsystems, it is necessary to check the power balance constraints within the subsystem. If the connected load power exceeds the power that the unit can provide, delete the unimportant load nodes at the edge of the partition. The specific constraints are as follows:

[0126]

[0127] Where: a ij Indicates whether node j is in partition i; Indicates whether the power supply of node j is connected; Indicates the active power of the power source connected to node i.

[0128] After obtaining the final partitioning result, it is necessary to perform independent load recovery for each partition. The specific model is shown below.

[0129] 2.1 Objective Function

[0130] The goal of distribution network restoration is to restore important loads as quickly and as much as possible while ensuring system safety. The specific objective function is as follows:

[0131]

[0132] In the formula: m represents the number of partitions; n represents the number of nodes; T represents the number of time periods; ΔT represents the time interval; represents the weight of the load at node j; Represents the active load connected to node i.

[0133] 2.2 Constraints

[0134] (1) Power flow equation constraints

[0135] The basic power flow equation constraints need to be met during the operation of the power system. The simplified DistFlow power flow equation has been widely used in radial distribution networks. The assumption is that the nonlinear terms in the DistFlow equation are much smaller than the linear terms and can be ignored. The specific constraints are as follows:

[0136]

[0137] -M·μ jk，t ≤P jk，t≤M·μ jk，t (j,k)∈A (8)

[0138] -M·μ jk，t ≤Q jk，t ≤M·μ jk，t , (j, k)∈A (9)

[0139]

[0140] Where: and are the active output and reactive output of the distributed generation at node j at time t respectively; and are respectively the active power demand and reactive power demand of the load at node j at time t; P jk，t , Q jk，t 、r jk and x jk are the active power, reactive power, resistance and reactance on line (j, k) respectively; μ jk，t is a binary variable representing the recovery state of line (j, k) in time period t, μ jk，t =1 indicates that line (j, k) is restored in time period t; V j，t is the square of the node voltage amplitude, and M is a constant with a sufficiently large value.

[0141] Constraints (5) and (6) represent the active and reactive power balance of the power system, respectively; constraints (7) and (8) indicate that only energized lines can supply power. Constraints (9) and (10) define the voltage amplitude difference between the two end nodes of each energized line.

[0142] (2) Distributed power output constraints

[0143] The output of distributed power sources must meet certain upper and lower limit constraints. The specific constraints are as follows:

[0144]

[0145] Where: and They represent the lower and upper limits of the active output of the kth distributed generation respectively; and They represent the lower and upper limits of reactive power output of the kth distributed generation respectively; represents the startup state of the kth distributed power source; G is the set of distributed power sources.

[0146] Constraints (13) and (14) indicate that the output of distributed generation must satisfy certain formula conditions.

[0147] (3) Load constraints

[0148] The load demand of each node is considered to be fixed. When a system failure occurs and all loads are cut off, it takes a long time to restore the load. Which node's load should be restored first and how much load should be restored for each node need further optimization. During the load recovery process, the restored load cannot exceed the load demand of the node. The specific constraints are as follows:

[0149]

[0150]

[0151] Where: Indicates whether the jth load t period is put into use, if yes, it takes 1, otherwise it takes 0; P j，max , Q j，max Respectively represent the active and reactive power demands of the load.

[0152] Constraints (15) and (16) stipulate that the load restored in each time interval should not exceed the load demand.

[0153] (4) Distributed power supply ramp constraints

[0154] The output ramp of distributed power sources needs to meet certain constraints. The specific formula is as follows:

[0155]

[0156] Where: is the maximum ramp rate of the kth unit. Constraint (17) limits the ramp speed of distributed generation.

[0157] (5) Node voltage constraints

[0158] In order to ensure safe and reliable operation of the power grid during the recovery process, the node voltage needs to be kept within a certain range. The specific constraints are as follows:

[0159] μ i，t (V i min ) 2 ≤V i，t ≤μ i，t (V i max ) 2 , i∈N D (18)

[0160] Where: V i min and V i max Respectively represent the maximum and minimum node voltage. i，tIndicates whether node i is put into operation in time period t, if yes, it takes 1, otherwise it takes 0. Constraint (18) ensures that the voltage amplitude of each node should be kept within an acceptable range.

