Distribution network partition voltage optimization method based on distributed photovoltaic cluster

Abstract

The invention relates to the technical field of power distribution network voltage control, and discloses a power distribution network partition voltage optimization method based on a distributed photovoltaic cluster, and the method comprises the steps: obtaining the reactive voltage sensitivity between nodes, calculating the electrical distance between the nodes, carrying out the cluster division of a power distribution network, and obtaining the cluster of the power distribution network; voltage optimization is carried out by taking the cluster as a basic regulation and control unit, and when reactive power in the cluster is sufficient, the photovoltaic reactive power support quantity in the cluster is calculated based on the regional reactive power-voltage sensitivity of the cluster to which the voltage out-of-limit node belongs; when reactive power in the cluster is insufficient, mapping an optimization target of a voltage out-of-limit node of the original cluster to a boundary node of an adjacent cluster, only exchanging information of the boundary node with the adjacent cluster, and suppressing voltage out-of-limit by adjusting the voltage of the boundary node of the adjacent cluster; according to the method, reactive power resources can be effectively utilized and voltage out-of-limit is suppressed, so that voltage optimization is realized, more reactive power reserve can be reserved while network loss is optimized, and the voltage control time is remarkably shortened.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network voltage control, and in particular to a distribution network partition voltage optimization method based on a distributed photovoltaic cluster. Background Art

[0002] As the scale of distributed photovoltaic access to distribution networks continues to increase, the volatility and intermittency of photovoltaic power generation have led to serious voltage over-limit problems in distribution networks. At the same time, the rapid reactive power regulation capability of distributed photovoltaic inverters enables them to participate in distribution network voltage optimization and reduce voltage over-limit risks. Therefore, how to give full play to the regulation potential of distributed photovoltaics and propose a distribution network voltage optimization method that is suitable for large-scale distributed photovoltaic access has become the focus of current research.

[0003] At present, the voltage optimization of distribution networks mostly adopts centralized control methods. Centralized control collects and analyzes the global information of the distribution network through the central controller, and then transmits control instructions to each controllable unit. Centralized control requires all controllable devices to be connected to the distribution network control master station, and has high requirements for communication conditions. Due to the lack of complete communication conditions in the distribution network, the implementation of centralized control is difficult. Summary of the invention

[0004] In view of the above-mentioned deficiencies in the prior art, the present invention provides a distribution network partition voltage optimization method based on a distributed photovoltaic cluster, which is used to solve the problem of voltage optimization difficulty caused by incomplete distribution network communication conditions in the prior art.

[0005] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0006] A distribution network partition voltage optimization method based on a distributed photovoltaic cluster comprises the following steps:

[0007] S1, obtain the reactive voltage sensitivity between each node in the distribution network and calculate the electrical distance between each node;

[0008] S2, clustering the distribution network according to the electrical distance between each node to obtain the cluster of the distribution network;

[0009] S3, determine whether there is a voltage over-limit and the voltage over-limit duration exceeds the set threshold, if so, execute step S4, otherwise, return to step S2;

[0010] S4. Sort the voltages of all nodes whose voltage over-limit duration exceeds the set threshold from small to large, obtain the node with the most serious voltage over-limit, obtain the cluster to which the node with the most serious voltage over-limit belongs and the numbers of all nodes in the cluster, use the cubic exponential smoothing method to predict the voltage over-limit duration of the node with the most serious voltage over-limit and the cubic exponential smoothing value of the voltage of the node with the most serious voltage over-limit, and calculate the voltage prediction value of the node with the most serious voltage over-limit during the voltage over-limit duration;

[0011] S5, determining whether the voltage prediction value of the node with the most serious voltage over-limit duration is less than the first set value or greater than the second set value, if so, executing step S6, otherwise, returning to step S2;

[0012] S6, receiving a voltage optimization start signal, obtaining reactive voltage sensitivities of the adjustable photovoltaic nodes in the cluster to which the node with the most serious voltage over-limit belongs and the node with the most serious voltage over-limit according to the reactive voltage sensitivity between the nodes in the distribution network, and adjusting the reactive power of the node with the most serious voltage over-limit in sequence by using the adjustable photovoltaic nodes with the largest to smallest reactive sensitivities;

[0013] S7. Determine whether the reactive power of the adjustable photovoltaic node in the cluster to which the node with the most serious voltage over-limit belongs is exhausted. If so, map the node with the most serious voltage over-limit to the boundary node of the adjacent cluster, and use the boundary node of the adjacent cluster as the node with the most serious voltage over-limit, and execute step S6. Otherwise, continue to execute step S6 to realize reactive power regulation of the node with the voltage over-limit.

