Power distribution network multi-target cluster division method and system suitable for risk prevention and control, computer equipment and storage medium

Through the multi-objective cluster division method and K-means discrete particle swarm algorithm, the distribution network parameters are optimized, the voltage problem caused by distributed photovoltaic access is solved, the control efficiency and internal adjustable capabilities of the cluster are improved, and effective voltage management is achieved .

CN119051036APending Publication Date: 2024-11-29STATE GRID LIAONING ECONOMIC TECHN INST +5
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411010437.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-26
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

The existing centralized methods cannot effectively solve the problems of increased voltage of some nodes, excessive line end voltage, low control efficiency, large calculation volume, and communication delay caused by large-scale distributed photovoltaic access.

Method used

Using the multi-objective cluster division method, the weighting sum model of comprehensive electrical distance, comprehensive reserve degree, strong coupling within the cluster, and weak coupling between clusters based on voltage sensitivity is optimized. Network parameters and adjustable resource information, solve the multi-objective cluster division model, and realize cluster division.

Benefits of technology

It effectively solves the voltage problem caused by large-scale access to distributed photovoltaics, improves control efficiency, reduces calculation amount and communication delay, and improves the adjustable capability and control effect within the cluster.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119051036A_ABST
    Figure CN119051036A_ABST
Patent Text Reader

Abstract

The invention discloses a power distribution network multi-target cluster division method suitable for risk prevention and control, and the method comprises the steps: setting a target function of a multi-target cluster division model, the target function being a weighted sum of multiple targets, such as a comprehensive electrical distance, a comprehensive reserve degree, strong coupling in clusters, weak coupling between clusters, and the like, based on voltage sensitivity; establishing constraint conditions of a multi-target cluster division model, wherein the constraint conditions comprise constraint through the minimum number of nodes in the cluster and constraint through the number of adjustable resource nodes in the cluster; based on the target function and the constraint condition of the multi-target cluster division model, constructing the multi-target cluster division model; and solving the multi-target cluster division model based on a discrete particle swarm cluster division algorithm of K-means. According to the method, the problems that the voltage of part of nodes rises, the voltage at the tail end of a line exceeds the standard, direct regulation and control are difficult, the control efficiency is low, the calculated amount is large, and communication is delayed due to the fact that an existing centralized method cannot meet the requirement for massive distributed photovoltaic access can be solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of power system dispatching, and in particular to a distribution network multi-objective cluster division method, division system, computer equipment and storage medium suitable for risk prevention and control. Background Art

[0002] Large-scale distributed photovoltaic access has changed the distribution characteristics of the traditional distribution network flow, with the characteristics of "bidirectional flow" and "active distribution network", which has led to problems such as voltage increase at some nodes and voltage exceeding the standard at the end of the line. However, distributed resources are large in quantity and dispersed, with strong volatility, making them difficult to directly regulate. Traditional centralized control methods are prone to problems such as low control efficiency, large amount of calculation, and communication delay, and can no longer effectively solve the voltage limit problem caused by large-scale distributed photovoltaic access.

[0003] Therefore, a cluster partitioning method is needed to deal with the problem that distributed resources are large, dispersed, volatile, and difficult to directly control. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the problem that the present invention seeks to solve is that the current centralized method cannot meet the problems caused by the large-scale access of distributed photovoltaics, such as the increase in voltage at some nodes, the excessive voltage at the end of the line, the difficulty in direct regulation, the low control efficiency, the large amount of calculation, and the communication delay.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a distribution network multi-objective clustering method suitable for risk prevention and control, comprising:

[0008] Setting an objective function of the multi-objective cluster partitioning model, wherein the objective function is a weighted sum of multiple objectives including a comprehensive electrical distance based on voltage sensitivity, a comprehensive reserve degree, strong coupling within a cluster, and weak coupling between clusters;

[0009] Establishing constraint conditions of the multi-objective cluster partitioning model, the constraint conditions including a minimum number of nodes in the cluster and a number of adjustable resource nodes in the cluster;

[0010] Based on the objective function and constraint conditions of the multi-objective cluster partitioning model, a multi-objective cluster partitioning model is constructed, a discrete particle swarm cluster partitioning algorithm based on K-means is adopted, distribution network parameters and adjustable resource information in the multi-objective cluster partitioning model are input, the multi-objective cluster partitioning model is solved, and a partitioning result is obtained.