[0161] (6) Line power constraints

[0162] The active and reactive power flowing through the line cannot exceed a certain upper limit, otherwise the line may be overloaded and damaged. The specific constraints are as follows:

[0163]

[0164] Where: is the maximum transmission capacity of branch (j, k). Note that this formula is a quadratic term, so the polygon-based linearization method introduced in the literature is used to approximate the quadratic term. Polygon S jk The radius is chosen to be:

[0165]

[0166] Where n is the number of edges of the polygon used for linearization. The linear constraint when n=6 can be expressed as:

[0167]

[0168] (7) Restoring order constraints

[0169] 1) Non-black start power can only be restored after the node where it is located is restored

[0170]

[0171] Where: G N It is a collection of units that do not have black start capability.

[0172] 2) The line can only be powered when both ends of the node are powered

[0173] μ jk，t ≤μ j，t , (j, k)∈A (25)

[0174] μ jk，t ≤μ k，t , (j, k)∈A (26)

[0175] Where: A represents the set of lines.

[0176] 3) If the node is connected to a black start distributed generation or substation node, the node will be powered on

[0177]

[0178] Where: G YAssemble the units for black start.

[0179] 4) The load can only be powered when connected to a powered node

[0180]

[0181] Where: L is the load set.

[0182] 5) To ensure that the power grid can be restored safely, the number of lines closed at the same time needs to meet certain constraints

[0183]

[0184] Where: N lim To allow the number of lines to be put into operation simultaneously.

[0185] 6) Add some additional constraints to ensure that the power system components will not be cut off after power is turned on

[0186] μ jk，t -μ jk，t-1 ≥0, (j, k)∈A (30)

[0187]

[0188] (8) Network topology constraints

[0189] The distribution network generally adopts a ring network design and operates in a radial manner, so its topological structure during operation must meet the relevant constraints in graph theory. The specific formula is as follows:

[0190]

[0191] (9) Initial condition constraints

[0192] All lines are opened in the first step

[0193] μ ij，1 =0,(i,j)∈A (34)

[0194] Black start power supply is powered on in the first step

[0195]

[0196] Step 3: Node test case verification.

[0197] 3.1 Example Data

[0198] In this paper, distributed power sources are connected to nodes 15, 25, 32, 41, 47, 56, 61, and 65 in a certain node distribution system to verify the effectiveness of the proposed distribution network partition recovery strategy. An optimization model is established on the Matlab platform and ILOGCPLEX is called to solve it. The loads in the distribution system are classified according to their importance: the first-level load nodes are 6-9, 13-15, 20, 22, 33-35, 40-42, 48, 51, 52, 54, 55, 58, 62-64, and 68; the second-level load nodes are 10-12, 16-19, 31, 43-46, 49, 56, and 59; the rest are third-level loads, and the first-level load weight is set to 10, the second-level load weight is set to 1, and the third-level load weight is set to 0.1. Assume that node 1 is disconnected from the external network due to a fault, that is, the distribution network needs to rely on its own available distributed power sources to supply power to important loads as much as possible before the fault is eliminated and connected to the external network. Set 10 time steps, each time step is 5 minutes.

[0199] The parameters of the distributed power sources are shown in Table 1, and their status indicates the controllability of each distributed power source, where “1” indicates

[0200] "1 / 0" indicates a black start distributed power source that can be started at step t=1, and "1 / 0" indicates a non-black start distributed power source that can be started once a control command is issued. The voltage of the black start distributed power source is adjusted at 1.05 pu.

[0201] Table 1 Unit parameters

[0202]

[0203] The parameters related to the active power demand and reactive power demand of the load are shown in Table 2. The weight coefficients have been given above and will not be repeated here.

[0204] Table 2 Load parameters

[0205]

[0206] 3.2 Results Analysis

[0207] According to the partitioning method proposed in this paper, a partitioning scheme is obtained, such as Figure 3As shown. The scheme divides the distribution system of 69 nodes into four separately operated partitions, each partition includes a black start unit. Partition 1 consists of disconnected lines (5, 6) and lines (15, 16), including 1 black start unit DG1 and 3 units to be restored DG4, DG5, and DG6, with a total load to be restored of 2900kW. For partition 2, the lines (21, 22) are disconnected, including the black start unit DG2, and the total load to be restored is 500kW. Partition 3 is formed by disconnecting lines (30, 31), including a black start unit DG3, and the total load to be restored is 300kW. Partition 4 is formed by disconnecting lines (3, 59) and lines (68, 69), including black start unit DG7 and unit DG8 to be restored, and the total load to be restored is 1300kW. Some unimportant loads (such as L4, L5, L17, and L18) are not included in the partition because they are unimportant and far away from the black-start units, and other loads are restored first.