[0014] Furthermore, the reactive voltage sensitivity between nodes in the distribution network in step S1 is:

[0015] S VQ,ij =-[(G ij -P ij )(B ij +Q ij ) -1 (G ij +P ij )+(B ij -Q ij )] -1

[0016] Among them, S VQ,ij represents the reactive voltage sensitivity of node i to node j, G ij represents the line conductance from node i to node j, B ij represents the susceptance from node i to node j, P ij represents the active power flowing on the line from node i to node j, Q ij Represents the reactive power flowing on the line from node i to node j.

[0017] Furthermore, the calculation formula for the electrical distance between nodes in step S1 is:

[0018]

[0019] Where N represents the total number of nodes, L ij represents the electrical distance between node i and node j, D ij represents the sensitivity ratio of node i to node j to the voltage change of node j, lg represents the logarithmic function, S VQ,jj represents the reactive voltage sensitivity of node j to node j, S VQ,ij represents the reactive voltage sensitivity of node i to node j, D i1 It represents the sensitivity ratio of node i to node 1 to the voltage change of node 1, D j1 The sensitivity ratio of node j to node 1 to the voltage change of node 1, D iN It represents the sensitivity ratio of node i to node N to the voltage change of node N, D jN It represents the sensitivity ratio of node j to node N to the voltage change of node N.

[0020] Furthermore, step S2 specifically includes:

[0021] S21, set the total number of clusters to K, the total number of iterations to T, randomly select K nodes as cluster centers and initialize them;

[0022] S22, comparing the electrical distances from each node to the center of each cluster in turn, and grouping each node into the cluster closest to the center of each cluster, to obtain K clusters;

[0023] S23. Randomly reselect the cluster center based on all the nodes of the K clusters, repeat step S22, and determine whether the total number of iterations has been reached. If so, the cluster of the distribution network is obtained. Otherwise, continue to execute step S22.

[0024] Furthermore, in step S3, the threshold is set to 3 minutes.

[0025] Furthermore, the triple exponential smoothing value of the node voltage with the most serious voltage over-limit in step S4 is:

[0026]

[0027] Among them, W t (1) , W t (2) , W t (3) They represent the first, second, and third exponential smoothing values ​​of the node voltage with the most serious voltage over-limit at time t, They represent the first, second, and third exponential smoothing values ​​of the node voltage with the most serious voltage over-limit at time t, α represents the smoothing coefficient, V t Indicates the voltage amplitude of the node with the most serious voltage limit violation at time t.

[0028] Furthermore, the calculation formula for the voltage prediction value of the node with the most serious voltage over-limit during the voltage over-limit duration in step S4 is:

[0029]

[0030] Where k represents the duration of voltage exceeding the limit, V t+k A represents the voltage prediction value of the node with the most serious voltage over-limit at time t+k, that is, the voltage amplitude of the node with the most serious voltage over-limit during the duration of voltage over-limit, t , B t , C t Both represent intermediate variables.

[0031] Furthermore, in step S5, the first set value and the second set value are both voltage optimization start thresholds.