[0011] As a preferred solution of the multi-objective cluster division method for distribution network suitable for risk prevention and control described in the present invention, the objective function is a weighted sum of indicators such as comprehensive electrical distance, comprehensive reserve degree, strong coupling within the cluster, and weak coupling between clusters based on voltage sensitivity, and is expressed as:

[0012]

[0013] Among them, F is the overall evaluation index, D k , R k , C h , C l are the normalized maximum values ​​of comprehensive electrical distance, comprehensive reserve, strong coupling within clusters, weak coupling between clusters, etc.; ζ d , r , ch and cl are the weight coefficients of the four evaluation indicators, which are determined by combining subjective weighting and objective weighting.

[0014] As a preferred solution of the multi-objective cluster division method of the distribution network suitable for risk prevention and control described in the present invention, wherein: the comprehensive electrical distance based on voltage sensitivity specifically includes:

[0015] The changes in voltage amplitude V and phase angle θ (ΔV, Δθ) and the changes in active power P and reactive power Q (ΔP, ΔQ) can be related through formula (2). The corresponding sensitivity calculation formula is shown in formula (3), which is the corresponding item of the Jacobian matrix of the system power flow equation;

[0016]

[0017] In the formula, S θP and S θQ Represent the phase angle sensitivity matrix related to active power and reactive power respectively; S VP and S VQ Represent the voltage amplitude sensitivity matrix related to active and reactive power respectively;

[0018] Under the influence of reactive power, the voltage change relationship between any two points defined by voltage sensitivity is:

[0019]

[0020] Where ΔV i Q , ΔV j Q are the voltage changes of node i and node j under the influence of reactive power; ΔQ j is the reactive variable of node j, is the reactive voltage change sensitivity of node i to node j. Similarly, the active voltage change sensitivity of node i to node j is derived. As shown in formula (5):

[0021]

[0022] Based on reactive voltage sensitivity and active voltage sensitivity The expression of the electrical distance between nodes can be constructed to define the comprehensive electrical distance d between any two nodes. ij for:

[0023]

[0024] Where, d ij is the electrical distance between nodes i and j. The greater the voltage sensitivity between nodes i and j, the smaller the electrical distance. According to formula (6), the comprehensive electrical distance D of cluster k can be obtained: k for:

[0025]

[0026] As a preferred solution of the multi-objective cluster division method of the distribution network suitable for risk prevention and control described in the present invention, the comprehensive reserve degree is defined as follows:

[0027]

[0028] Where U MK Control voltage setpoint for the leading node of cluster k; are the voltage values ​​of the dominant nodes when all distributed generation nodes in cluster k operate at the upper limit of feasible active / reactive power and the lower limit of feasible active / reactive power respectively; are the upward and downward adjustable capacities of the leading node of cluster k, respectively;

[0029] Comprehensive reserve ratio R k The smaller the value, the stronger the adjustable capability of the cluster. For a cluster without adjustable resources, since it has no adjustable capability, the comprehensive reserve is 1.

[0030] As a preferred solution of the multi-objective clustering method for distribution network suitable for risk prevention and control described in the present invention, wherein: the strong coupling within the cluster and the weak coupling between clusters specifically include:

[0031] Introducing strong and weak coupling indicators within the cluster To characterize the degree of coupling between nodes in the region, the calculation method is shown in equations (11)-(13):

[0032]

[0033]

[0034] In the formula, is the average electrical distance between the adjustable resource nodes and the load nodes in cluster k; N kg 、N kl are the sets of adjustable resources and pure load nodes in cluster k; It is a strong coupling index within the cluster. Its physical meaning is the coupling relationship between the adjustable resource nodes within cluster k and the load nodes within its own cluster. The smaller the value, the stronger the coupling of nodes in the cluster and the closer the connection. is the weak coupling index between clusters, and its physical meaning is the coupling relationship between the adjustable resource nodes in cluster k and the load nodes outside the cluster. The smaller it is, the weaker the coupling between clusters.