[0208] After the partitioning scheme is obtained, each partition is restored independently and in parallel. According to the parallel recovery optimization model established in this paper, the recovery results of each partition are as follows: Figure 4 shown.

[0209] From the calculation results, it can be seen that due to the large number of distributed power sources and load nodes in partition 1, the overall recovery process is slow. After 10 time steps, the total restored load is 2900kW. Partition 2 is completed after 3 time steps. During the load recovery process, the total restored load is 500kW. The recovery process of partition 3 is completed within 3 time steps, and the total restored load is 300kW. The recovery process of partition 4 is completed within 5 time steps, and the total restored load is 1300kW. Taking partition 4 as an example, it can be seen from the table that all lines in this partition are restored in the third step, while loads L66, L67, and L68 are only restored in the fourth and fifth steps. The recovery only depends on the constraints of the recovery order and the maximum number of lines allowed to be put into operation in each time step. The recovery of the load also needs to consider the output constraints and climbing constraints of the unit. The recovery of the other three partitions is consistent with the principle of partition 4, which will not be repeated.

[0210] This implementation also obtained the following results for a 10kV distribution network system with 39 nodes in a certain area (distribution network under N25HPY switch): Figure 5 The radial network structure shown in the figure makes the following assumptions:

[0211] (1) Assume that the power grid is in a "total blackout" state due to an extreme weather accident, the power infrastructure is severely damaged, the transmission network fails, and the entire distribution area cannot obtain power from the upper power grid. After emergency repairs, all faults in the distribution system have been repaired, and the recovery process of the distribution system can only be supported by local multi-type distributed power sources;

[0212] (2) After investigating the actual production and application of waste incineration power plants in Shanghai, a waste incineration power plant that matches the voltage level and total capacity of the power distribution system is selected as the power source in the calculation example;

[0213] (3) All nodes have load access, and the load scale and load priority in the recovery are ideal assumptions. According to the requirements for power supply reliability and load importance, the load can be divided into primary load, secondary load and tertiary load, and each level of load is controllable.

[0214] (4) Assume that the interval between two consecutive time steps in the sequential recovery process is set to 5 minutes, the total recovery time is 10 time steps, and the total time from fault outage to recovery is 50 minutes.

[0215] (5) Transient quantities such as the excitation surge current generated by starting the substation during emergency recovery and the impact current generated by the asynchronous motor are not considered.

[0216] Table 3 below shows the generator set recovery characteristic data.

[0217] Table 3 Generator set recovery characteristic data

[0218]

[0219] Through the above method, the preliminary division results are as follows Figure 6 As shown, the partitioning takes into account the electrical parameters of the line and the "source-load" matching within the subsystem, and further adjusts the initial partitioning based on the power supply radius.

[0220] Prioritize the search for the first and second level loads to determine the power supply range of the main power supply. After judging that the power of the important loads in the sub-partition is within the limit, the rationality of the partition is verified and a partition scheme is formed. The partition results are shown in Table 4.

[0221] The detailed recovery process of each partition is shown in Table 5 to Table 9. The final recovery effect of each partition is shown in Figures 7 to 11 shown.

[0222] Table 4 Partition results

[0223] Partition Load Node Partition 1 1、2、3、4、5、6、7、27、28、33、34、35、36、37、38、39 Zone 2 8、9、10、11、12、13 Zone 3 14、15、16、17、18、19 Zone 4 20、21、22、23、24、25、26 Zone 5 29、30、31、32

[0224] Table 5 Partition 1 Recovery Nodes, Lines and Loads in Each Step

[0225]

[0226] Table 6 Partition 2 recovery steps for nodes, lines and loads

[0227]

[0228] Table 7 Partition 3 steps to restore nodes, lines and loads

[0229]

[0230] Table 8 Partition 4 steps to restore nodes, lines and loads

[0231]

