CN111416359A - Power distribution network reconstruction method considering weighted power flow entropy - Google Patents

Abstract

The invention relates to a power distribution network reconstruction method considering weighted power flow entropy, which includes: dividing an islanded distribution network after a fault to obtain a regional network, wherein, if there is an isolated island, the regional network is the distribution network to remove the isolated island Otherwise, the entire distribution network after the fault is a regional network; the traditional power supply index and power flow entropy are used as evaluation indicators to build a network reconstruction optimization model. The traditional power supply index includes network loss and node voltage drop, power flow entropy It is used to quantitatively evaluate the inhomogeneity of the line load distribution; based on the CBPSO algorithm and the constructed network reconstruction optimization model, the regional network is solved to obtain the global optimal solution, which is the reconstructed network topology. Compared with the prior art, the present invention can improve the robustness of the reconstructed network and meet the requirements of stable operation of the reconstructed network by introducing the weighted power flow entropy as a system robustness evaluation index into the optimization objective function.

Description

A Distribution Network Reconfiguration Method Considering Weighted Power Flow Entropy

technical field

The invention relates to the technical field of power system distribution network reconfiguration, in particular to a distribution network reconfiguration method considering weighted power flow entropy.

Background technique

With the development of power system distribution network, the network structure has become more complex. When the power distribution network fails, because the network contains a large number of load nodes, the unbalanced power flow distribution of the line will reduce the robustness of the power grid. For the network reconstruction during the fault recovery process, the power flow is mainly adjusted by adjusting the on-off of the line switch, which not only needs to reduce the line loss and ensure the voltage quality, but also should improve the line load balance to reduce the probability of the grid falling into the self-organized critical state.

The research on network reconfiguration is mainly divided into two types: dynamic reconfiguration and static reconfiguration. Static reconfiguration assumes that the load is constant throughout the time period and only considers the reconfiguration of a single time section; dynamic reconfiguration considers the temporal changes of the load. , to dynamically and continuously optimize the network topology over a period of time. At present, the research on network reconstruction mainly focuses on three aspects, including:

(1) The problem of island division caused by the connection of DG and energy storage devices to the distribution network.

(2) The choice of network optimization objective function.

(3) Research on the network reconstruction algorithm.

Due to the unbalanced distribution of line loads in the current distribution network, the distribution network is prone to fall into a self-organized critical state. Most of the existing network reconstruction methods only consider the power supply index, which cannot effectively ensure the stable operation of the network after reconstruction.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a power distribution network reconstruction method considering the weighted power flow entropy in order to overcome the above-mentioned defects of the prior art. Combined with island division, and according to the characteristics of line load distribution unbalanced, based on the line load rate, construct The weighted power flow entropy, combined with the traditional power supply index, is used as the evaluation index of network reconfiguration, thereby improving the robustness of the reconfigured network and ensuring the stable operation of the distribution network.

The purpose of the present invention can be achieved through the following technical solutions: a method for reconfiguring a power distribution network considering weighted power flow entropy, which is used to reconfigure the distribution network after a fault, including the following steps:

S1. Divide the islanded distribution network after the fault to obtain the regional network, wherein, if there is an isolated island, the regional network is the distribution network except the islanded network, otherwise the entire distribution network after the fault is the regional network;

S2. The traditional power supply index and power flow entropy are used as evaluation indicators to construct a network reconfiguration optimization model. The traditional power supply index includes network loss and node voltage drop, and the power flow entropy is used to quantitatively evaluate the inhomogeneity of line load distribution;

S3. Based on the CBPSO (Chaotic Binary Panicle Swarm Optimization) algorithm and the constructed network reconstruction optimization model, the regional network is solved to obtain the global optimal solution, which is the reconstructed network topology.

Further, the step S1 specifically includes the following steps:

S11. Obtain the capacity and location of the distributed power supply participating in the grid connection in the distribution network;

S12. If the capacity of the distributed power supply participating in the grid connection meets the power demand of the load node, step S13 is performed; otherwise, it is judged that there is no island in the distribution network after the failure, and the entire distribution network after the failure is used as the regional network, and then execute step S2;

S13 , searching for the island area according to the output power of the distributed power source, and supplying power to the island node load by the distributed power source, removing the island after the failure of the distribution network to obtain an area network, and then performing step S2 .