[0035] As a preferred solution of the multi-objective cluster division method for distribution network applicable to risk prevention and control described in the present invention, wherein: establishing a constraint on the minimum number of nodes in the cluster specifically includes:

[0036] N kg +N kl ≥2 (14)

[0037] Formula (14) indicates that a single node cannot form a cluster;

[0038] As a preferred solution of the multi-objective cluster division method for distribution network applicable to risk prevention and control described in the present invention, wherein: establishing a constraint on the number of adjustable resource nodes in the cluster specifically includes:

[0039] N kl ≥1 (15)

[0040] Formula (15) indicates that a single node cannot form a cluster;

[0041] In a second aspect, the present invention provides a distribution network multi-objective cluster division system suitable for risk prevention and control, which includes:

[0042] The objective function building module of the multi-objective cluster partitioning model is used to obtain the weighted sum of the comprehensive electrical distance, comprehensive reserve degree, strong coupling within the cluster, and weak coupling between clusters based on voltage sensitivity;

[0043] The constraint condition establishment module of the multi-objective cluster partitioning model is used to establish the minimum number of nodes in the cluster and the number of adjustable resource nodes in the cluster;

[0044] The multi-objective clustering model building module is used to establish a discrete particle swarm clustering algorithm based on K-means, integrate the objective function and constraint conditions of the multi-objective clustering model, and solve the multi-objective clustering model.

[0045] As a preferred solution of the multi-objective clustering system for distribution network suitable for risk prevention and control described in the present invention, in the multi-objective clustering model establishment module, the particle positions are first initialized based on the K-means clustering algorithm, and then the discrete particle swarm optimization algorithm is used for global iterative optimization.

[0046] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the processor executes the computer program, it implements any step of the above-mentioned distribution network multi-objective cluster division method suitable for risk prevention and control.

[0047] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the above-mentioned distribution network multi-objective cluster division method suitable for risk prevention and control is implemented.

[0048] Beneficial effects of the present invention: The present invention constructs a weighted sum model of multiple objectives such as comprehensive electrical distance based on voltage sensitivity, comprehensive reserve degree, strong coupling within the cluster, weak coupling between clusters, etc. as a distribution network multi-objective cluster division model suitable for risk prevention and control, and considers the constraints of the minimum number of nodes in the cluster and the number of adjustable resource nodes in the cluster, and finally determines the multi-objective cluster division scheme based on a pilot scenario of distributed photovoltaic development on the roof of a whole county. It can solve the problems of voltage increase of some nodes, voltage exceeding the standard at the end of the line, difficulty in direct regulation, low control efficiency, large amount of calculation, communication delay, etc. caused by the current centralized method being unable to meet the large-scale access of distributed photovoltaics. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0050] Figure 1 A schematic diagram of a K-means-based discrete particle swarm optimization algorithm flow chart of a multi-objective clustering method for distribution network suitable for risk prevention and control according to an embodiment of the present invention;

[0051] Figure 2A schematic diagram of a pilot scenario calculation example of a distributed photovoltaic development on rooftops of a whole county for a multi-objective cluster division method of a distribution network suitable for risk prevention and control according to an embodiment of the present invention;

[0052] Figure 3 A cluster partitioning diagram of a K-means-based discrete particle swarm algorithm for a multi-objective cluster partitioning method for distribution network suitable for risk prevention and control according to an embodiment of the present invention;

[0053] Figure 4 A cluster partitioning diagram based on an electrical distance algorithm is provided for comparing a method for multi-objective cluster partitioning of a distribution network suitable for risk prevention and control according to an embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0055] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0056] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0057] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0058] At the same time, in the description of the present invention, it should be noted that the orientations or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the system or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0059] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0060] Example 1

[0061] Reference Figure 1 , is an embodiment of the present invention, and provides a distribution network multi-objective cluster division method suitable for risk prevention and control, comprising:

[0062] S1: Set the objective function of the multi-objective cluster partitioning model, which is the weighted sum of multiple objectives such as comprehensive electrical distance based on voltage sensitivity, comprehensive reserve, strong coupling within the cluster, and weak coupling between clusters;

[0063] Furthermore, the objective function is a weighted sum of multiple objectives such as comprehensive electrical distance based on voltage sensitivity, comprehensive reserve, strong coupling within the cluster, and weak coupling between clusters:

[0064]

[0065] Among them, F is the overall evaluation index, D k , R k , C h , C l are the normalized maximum values ​​of comprehensive electrical distance, comprehensive reserve, strong coupling within clusters, weak coupling between clusters, etc.; ζ d , r , ch and cl They are the weight coefficients of the four evaluation indicators, which are reasonably determined by combining subjective weighting and objective weighting.