[0032] Furthermore, in step S7, the mapping formula for mapping the node with the most serious voltage over-limit to the boundary node of the adjacent cluster is:

[0033]

[0034] Where ΔV represents the voltage difference that needs to be adjusted at the node with the most serious voltage over-limit, V obj,a Indicates the target voltage of node a, which has the most serious voltage over-limit, V obj,b represents the target voltage mapped from the boundary node b in the adjacent cluster, k ′ 、i ′ All represent the nodes between node a with the most serious voltage over-limit and node b on the border of the adjacent cluster, and n represents the nodes between node a and node k. ′ All nodes connected to the node, x, r represent the nodes connected to k ′ The resistance and reactance of the branches connected to the node, Respectively represent k ′ The equivalent active power and reactive power of the node, V i′-1 Represents node i ′ -1 voltage.

[0035] The present invention has the following beneficial effects:

[0036] The invention proposes a distribution network partition voltage optimization method based on distributed photovoltaic clusters, which can effectively utilize reactive power resources and effectively suppress voltage over-limit problems, thereby realizing voltage optimization. It can also retain more reactive power reserves while optimizing network losses and significantly reduce the voltage control time during voltage optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A schematic diagram of a flow chart of a distribution network partition voltage optimization method based on a distributed photovoltaic cluster proposed in the present invention;

[0038] Figure 2 This is a schematic diagram of the cluster division result of the IEEE33 standard distribution network nodes in the embodiment;

[0039] Figure 3 A schematic diagram showing a comparison between a cubic exponential smoothing value of a node voltage at which the voltage exceeds the limit and an actual voltage value in an embodiment;

[0040] Figure 4 It is a schematic diagram of the change of the voltage of the node 18 and the reactive power margin in the cluster with the control times in the embodiment;

[0041] Figure 5 Schematic diagram of reactive power and voltage changes of cluster B during inter-cluster collaborative control in the embodiment;

[0042] Figure 6 Schematic diagram of voltage comparison after intra-cluster voltage control and inter-cluster reactive power support in the embodiment. DETAILED DESCRIPTION

[0043] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0044] like Figure 1 As shown, a distribution network partition voltage optimization method based on a distributed photovoltaic cluster includes the following steps S1-S7:

[0045] S1. Obtain the reactive voltage sensitivity between nodes in the distribution network and calculate the electrical distance between nodes.

[0046] In this embodiment, the purpose of calculating the electrical distance between each node is to facilitate clustering of each node in subsequent steps.

[0047] Specifically, the reactive voltage sensitivity between nodes in the distribution network in step S1 is:

[0048] S VQ,ij =-[(G ij -P ij )(B ij +Q ij ) -1 (G ij +P ij )+(B ij -Q ij )] -1

[0049] Among them, S VQ,ij represents the reactive voltage sensitivity of node i to node j, G ij represents the line conductance from node i to node j, B ij represents the susceptance from node i to node j, P ij represents the active power flowing on the line from node i to node j, Q ij Represents the reactive power flowing on the line from node i to node j.

[0050] Specifically, the calculation formula of the electrical distance between each node in step S1 is:

[0051]

[0052] Where N represents the total number of nodes, L ij represents the electrical distance between node i and node j, D ij represents the sensitivity ratio of node i to node j to the voltage change of node j, lg represents the logarithmic function, S VQ,jj represents the reactive voltage sensitivity of node j to node j, S VQ,ij represents the reactive voltage sensitivity of node i to node j, D i1 It represents the sensitivity ratio of node i to node 1 to the voltage change of node 1, D j1 The sensitivity ratio of node j to node 1 to the voltage change of node 1, D iN It represents the sensitivity ratio of node i to node N to the voltage change of node N, D jN It represents the sensitivity ratio of node j to node N to the voltage change of node N.

[0053] S2. According to the electrical distance between each node, the distribution network is clustered to obtain the clusters of the distribution network.

[0054] In this embodiment, the purpose of clustering the distribution network is to overcome the inherent communication bottleneck and computational burden limitations of the centralized control method currently used for voltage regulation in the distribution network.

[0055] Specifically, step S2 specifically includes S21-S23:

[0056] S21. Set the total number of clusters to K and the total number of iterations to T, randomly select K nodes as cluster centers and initialize them.

[0057] In this embodiment, T is greater than 50 times to ensure the correctness of cluster selection.

[0058] S22. Compare the electrical distances from each node to the center of each cluster in turn, and group each node into the cluster closest to the center of each cluster, to obtain K clusters.