Further, in the step S12, the conditions for the capacity of the distributed power source to meet the electricity demand of the load node are specifically:

Among them, S _i is the load of node i, ΔS _j is the line power loss, _{SD DG} is the capacity of the distributed power supply, and g is the maximum number of load-bearing nodes in the island area.

Further, the step S2 specifically includes the following steps:

S21, establishing a multi-objective optimization function by combining network loss, node voltage drop and power flow entropy, wherein the network loss is specifically the active network loss, the node voltage drop is specifically the minimum node voltage deviation, and the power flow entropy is specifically the weighted power flow entropy;

S22. Set constraints, where the constraints include network power flow constraints, voltage constraints, line power constraints, and topology constraints.

Further, the multi-objective optimization function in the step S21 is specifically:

minf={minf ₁ , minf ₂ , minf ₃ }

f=ω ₁ ·f ₁ +ω ₂ ·f ₂ +ω ₃ ·f ₃

ω ₁ +ω ₂ +ω ₃ =1

Among them, f is the optimized value of the objective function, f ₁ is the active network loss, f ₂ is the minimum node voltage deviation, f ₃ is the weighted power flow entropy, ω ₁ , ω ₂ , ω ₃ are f ₁ , f ₂ , f ₃ respectively the weight of.

Further, the active network loss in the step S21 is specifically:

Among them, R _ij is the line resistance between node i and node j, P _ij and Q _ij are the active and reactive power flowing through line ij, respectively, P _DGi and Q _DGi are the active power of the distributed power source DG at node i, respectively Output and reactive power output, if the corresponding node has no distributed power supply DG, then P _DGi =0, Q _DGi =0, U _i is the voltage of node i, M is the total number of lines, and N is the total set of nodes;

The minimum node voltage deviation is specifically:

Among them, U _min is the minimum node voltage, U _e is the node rated voltage;

The weighted power flow entropy is specifically:

Among them, w is the power flow entropy weight, P _l is the actual active power flow value of line l, P _max is the maximum active power flow value of all lines, P _min is the minimum active power flow value of all lines, and m is the number of state classifications;

P(X _a ) is the probability of the occurrence of the a-th state, that is, the probability that the line load rate is in any power equal difference interval:

η _l ∈[D _k , D _k+1 ], k<K

D=[D ₀ , D ₁ , ..., D _K ]

In the formula, η _l is the load rate of line l, D is the continuous equal difference interval formed according to the line operating power limit requirements, and K is the total number of power equal difference intervals.

Further, the network power flow constraint in the step S22 is specifically:

Among them, ΔP _i and ΔQ _i are the active and reactive power losses of node i, respectively, P _i and Q _i are the active and reactive power injected by node i, respectively, U _i , U _j are the voltage values of nodes i and j, respectively, G _ij , B _ij , and θ _ij are the conductance, susceptance, and voltage phase angle differences of the branch between nodes i and j, respectively;

The voltage constraints are specifically:

U _i.min ≤U _i ≤U _i.max

Among them, U _i.min and U _i.max are the upper and lower voltage limits of node i, respectively;

The line power constraints are specifically:

S _l ≤1.5×S _l.max

Among them, S _l is the power of line i, and S _l.max is the maximum power allowed by line i;

The topological constraints are specifically:

Among them, K _q,r is the switch state of the branch switch (q, r), that is, the branch switch state with q as the head node and r as the tail node, K _q,r =1 when the switch is closed, and K when the switch is open _q,r = 0, N _n is the total number of nodes in the topology, N _f is the number of root nodes in the topology, N _g is the number of island nodes in the topology, B is the set of branches in the topology, R is the root node and island nodes except for The set of nodes, N is the total set of nodes, F is the set of root nodes, and G is the set of island nodes;

In topological constraints:

K _q,r = 1 means that a single branch with r as the tail node must be closed,

即要求所有支路均是闭合的。

That is, all branches are required to be closed.