[0066] Furthermore, the comprehensive electrical distance based on voltage sensitivity includes:

[0067] The comprehensive electrical distance based on voltage sensitivity describes the closeness of the electrical connection between any two nodes in the system. The change in voltage amplitude V and phase angle θ (ΔV, Δθ) and the change in active power P and reactive power Q (ΔP, ΔQ) can be linked by equation (2). The corresponding sensitivity calculation formula is shown in equation (3), which is the corresponding term of the Jacobian matrix of the system power flow equation.

[0068]

[0069] In the formula, S θP and S θQ Represent the phase angle sensitivity matrix related to active power and reactive power respectively; S VP and S VQ Represent the voltage amplitude sensitivity matrix related to active and reactive power respectively.

[0070] The R / X ratio of the distribution network is relatively large, and the change in node voltage is related to both active and reactive power. First, the voltage change relationship between any two points defined by voltage sensitivity under the influence of reactive power is analyzed:

[0071]

[0072] Where ΔV i Q , ΔV j Q are the voltage changes of node i and node j under the influence of reactive power; ΔQ j is the reactive variable of node j. is the sensitivity of node i to the reactive voltage change of node j. Similarly, the sensitivity of node i to the active voltage change of node j is derived As shown in formula (5).

[0073]

[0074] Based on reactive voltage sensitivity and active voltage sensitivity An expression for the electrical distance between nodes can be constructed. Since the electrical distance represents the coupling relationship between two nodes in the system, it needs to be strictly symmetrical. and and The weights are the same, defining the comprehensive electrical distance d between any two nodes ij for

[0075]

[0076] Where, d ijis the electrical distance between nodes i and j. The greater the voltage sensitivity between nodes i and j, the smaller the electrical distance. According to formula (6), the comprehensive electrical distance D of cluster k can be obtained: k for

[0077]

[0078] Furthermore, the comprehensive reserve ratio includes:

[0079] In order to effectively control the cluster voltage, it is necessary to ensure that there is sufficient power reserve in the cluster, that is, there must be sufficient adjustable capacity. The comprehensive reserve is defined as follows:

[0080]

[0081] Where U MK Control voltage setpoint for the leading node of cluster k; are the voltage values ​​of the dominant nodes when all distributed generation nodes in cluster k operate at the upper limit of feasible active / reactive power and the lower limit of feasible active / reactive power respectively; are the upward and downward adjustable capacities of the leading node in cluster k, respectively.

[0082] Comprehensive reserve ratio R k The smaller the value, the stronger the adjustable capability of the cluster. For a cluster without adjustable resources, since it has no adjustable capability, the comprehensive reserve is 1.

[0083] Furthermore, the comprehensive reserve ratio includes:

[0084] In order to improve the voltage control effect within the cluster and avoid the control within the cluster affecting other external clusters, it is necessary to ensure that the internal coupling of the cluster is large and the coupling between clusters is small. To characterize the degree of coupling between nodes in the region, the calculation method is shown in equations (11)-(13).

[0085]

[0086] In the formula, is the average electrical distance between the adjustable resource nodes and the load nodes in cluster k; N kg 、N kl are the sets of adjustable resources and pure load nodes in cluster k; It is a strong coupling index within the cluster. Its physical meaning is the coupling relationship between the adjustable resource nodes within cluster k and the load nodes within its own cluster. The smaller the value, the stronger the coupling of nodes in the cluster and the closer the connection. is the weak coupling index between clusters, and its physical meaning is the coupling relationship between the adjustable resource nodes in cluster k and the load nodes outside the cluster. The smaller it is, the weaker the coupling between clusters.

[0087] S2: Establish the constraints of the multi-objective cluster partitioning model, including the minimum number of nodes in the cluster and the number of adjustable resource nodes in the cluster;

[0088] Furthermore, we establish the minimum number of nodes in the cluster, including:

[0089] N kg +N kl ≥2 (14)

[0090] Formula (14) indicates that a single node cannot form a cluster;

[0091] Furthermore, we establish the constraints on the number of adjustable resource nodes in the cluster, including:

[0092] N kl ≥1 (15)

[0093] Formula (15) indicates that a single node cannot form a cluster;

[0094] S3: Based on the objective function and constraints of the multi-objective clustering model, a multi-objective clustering model is constructed.