[0059] S23. Randomly reselect the cluster center based on all the nodes of the K clusters, repeat step S22, and determine whether the total number of iterations has been reached. If so, the cluster of the distribution network is obtained. Otherwise, continue to execute step S22.

[0060] In this embodiment, taking the IEEE33 standard distribution network node as an example, the reference voltage is 12.66kV, the total photovoltaic access capacity is 5600kVA, and there are 13 350kVA photovoltaic units in total. The cluster division results are as follows: Figure 2 As shown, cluster A is {1-2, 19-22}, cluster B is {3-4, 23-25}, and cluster C is {6-18, 26-33}.

[0061] S3, determine whether there is voltage over-limit and the duration of voltage over-limit exceeds the set threshold. If so, execute step S4; otherwise, return to step S2.

[0062] Specifically, in step S3, the threshold is set to 3 minutes.

[0063] In this embodiment, in order to ensure the safety of power grid operation, it is necessary to detect voltage over-limit; at the same time, it is not necessary to immediately adjust reactive equipment when voltage over-limit occurs, that is, in the actual operation of the power system, while ensuring safety, it is also necessary to take into account economy. Therefore, voltage control is required only when the voltage over-limit lasts for a certain duration, that is, when the voltage over-limit duration exceeds 3 minutes, it can be judged by calculating the voltage amplitude after three smoothings to start reactive compensation, that is, execute step S4.

[0064] S4. Sort the voltages of all nodes whose voltage over-limit duration exceeds the set threshold from small to large, obtain the node with the most serious voltage over-limit, obtain the cluster to which the node with the most serious voltage over-limit belongs and the numbers of all nodes in the cluster, use the cubic exponential smoothing method to predict the voltage over-limit duration of the node with the most serious voltage over-limit and the cubic exponential smoothing value of the voltage of the node with the most serious voltage over-limit, and calculate the voltage prediction value of the node with the most serious voltage over-limit during the voltage over-limit duration.

[0065] Specifically, the triple exponential smoothing value of the node voltage with the most serious voltage over-limit in step S4 is:

[0066]

[0067] Among them, W t (1) , W t (2) , W t (3) They represent the first, second, and third exponential smoothing values ​​of the node voltage with the most serious voltage over-limit at time t, They represent the first, second, and third exponential smoothing values ​​of the node voltage with the most serious voltage over-limit at time t, α represents the smoothing coefficient, V t Indicates the voltage amplitude of the node with the most serious voltage limit violation at time t.

[0068] In this embodiment, the value of α is 0.1.

[0069] Specifically, the calculation formula for the voltage prediction value of the node with the most serious voltage over-limit during the voltage over-limit duration in step S4 is:

[0070]

[0071] Where k represents the duration of voltage exceeding the limit, V t+k A represents the voltage prediction value of the node with the most serious voltage over-limit at time t+k, that is, the voltage amplitude of the node with the most serious voltage over-limit during the duration of voltage over-limit, t , B t , C t They all represent intermediate variables, which means they have no practical meaning.

[0072] In this embodiment, the IEEE33 standard distribution network node is still taken as an example. According to the real-time load and photovoltaic data provided in Table 1, it is found that at 18:57, the voltage of node 12 to node 18 exceeds the limit. At the same time, according to the real-time data provided by the distribution network measurement and the previous historical data, it is predicted that this limit exceeding will last for more than 3 minutes. At this time, the voltage autonomy strategy in the cluster is started, that is, the triple exponential smoothing method is used to predict the triple exponential smoothing value of the node voltage where the voltage exceeds the limit. The triple exponential smoothing result is as follows: Figure 3 As shown. Among them, Table 1 is as follows:

[0073] Table 1 Real-time load and photovoltaic data table of IEEE33 standard distribution network nodes

[0074]

[0075]

[0076] S5. Determine whether the voltage prediction value of the node with the most serious voltage over-limit during the voltage over-limit duration is less than the first set value or greater than the second set value. If so, execute step S6; otherwise, return to step S2.