Further, the step S3 specifically includes the following steps:

S31, initialize the parameters of the regional network nodes, and set the global parameters of the CBPSO algorithm to initialize the position and speed of the particle swarm;

S32, based on the network reconstruction optimization model, combined with the coding rules of the CBPSO algorithm, update the particle position and speed;

S33, judging whether the updated particle complies with the topology constraint, if it is judged to be yes, then execute step S34, otherwise return to step S32, wherein, one particle corresponds to a network topology, and the network topology is composed of different branch switch state data;

S34. Calculate the fitness of the particle swarm, and update the optimal fitness of each particle and the global optimal fitness, where the fitness corresponds to the optimal value of the objective function, and the optimal fitness corresponds to the minimum optimal value of the objective function;

S35: Determine whether the solution process satisfies the convergence condition, and if so, output the optimal particle, which is to reconstruct the network topology, and if not, return to step S32.

Further, the global parameters of the CBPSO algorithm in the step S31 include the maximum number of iterations, the learning factor, the number of populations and the chaotic mapping parameters.

Further, the step S35 specifically includes the following steps:

S351. Determine whether the current number of iterations reaches the set maximum number of iterations, and if the determination is yes, output the optimal particle, otherwise, perform step S352;

S352: Determine whether the global optimal fitness obtained by the current solution satisfies the convergence accuracy requirement, and if the determination is yes, output the optimal particle, otherwise return to step S32.

Compared with the prior art, the present invention has the following advantages:

1. The present invention proposes a multi-objective optimization model that takes the power flow entropy as the evaluation index and takes into account the network loss and node voltage drop, and combines the island division, By constructing a more reasonable network topology, in the process of power grid reconstruction, as many loads affected by faults can be restored as possible, and more important loads can be restored first, thereby effectively improving the robustness of the reconstructed network. Guarantee the operational stability of the reconstructed network.

2. The present invention considers that there are some branch faults with uneven power flow distribution in the distribution network after the fault, that is, the line load rate is inconsistent. In order to avoid the calculation of the power flow entropy being too large, the present invention adopts the method of weighted power flow entropy, which can effectively Distinguish the severity of load lines, thereby improving the accuracy of power flow entropy calculation and further ensuring the operational stability of the reconstructed network.

Description of drawings

Fig. 1 is the method flow schematic diagram of the present invention;

Fig. 2 is a schematic diagram of the area network solution process;

Fig. 3 is the original network structure diagram of 37 branches in the embodiment;

4 is a structural diagram of a reconstructed network in which 31 nodes are connected to a distributed power source DG in an embodiment;

Fig. 5 is the reconstruction network structure diagram that comprises an island in the embodiment;

6 is a schematic diagram of an isolated island corresponding to FIG. 5;

7 is a structural diagram of a reconstructed network comprising two isolated islands in an embodiment;

8 is a schematic diagram of an isolated island corresponding to FIG. 7;

FIG. 9 is a schematic diagram showing the comparison of minimum node voltages before and after network reconfiguration.

Detailed ways

The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

Example

As shown in Figure 1, a distribution network reconstruction method considering weighted power flow entropy includes the following steps:

S1. Divide the islanded distribution network after the fault to obtain the regional network, wherein, if there is an isolated island, the regional network is the distribution network except the islanded network, otherwise the entire distribution network after the fault is the regional network;

S2. The traditional power supply index and power flow entropy are used as evaluation indicators to construct a network reconfiguration optimization model. The traditional power supply index includes network loss and node voltage drop, and the power flow entropy is used to quantitatively evaluate the inhomogeneity of line load distribution;

S3. Based on the CBPSO algorithm and the constructed network reconstruction optimization model, the regional network is solved to obtain a global optimal solution, which is to reconstruct the network topology.

Based on the reconstructed network topology obtained by the solution, the switch state of each branch of the distribution network after the fault is controlled to complete the network reconstruction.