[0232] Table 9 Partition 5 steps to restore nodes, lines and loads

[0233]

[0234] Example 2

[0235] This embodiment provides a system for coordinated restoration of urban power grid zones taking into account source-grid-load interaction, including:

[0236] Topology building module: used to build the topology of the urban distribution network, wherein the topology represents the connection relationship between load nodes and power supply nodes;

[0237] Partition division module: used for dividing the urban distribution network based on the topological structure by using the shortest path algorithm to obtain multiple partitions;

[0238] Model building module: used to establish a distribution network restoration mathematical model taking into account partitions with the goal of maximizing the cumulative weighted load restoration amount;

[0239] Restoration module: used for solving the distribution network restoration mathematical model based on the multiple partitions to restore each partition in parallel.

[0240] The rest is the same as in Example 1.

[0241] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0242] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The schemes in the embodiments of the present invention may be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0243] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0244] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0245] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0246] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0247] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for coordinated restoration of urban power grid zones considering source-grid-load interaction, characterized in that: The following steps are involved: Establishing a topological structure of a city distribution network, wherein the topological structure represents a connection relationship between load nodes and power supply nodes; Based on the topological structure, the urban distribution network is divided into multiple partitions using the shortest path algorithm; A distribution network restoration mathematical model considering partitions is established with the goal of maximizing the cumulative weighted load restoration. Based on the multiple partitions, the distribution network restoration mathematical model is solved to restore each partition in parallel.

2. According to claim 1, a method for coordinated restoration of urban power grid zones taking into account source-grid-load interaction is characterized in that: The power supply nodes include black start power supply nodes and non-black start power supply nodes.

3. A method for coordinated restoration of urban power grid zones considering source-grid-load interaction according to claim 2, characterized in that: The step of obtaining multiple partitions comprises: Modeling the topology as an undirected weighted graph; Determining the number of partitions and black start power supply nodes in each partition based on the undirected weighted graph; Based on the undirected weighted graph, the shortest path algorithm is used to solve the shortest distance from each node to be restored to the black start power supply node, so as to divide the node to be restored into the partition where the start power supply node with the shortest distance is located, and obtain an initial partition result; Based on the initial partitioning result, a power balance check is performed within the partition to obtain a final partitioning result.

4. A method for coordinated restoration of urban power grid zones considering source-grid-load interaction according to claim 3, characterized in that: The weight of the edge in the undirected weighted graph is adjusted according to the electrical distance of the line, and the expression of the edge weight is: ω ij =Z line l ij In the formula, ω ij is the weight of edge (i, j), Z line Indicates the impedance value corresponding to the line per unit distance, l ij Indicates the length of line (i, j).

5. A method for coordinated restoration of urban power grid zones considering source-grid-load interaction according to claim 3, characterized in that: The expression for determining the number of partitions is: m=min{N black ,N load }m In the formula, is the start-stop state of the dispatchable DG without black start function, μ g，t is the start and stop status of the unit, m is the maximum number of partitions, N black 、N load They represent the number of black start nodes and load nodes in the system respectively.

6. A method for coordinated restoration of urban power grid zones considering source-grid-load interaction according to claim 3, characterized in that: The step of performing intra-partition power balance verification comprises: Determine whether the load power connected to each initial partition exceeds the total power that each initial partition can provide. If so, delete the unimportant load nodes on the edge of each initial partition to meet the power requirements. If not, do nothing.

7. A method for coordinated restoration of urban power grid zones taking into account source-grid-load interaction according to claim 6, characterized in that: The constraint expression of the power balance check is: In the formula, a ij Indicates whether node j is in partition i, n is the total number of nodes, is the load input state of node j, is the active load connected at node j, represents the active power of the power source connected to node i, Indicates whether the power supply of node j is connected, and m is the number of partitions.