[0077] In this embodiment, the first set value is 0.95 pu, and the second set value is 1.05 pu.

[0078] Specifically, in step S5, the first set value and the second set value are both voltage optimization start thresholds.

[0079] S6. Receive a voltage optimization start signal, and obtain the reactive voltage sensitivity of the adjustable photovoltaic node in the cluster to which the node with the most serious voltage over-limit belongs and the node with the most serious voltage over-limit according to the reactive voltage sensitivity between the nodes in the distribution network, and adjust the reactive power of the node with the most serious voltage over-limit in sequence using the adjustable photovoltaic nodes with reactive sensitivity from large to small.

[0080] In this embodiment, voltage exceeding the limit is avoided by adjusting the reactive voltage.

[0081] S7. Determine whether the reactive power of the adjustable photovoltaic node in the cluster to which the node with the most serious voltage over-limit belongs is exhausted. If so, map the node with the most serious voltage over-limit to the boundary node of the adjacent cluster, and use the boundary node of the adjacent cluster as the node with the most serious voltage over-limit, and execute step S6. Otherwise, continue to execute step S6 to realize reactive power regulation of the node with the voltage over-limit.

[0082] In this embodiment, by detecting the reactive surplus capacity inside the cluster, it is decided whether reactive power needs to be used from adjacent clusters to suppress the voltage exceeding the limit inside the cluster.

[0083] Specifically, in step S7, the mapping formula for mapping the node with the most serious voltage over-limit to the boundary node of the adjacent cluster is:

[0084]

[0085] Where ΔV represents the voltage difference that needs to be adjusted at the node with the most serious voltage over-limit, V obj,a Indicates the target voltage of node a, which has the most serious voltage over-limit, V obj,b represents the target voltage mapped from the boundary node b in the adjacent cluster, k ′ 、i ′ All represent the nodes between node a with the most serious voltage over-limit and node b on the border of the adjacent cluster, and n represents the nodes between node a and node k. ′ All nodes connected to the node, x, r represent the nodes connected to k ′ The resistance and reactance of the branches connected to the node, Respectively represent k ′The equivalent active power and reactive power of the node, V i′-1 Represents node i ′ -1 voltage.

[0086] In this embodiment, taking the IEEE33 standard distribution network node as an example, it is found that the voltage of node 18 is the most serious, and the method proposed in step S5 is used for adjustment, as shown in detail. Figure 4 As shown in FIG. 1 , the variation process of the voltage of node 18 and the reactive power margin in the cluster with the number of control times is shown. It can be seen that when the autonomous method within the cluster is adopted, the photovoltaic nodes 18, 17, 13 and 11, which are closer to the node with the most serious limit violation, successively undertake the voltage regulation task until the power is exhausted after 6 controls. At the same time, the photovoltaic nodes 9, 28, 31 and 33, which are farther away, are exhausted after 14 controls. Therefore, it is necessary to request reactive power support from the adjacent cluster (cluster B), that is, to execute the operation of step S6, as shown in FIG. Figure 5 As shown in the figure, the reactive power and voltage changes of cluster B during inter-cluster collaborative control are shown. When cluster C is insufficient in reactive power, the reactive power support task is transferred to cluster B. It is calculated that cluster B needs to adjust the voltage of boundary node 5 to 1.0086pu. At this time, the problem of reactive power support between clusters is transformed into an intra-cluster autonomy problem based on cluster B, thereby starting step S5, that is, the node that needs to be adjusted in cluster B is node 5, and the photovoltaics at nodes 4, 25 and 24 provide reactive power support for it in turn, and finally complete the reactive power support between clusters. In addition, Figure 6 The voltage changes of each node before and after cluster control are shown, from which it can be seen that the method proposed in the present invention can effectively suppress the voltage over-limit problem, thereby achieving voltage optimization.