Specifically, applying the above method to practice mainly includes the following steps:

Step 1: Islanding the distribution network after the fault, and establishing an optimization model including network loss and node voltage drop, with power flow entropy as the robustness evaluation index;

Step 1.1, divide the island according to the capacity and location of the grid-connected DG:

(1) When the DG capacity is small and the output power is not enough to meet the power generation demand of the load node during normal operation, only the corresponding load is offset without forming an island;

(2) When the DG capacity is large, the normal output power can meet the power demand of the load node. Therefore, when the front-end line is powered off, the DG can be used as the power source to supply power to the node load, namely:

Among them, S _i is the load of node i, ΔS _j is the line power loss, _{SD DG} is the capacity of the distributed power supply, and g is the maximum number of load-bearing nodes in the island area;

Step 1.2, establish a multi-objective optimization scheme that comprehensively considers network network loss, node voltage deviation and power flow entropy: in the process of distribution network reconstruction, ensure that as many loads affected by the fault are restored as possible, and priority restoration is more important. load,

Multi-objective optimization function:

minf={minf ₁ , minf ₂ , minf ₃ }

In the formula, f is the optimized value of the objective function, and the weight method is used to convert the multi-objective function into a single-objective function, namely:

f=ω ₁ ·f ₁ +ω ₂ ·f ₂ +ω ₃ ·f ₃

In the formula, ω ₁ , ω ₂ , and ω ₃ are the weights of each sub-objective f ₁ , f ₂ , and f ₃ respectively, which must satisfy ω ₁ +ω ₂ +ω ₃ =1. In this embodiment, ω ₁ =0.5, ω ₂ =0.1, ω ₃ =0.4;

The sub-goals are:

1) Active network loss

If DG is connected to the network, but the DG capacity is not enough to meet the power demand of the load:

Among them, R _ij is the line resistance between node i and node j, P _ij and Q _ij are the active and reactive power flowing through line ij, respectively, P _DGi and Q _DGi are the active power of the distributed power source DG at node i, respectively Output and reactive power output, if the corresponding node has no distributed power supply DG, then P _DGi =0, Q _DGi =0, U _i is the voltage of node i, M is the total number of lines, and N is the total set of nodes;

2) Minimum node voltage drop

Among them, U _min is the minimum node voltage, U _e is the node rated voltage;

3) Weighted power flow entropy

As a state scalar, entropy is used to measure the chaotic degree and disorder state of the motion of a large number of microscopic particles in space. Power flow entropy is used to quantitatively evaluate the inhomogeneity of network line load distribution.

实际运行中线路潮流值为

则线路的负载率η_l为：Let the maximum load capacity of line l be

In actual operation, the line power flow value is

Then the load rate η _l of the line is:

In the formula, M is the number of lines;

A continuous equidistant interval D=[D ₀ , D ₁ ,..., _K ] is formed according to the line operating power limit requirements, and D _K can be set according to the actual operation requirements of the network. During the network reconstruction process, the line load exceeds the interval. The scheme is not considered, so it is defined that the load rate of line l is in the interval η _l ∈ [D _k , D _k+1 ], and the probability of k<K is P(X _a ), then the power flow entropy is defined as follows:

In the formula, C is a constant, m is the number of state classifications, P(X _a ) is the probability of the occurrence of the a-th state,

The larger the power flow entropy is, the higher the inconsistency of the system line load rate is. The excessively scattered line load rate will increase the probability of the system entering a self-organized critical state by small disturbances, thus affecting the robustness of the system. When the line load rate is concentrated in the same interval, the power flow entropy of the line is 0. At this time, the power flow distribution of the line is the most orderly. When the system is disturbed, cascading failures are not easy to occur. Both will cause the power flow entropy to be too large. Obviously, the latter cannot correctly reflect the physical characteristics of the power flow entropy. The weighted power flow entropy can effectively distinguish the severity of the load line and improve the accuracy of the power flow entropy value calculation. The calculation formula of the power flow entropy weight is as follows :

In the formula, w is the power flow entropy weight, P _l is the actual active power flow value of line l, P _max is the maximum active power flow value of all lines, and P _min is the minimum active power flow value of all lines;

Therefore, the weighted power flow entropy is calculated as follows:

Step 1.3, add network power flow constraints

The network power flow constraints must be satisfied during the reconstruction process:

Voltage constraint: U _i.min ≤U _i ≤U _i.max

Line power constraint: S _l ≤1.5×S _l.max

Topological constraints:

In the formula, ΔP _i and ΔQ _i are the active and reactive power losses of node i, respectively, P _i and Q _i are the active and reactive power injected by node i, respectively, U _i , U _j are the voltage values of nodes i and j, respectively , G _ij , B _ij , and θ _ij are the conductance, susceptance, and voltage phase angle difference of the branch between nodes i and j, respectively; S _l is the load of node l, and S _l.max is the maximum operating power of node l , K _q,r is the switching state of the branch switch (q, r), K _q,r =1 when the switch is closed, K _q,r =0 when the switch is open, N _n is the total number of nodes in the topology, N _f is The number of root nodes in the topology, N _g is the number of island nodes in the topology, B is the set of branches in the topology, R is the set of nodes except the root node and the island node, N is the total set of nodes, F is the set of root nodes, G A collection of island nodes.

Step 2, using the chaotic binary particle swarm algorithm for optimization, as shown in Figure 2, including:

Step 2.1, initialize the distribution network node parameters, set the global parameters of the chaotic binary particle swarm algorithm, and initialize the particle swarm position and speed;

Step 2.2, according to the coding rules, update the particle swarm position and velocity;

Step 2.3, judging whether each particle meets the network topology requirements, a particle is specifically a network topology including the switch states of each branch, and a network topology is composed of a plurality of branch switch states;

Step 2.4, calculate the particle swarm fitness, update the optimal fitness of each particle and the global optimal fitness;

Step 2.5, judge whether the convergence condition is met;

Step 2.6, output the optimal particle swarm composition.

In this embodiment, the method proposed in the present invention is used to compare and analyze the network reconfiguration results in different situations such as the distribution location of the distributed power DG and whether the node forms an island, so as to provide a new idea for the actual operation of the distribution network and the configuration of the DG.

This embodiment uses the IEEE 33 standard test system as an example. As shown in FIG. 3 , the network includes 37 branches (32 conventional branches and 5 tie switch branches) and 33 nodes.

The network reference voltage is 12.66KV, the total load is 3715+j2300KVA, the initial parameters of the chaotic binary particle swarm algorithm: the maximum number of iterations is 100, the learning factor c ₁ =c ₂ =2, the population number is 50, and the particle velocity V _i ⁿ ∈[-4 , 4], chaotic mapping parameters: Y ₁ =0.5, μ=3.99.

It is assumed that the DG system added to the network includes photovoltaic power generation and wind power generation, and both are equipped with enough energy storage devices to operate stably with independent load. It is assumed that the average output power of the DG system is 500kVA when the reconstruction occurs. The distribution network containing DG mainly guarantees the continuous power supply of part of the load in the event of a fault through the island operation mode, and improves the reliability of the power supply. However, if the DG capacity is small and the output power is not enough to bear the load of the node, the island will not be formed. When the DG capacity is large, the normal output power can meet the power demand of the load node, which can be regarded as the power supply for the island load.

Except for the network reconfiguration without DG, this embodiment mainly analyzes the following situations:

1) There is only one distributed power system. Assuming that the DG is installed in 31 nodes, but does not form an island, after the network is reconfigured, as shown in Figure 4, Figure 9 and Table 1.

2) If the DG capacity meets the normal power demand of the load node 31, the node 31 can form an island independently, and the power is provided by the DG output power. Similarly, if the DG power generation can still meet the load demand of other nodes connected to node 31, priority is given to restoring load nodes with higher importance, such as node 30, to form an island area; after eliminating the island area, the remaining branches will be connected to the power grid. After reconstruction, the results are shown in Figure 5, Figure 6, Figure 9 and Table 1.

3) There are two distributed power systems DG1 and DG2. If DG1 and DG2 are connected to nodes 25 and 31 respectively, a new island area can be formed according to the output of DG and the importance of the load. The remaining branches are reconstructed, and the results are shown in Figure 7, Figure 8, Figure 9 and Table 1.

Table 1

It can be seen from Table 1 that even if the network is reconfigured without DG, the network loss and power flow entropy can be significantly reduced, while the minimum node voltage can be significantly increased. When DG is added at node 31 without forming an island, the power flow entropy is slightly increased, that is, the uniformity of power flow distribution is slightly worse, but the voltage drop is reduced, and the network loss is lower;

When the DG and its branches near the connecting node form an island powered by the DG, the power supply length of some lines increases, so the minimum node voltage decreases significantly. In this case, you can choose to continue to install the DG in the long line and at the end;

When adding DG to multiple nodes in the system, the network loss is significantly reduced, but the line power flow entropy may increase. This is caused by the unreasonable installation position of the DG. At this time, the DG installation position can be adjusted or the DG installation point can be added. To optimize the network power flow entropy.