8. The method for coordinated restoration of urban power grid zones considering source-grid-load interaction according to claim 1 is characterized in that: The distribution network restoration mathematical model includes an objective function and constraints, and the objective function is: In the formula, f is the objective function, n is the number of nodes, m is the number of partitions, T is the number of time periods, ΔT is the time taken for each time step, and a is the ij Indicates whether node j is in partition i, represents the weight of the load at node j, is the active power demand of the load at node j at time t, and L is the load set; The constraints include: (1) Power flow equation constraints -M·m jk，t ≤P jk，t ≤M·μ jk，t ,(j,k)∈A -M·m jk，t ≤Q jk，t ≤M·μ jk，t ,(j,k)∈A V j，t -V k，t ≤2(r jk ·P jk，t +x jk ·Q jk，t )+M(1-μ jk，t ), j, k = 1, 2, ..., n V j，t -V k，t ≥2(r jk ·P jk，t +x jk ·Q jk，t )-M(1-μ jk，t ), j, k = 1, 2, ..., n In the formula, is the reactive power demand of the load at node j at time t, and are the active and reactive outputs of the distributed generation at node j at time t, M is a constant, μ jk , t is a binary variable representing the recovery state of line (j, k) in time period t, μ jk，t =1 indicates that line (j, k) is restored in time period t, P jk，t , Q jk，t 、r jk and x jk are respectively the active power, reactive power, resistance and reactance on line (j, k) in time period t, V j，t is the square of the voltage amplitude of the first node j of line (j, k) in time period t, A is the set of lines, V k，t is the square of the voltage amplitude at the end node k of line (j, k) in time period t; (2) Distributed power output constraints In the formula, represents the startup state of the kth distributed generation in time period t, and They represent the lower and upper limits of the active output of the kth distributed generation, and They represent the lower and upper limits of reactive power output of the kth distributed generation, respectively. G is the set of distributed generation. is the active power of the DG connected to node k in partition s at time t, is the reactive power of the DG connected to node k in partition s at time t; (3) Load constraints In the formula, Indicates whether the jth load t period is put into use, if yes, it takes 1, otherwise it takes 0, P j , max , Q j , max Respectively represent the active and reactive demands of the load; (4) Distributed power supply ramp constraints In the formula, is the maximum ramp rate of the kth unit, is the active power of the DG connected to node k at time t; (5) Node voltage constraints In the formula, and Respectively represent the maximum and minimum node voltage, μ i，t Indicates whether node i is engaged in time period t, if yes, it takes 1, otherwise it takes 0; (6) Line power constraints In the formula, is the maximum transmission capacity of branch (j, k), P ij , t is the active transmission capacity of branch (i, j) at time t, Q ij , t is the reactive power transfer capacity of branch (i, j) at time t, S jk is the radius of the polygon involved in the polygon-based linearization method to approximate the quadratic term, n is the number of sides of the polygon used for linearization; (7) Restoring order constraints 1) A non-black start power node can only be restored after the node it is on is restored: 2) The line can be powered only when both end nodes are powered: m jk，t ≤μ j，t ,(j,k)∈A m jk，t ≤μ k，t ,(j,k)∈A 3) If the node is connected to a black start power node or a substation node, it will be powered on: 4) The load can only be powered when connected to a powered node: 5) Constraints on the number of lines to be closed: 6) Additional constraints to ensure that distribution network components will not be cut off after power is applied: m jk，t -m jk，t-1 ≥0,(j,k)∈A In the formula, μ j , t is the input state of node j at time t, G N is the set of units without black start capability, G Y is the set of black start power nodes, N lim To allow the number of lines to be put into operation simultaneously; (8) Network topology constraints Where N D is the set of all nodes, μ ij , t is the input state of branch (i, j) at time t; (9) Initial condition constraints All lines are opened in the first step: m ij，1 =0,(i,j)∈A The black start power node is powered on in the first step: In the formula, μ ij , 1 is the initial state of branch (i, j), is the input state of the black start power supply on node i at the initial moment.

9. The method for coordinated restoration of urban power grid zones considering source-grid-load interaction according to claim 1, characterized in that: ILOG CPLEX is used for solving.

10. A coordinated restoration system for urban power grid zones considering source-grid-load interaction, characterized in that: include: Topology building module: used to establish the topology of the urban distribution network, wherein the topology represents the connection relationship between the load nodes and the power supply nodes; Partition division module: used for dividing the urban distribution network based on the topological structure by using the shortest path algorithm to obtain multiple partitions; Model building module: used to establish a distribution network restoration mathematical model taking into account partitions with the goal of maximizing the cumulative weighted load restoration amount; Restoration module: used for solving the distribution network restoration mathematical model based on the multiple partitions to restore each partition in parallel.

Images