[0087] At the same time, the reactive power margin and control duration under centralized control obtained by the traditional centralized control method are compared with the reactive power margin and control duration under partition control obtained by the method proposed in the present invention, as shown in Table 2:

[0088] Table 2 Comparison of reactive power margin and control time under centralized control and partition control mode

[0089]

[0090] It can be seen from Table 2 that the partition control method proposed in the present invention can utilize reactive resources more effectively and retain more reactive reserves while optimizing network losses. At the same time, the control time of the partition control method is only 1.682 seconds, while the control time of the centralized control method is 3.237 seconds, indicating that the method proposed in the present invention can significantly reduce the control time.

[0091] In summary, the present invention proposes a distribution network partition voltage optimization method based on a distributed photovoltaic cluster. According to the electrical distance between nodes and the voltage control capability of the cluster, a distributed photovoltaic cluster (cluster of the distribution network) with local autonomy is constructed, and the photovoltaic cluster is used as the basic control unit for voltage optimization. When the reactive power in the cluster is sufficient, based on the regional reactive power-voltage sensitivity of the cluster to which the voltage over-limit node belongs, the reactive power support of the distributed photovoltaic in the cluster is calculated without the need for inter-cluster communication. When the reactive power in the cluster is insufficient, the optimization target of the voltage over-limit node of the original cluster is mapped to the boundary node of the adjacent cluster, and only the information of the boundary node is exchanged with the adjacent cluster. By adjusting the voltage of the boundary node of the adjacent cluster, the voltage over-limit is indirectly suppressed. Reactive resources can be used more effectively and the voltage over-limit problem can be effectively suppressed, thereby achieving voltage optimization. While optimizing network loss, more reactive power reserves can be retained, and the voltage control time in the voltage optimization process can be significantly reduced.

[0092] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

[0093] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. A distribution network partition voltage optimization method based on distributed photovoltaic clusters, characterized in that: The following steps are involved: S1, obtain the reactive voltage sensitivity between each node in the distribution network and calculate the electrical distance between each node; S2, clustering the distribution network according to the electrical distance between each node to obtain the cluster of the distribution network; S3, determine whether there is a voltage over-limit and the duration of the voltage over-limit exceeds the set threshold. If so, execute step S4, otherwise, return to step S2; S4. Sort the voltages of all nodes whose voltage over-limit duration exceeds the set threshold from small to large, obtain the node with the most serious voltage over-limit, obtain the cluster to which the node with the most serious voltage over-limit belongs and the numbers of all nodes in the cluster, use the cubic exponential smoothing method to predict the voltage over-limit duration of the node with the most serious voltage over-limit and the cubic exponential smoothing value of the voltage of the node with the most serious voltage over-limit, and calculate the voltage prediction value of the node with the most serious voltage over-limit during the voltage over-limit duration; S5, determining whether the voltage prediction value of the node with the most serious voltage over-limit duration is less than the first set value or greater than the second set value, if so, executing step S6, otherwise, returning to step S2; S6, receiving a voltage optimization start signal, obtaining reactive voltage sensitivities of the adjustable photovoltaic nodes in the cluster to which the node with the most serious voltage over-limit belongs and the node with the most serious voltage over-limit according to the reactive voltage sensitivity between the nodes in the distribution network, and adjusting the reactive power of the node with the most serious voltage over-limit in sequence by using the adjustable photovoltaic nodes with the largest to smallest reactive sensitivities; S7. Determine whether the reactive power of the adjustable photovoltaic node in the cluster to which the node with the most serious voltage over-limit belongs is exhausted. If so, map the node with the most serious voltage over-limit to the boundary node of the adjacent cluster, and use the boundary node of the adjacent cluster as the node with the most serious voltage over-limit, and execute step S6. Otherwise, continue to execute step S6 to realize reactive power regulation of the node with the voltage over-limit.

2. The method for optimizing voltage in distribution network partitions based on distributed photovoltaic clusters according to claim 1 is characterized in that: The reactive voltage sensitivity between nodes in the distribution network in step S1 is: S VQ,ij =-[(G ij -P ij )(B ij +Q ij ) -1 (G ij +P ij )+(B ij -Q ij )] -1 Among them, S VQ,ij represents the reactive voltage sensitivity of node i to node j, G ij represents the line conductance from node i to node j, B ij represents the susceptance from node i to node j, P ij represents the active power flowing on the line from node i to node j, Q ij Represents the reactive power flowing on the line from node i to node j.