In addition, in this embodiment without DG, the BPSO (Discrete Binary Particle Swarm Optimization, binary particle swarm optimization) algorithm and the CBPSO algorithm are respectively used to solve the area network. The comparison results of the two algorithms are shown in Table 2. , from the results in Table 2, it can be seen that the reconstruction results of the two algorithms are basically the same, but the average convergence times of the CBPSO algorithm is significantly reduced, that is, the global search ability is stronger, which reflects the effectiveness of the algorithm, which also ensures the method of the present invention. reliability.

Table 2

algorithm BPSO CBPSO Average Iteration Convergence Times 47 16 Minimum node voltage p.u. 0.9348 0.9386 network loss 150.12 149.63 Power flow entropy 0.5910 0.5916

Claims (10)

1. a distribution network reconfiguration method considering weighted power flow entropy, is characterized in that, for carrying out network reconfiguration to the distribution network after the fault, comprising the following steps: S1. Divide the islanded distribution network after the fault to obtain the regional network, wherein, if there is an isolated island, the regional network is the distribution network except the islanded network, otherwise the entire distribution network after the fault is the regional network; S2. The traditional power supply index and power flow entropy are used as evaluation indicators to construct a network reconfiguration optimization model. The traditional power supply index includes network loss and node voltage drop, and the power flow entropy is used to quantitatively evaluate the inhomogeneity of line load distribution; S3. Based on the CBPSO algorithm and the constructed network reconstruction optimization model, the regional network is solved to obtain a global optimal solution, which is to reconstruct the network topology. 2. A power distribution network reconfiguration method considering weighted power flow entropy according to claim 1, wherein the step S1 specifically comprises the following steps: S11. Obtain the capacity and location of the distributed power supply participating in the grid connection in the distribution network; S12. If the capacity of the distributed power supply participating in the grid connection meets the power demand of the load node, step S13 is performed; otherwise, it is judged that there is no island in the distribution network after the failure, and the entire distribution network after the failure is used as the regional network, and then execute step S2; S13 , searching for the island area according to the output power of the distributed power source, and supplying power to the island node load by the distributed power source, removing the island after the failure of the distribution network to obtain an area network, and then performing step S2 . 3. A power distribution network reconfiguration method considering weighted power flow entropy according to claim 2, characterized in that, in the step S12, the condition that the capacity of the distributed power source meets the electricity demand of the load node is specifically:

Among them, S _i is the load of node i, ΔS _j is the line power loss, _{SD DG} is the capacity of the distributed power supply, and g is the maximum number of load-bearing nodes in the island area. 4. A power distribution network reconfiguration method considering weighted power flow entropy according to claim 2, wherein the step S2 specifically comprises the following steps: S21, establishing a multi-objective optimization function by combining network loss, node voltage drop and power flow entropy, wherein the network loss is specifically the active network loss, the node voltage drop is specifically the minimum node voltage deviation, and the power flow entropy is specifically the weighted power flow entropy; S22. Set constraints, where the constraints include network power flow constraints, voltage constraints, line power constraints, and topology constraints. 5. A power distribution network reconstruction method considering weighted power flow entropy according to claim 4, wherein the multi-objective optimization function in the step S21 is specifically: min f={min f ₁ , min f ₂ , min f ₃ } f=ω ₁ ·f ₁ +ω ₂ ·f ₂ +ω ₃ ·f ₃ ω ₁ +ω ₂ +ω ₃ =1 Among them, f is the optimized value of the objective function, f ₁ is the active network loss, f ₂ is the minimum node voltage deviation, f ₃ is the weighted power flow entropy, ω ₁ , ω ₂ , ω ₃ are f ₁ , f ₂ , f ₃ respectively the weight of. 6. A power distribution network reconfiguration method considering weighted power flow entropy according to claim 5, wherein the active network loss in the step S21 is specifically:

Among them, R _ij is the line resistance between node i and node j, P _ij and Q _ij are the active and reactive power flowing through line ij, respectively, P _DGi and Q _DGi are the active power of the distributed power source DG at node i, respectively Output and reactive power output, if the corresponding node has no distributed power supply DG, then P _DGi =0, Q _DGi =0, U _i is the voltage of node i, M is the total number of lines, and N is the total set of nodes; The minimum node voltage deviation is specifically:

Among them, U _min is the minimum node voltage, U _e is the node rated voltage; The weighted power flow entropy is specifically:

Among them, w is the power flow entropy weight, P _l is the actual active power flow value of line l, P _max is the maximum active power flow value of all lines, P _min is the minimum active power flow value of all lines, and m is the number of state classifications; P(X _a ) is the probability of the occurrence of the a-th state, that is, the probability that the line load rate is in any power equal difference interval: η _l ∈[D _k , D _k+1 ], k<K D=[D ₀ , D ₁ , ..., D _K ] In the formula, η _l is the load rate of line l, D is the continuous equal difference interval formed according to the line operating power limit requirements, and K is the total number of power equal difference intervals. 7. A power distribution network reconfiguration method considering weighted power flow entropy according to claim 6, characterized in that, in the step S22, the network power flow constraint is specifically:

Among them, ΔP _i and ΔQ _i are the active and reactive power losses of node i, respectively, P _i and Q _i are the active and reactive power injected by node i, respectively, U _i and U _j are the voltage values of nodes i and j, respectively, G _ij , B _ij , and θ _ij are the conductance, susceptance, and voltage phase angle differences of the branch between nodes i and j, respectively; The voltage constraints are specifically: U _i.min ≤U _i ≤U _i.max Among them, U _i.min and U _i.max are the upper and lower voltage limits of node i; The line power constraints are specifically: S _l ≤1.5×S _l.max Among them, Sl is the power of line ₁ , and _Sl.max is the maximum power allowed by line 1; The topological constraints are specifically:

Among them, K _q,r is the switch state of the branch switch (q, r), that is, the branch switch state with q as the head node and r as the tail node, K _q,r =1 when the switch is closed, and K when the switch is open _q,r = 0, N _n is the total number of nodes in the topology, N _f is the number of root nodes in the topology, N _g is the number of island nodes in the topology, B is the set of branches in the topology, R is the root node and island nodes except for The node set of , N is the total set of nodes, F is the root node set, and G is the island node set. 8. A power distribution network reconfiguration method considering weighted power flow entropy according to claim 1, wherein the step S3 specifically comprises the following steps: S31, initialize the parameters of the regional network nodes, and set the global parameters of the CBPSO algorithm to initialize the position and speed of the particle swarm; S32, based on the network reconstruction optimization model, combined with the coding rules of the CBPSO algorithm, update the particle position and speed; S33, judging whether the updated particle complies with the topology constraint, if it is judged to be yes, then execute step S34, otherwise return to step S32, wherein, one particle corresponds to a network topology, and the network topology is composed of different branch switch state data; S34. Calculate the fitness of the particle swarm, and update the optimal fitness of each particle and the global optimal fitness, where the fitness corresponds to the optimal value of the objective function, and the optimal fitness corresponds to the minimum optimal value of the objective function; S35: Determine whether the solution process satisfies the convergence condition, and if so, output the optimal particle, which is to reconstruct the network topology, and if not, return to step S32. 9. a kind of power distribution network reconstruction method considering weighted power flow entropy according to claim 8, is characterized in that, in described step S31, the global parameter of CBPSO algorithm comprises maximum number of iterations, learning factor, population number and chaotic map parameter. 10. A power distribution network reconfiguration method considering weighted power flow entropy according to claim 8, wherein the step S35 specifically comprises the following steps: S351. Determine whether the current number of iterations reaches the set maximum number of iterations, and if the determination is yes, output the optimal particle, otherwise, perform step S352; S352: Determine whether the global optimal fitness obtained by the current solution satisfies the convergence accuracy requirement, and if the determination is yes, output the optimal particle, otherwise return to step S32.

Images