3. The method for optimizing voltage in distribution network partitions based on distributed photovoltaic clusters according to claim 2 is characterized in that: The calculation formula of the electrical distance between each node in step S1 is: Where N represents the total number of nodes, L ij represents the electrical distance between node i and node j, D ij represents the sensitivity ratio of node i to node j to the voltage change of node j, lg represents the logarithmic function, S VQ,jj represents the reactive voltage sensitivity of node j to node j, S VQ,ij represents the reactive voltage sensitivity of node i to node j, D i1 It represents the sensitivity ratio of node i to node 1 to the voltage change of node 1, D j1 It represents the sensitivity ratio of node j to node 1 to the voltage change of node 1, D iN It represents the sensitivity ratio of node i and node N to the voltage change of node N, D jN It represents the sensitivity ratio of node j to node N to the voltage change of node N.

4. The method for optimizing voltage in distribution network partitions based on distributed photovoltaic clusters according to claim 3 is characterized in that: Step S2 specifically includes: S21, set the total number of clusters to K, the total number of iterations to T, randomly select K nodes as cluster centers and initialize them; S22, comparing the electrical distances from each node to the center of each cluster in turn, and grouping each node into the cluster closest to the center of each cluster, to obtain K clusters; S23. Randomly reselect the cluster center based on all the nodes of the K clusters, repeat step S22, and determine whether the total number of iterations has been reached. If so, the cluster of the distribution network is obtained. Otherwise, continue to execute step S22.

5. The method for optimizing voltage in distribution network partitions based on distributed photovoltaic clusters according to claim 4 is characterized in that: In step S3, the threshold is set to 3 minutes.

6. The method for optimizing voltage in distribution network partitions based on distributed photovoltaic clusters according to claim 5, characterized in that: The triple exponential smoothing value of the node voltage with the most serious voltage over-limit in step S4 is: Among them, W t (1) , W t (2) , W t (3) They represent the first, second, and third exponential smoothing values ​​of the node voltage with the most serious voltage over-limit at time t, They represent the first, second, and third exponential smoothing values ​​of the node voltage with the most serious voltage over-limit at time t, α represents the smoothing coefficient, V t Indicates the voltage amplitude of the node with the most serious voltage limit violation at time t.

7. The method for optimizing voltage in distribution network partitions based on distributed photovoltaic clusters according to claim 6, characterized in that: The calculation formula for the voltage prediction value of the node with the most serious voltage over-limit in step S4 during the voltage over-limit duration is: Where k represents the duration of voltage exceeding the limit, V t+k A represents the voltage prediction value of the node with the most serious voltage over-limit at time t+k, that is, the voltage amplitude of the node with the most serious voltage over-limit during the duration of voltage over-limit, t , B t , C t Both represent intermediate variables.

8. The method for optimizing voltage in distribution network partitions based on distributed photovoltaic clusters according to claim 7, characterized in that: In step S5, the first set value and the second set value are both voltage optimization start thresholds.

9. The method for optimizing voltage in distribution network partitions based on distributed photovoltaic clusters according to claim 8, characterized in that: In step S7, the mapping formula for mapping the node with the most serious voltage over-limit to the boundary node of the adjacent cluster is: Where ΔV represents the voltage difference that needs to be adjusted at the node with the most serious voltage over-limit, V obj,a Indicates the target voltage of node a, which has the most serious voltage over-limit, V obj,b represents the target voltage mapped from the boundary node b in the adjacent cluster, k ′ 、i ′ All represent the nodes between node a with the most serious voltage over-limit and node b on the border of the adjacent cluster, and n represents the nodes between node a and node k. ′ All nodes connected to the node, x, r represent the nodes connected to k ′ The resistance and reactance of the branches connected to the node, Respectively represent k ′ The equivalent active power and reactive power of the node, V i′-1 Represents node i ′ -1 voltage.

Images