[0095] After defining the objective function and constraints, the multi-objective clustering model of the distribution network suitable for risk prevention and control is constructed as follows:

[0096]

[0097] st(14)-(15)

[0098] The multi-objective clustering model is solved by a K-means-based discrete particle swarm clustering algorithm.

[0099] First, the particle positions are initialized based on the K-means clustering algorithm, and then the discrete particle swarm optimization algorithm is used for global iterative optimization. In this way, the position after initialization is closer to the global optimal position, which simplifies the complex dynamic inertia weight update problem at the algorithm level, and can effectively improve the algorithm's optimization ability, convergence speed and stability, and avoid falling into the local optimum.

[0100] The Kmeans-DPSO algorithm optimizes the partitioning steps as follows:

[0101] 1) Input distribution network parameters and adjustable resource information, and calculate data such as electrical distance and active and reactive power reserves based on voltage sensitivity.

[0102] 2) Perform K-means clustering partitioning according to the lower triangular matrix of electrical distance as the initial position of each particle and the global initialization partition, and randomly initialize the particle speed.

[0103] 3) Set the maximum number of iterations, the number of particle populations, and give the particle swarm algorithm learning factor, inertia weight and other parameters.

[0104] 4) Calculate the fitness value of the individual multi-objective function of the initialized particle population, and calculate the global optimal value of the entire population after initialization.

[0105] 5) Update the continuous velocity variables and continuous position variables of particles in the population.

[0106] 6) Discretize the corresponding continuous position variables in a specific way to determine the new particle positions and partitions.

[0107] 7) Calculate the individual multi-objective function fitness value of each particle population after update, and calculate the global optimal value of the entire population.

[0108] 8) Determine whether the global optimal value is less than the initialized particle position objective function value, and decide whether to update the partition based on the global optimal value.

[0109] 9) Determine whether the current number of iterations meets the end condition of the maximum number of iterations. If not, return to step 5; if yes, proceed to the next step.

[0110] 10) Select the dominant node based on the final partition situation and the observability and controllability within the partition.

[0111] 11) Output the partitioning results and the dominant node selection results.

[0112] 12) End.

[0113] Furthermore, this embodiment also provides a distribution network multi-objective cluster division system suitable for risk prevention and control, including:

[0114] The objective function building module of the multi-objective cluster partitioning model is used to obtain the weighted sum of the comprehensive electrical distance, comprehensive reserve degree, strong coupling within the cluster, and weak coupling between clusters based on voltage sensitivity;

[0115] The constraint condition establishment module of the multi-objective cluster partitioning model is used to establish the minimum number of nodes in the cluster and the number of adjustable resource nodes in the cluster;

[0116] The multi-objective clustering model building module is used to establish a discrete particle swarm clustering algorithm based on K-means, integrate the objective function and constraint conditions of the multi-objective clustering model, and solve the multi-objective clustering model.

[0117] This embodiment also provides a computer device, which is suitable for the case of a multi-objective cluster division method for distribution network suitable for risk prevention and control, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the multi-objective cluster division method for distribution network suitable for risk prevention and control as proposed in the above embodiment.

[0118] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0119] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for dividing distribution networks into multi-objective clusters suitable for risk prevention and control proposed in the above embodiment is implemented.

[0120] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be found in the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0121] Example 2

[0122] Reference Figure 2-4 And Tables 1-4 are an embodiment of the present invention, which provides a distribution network multi-objective cluster division method suitable for risk prevention and control. In order to verify its beneficial effects, the beneficial effects are explained by comparing data.

[0123] The main steps are:

[0124] Input the distribution network topology data of a county-wide rooftop distributed photovoltaic development pilot project.

[0125] Enter the distributed photovoltaic node number and distributed photovoltaic installed capacity.

[0126] Enter the number of cluster divisions.

[0127] The K-means-based discrete particle swarm clustering algorithm is called to solve the multi-objective clustering model, and a multi-objective clustering scheme for distribution network suitable for risk prevention and control is obtained.

[0128] Specifically, Figure 2 , taking the pilot scenario of distributed photovoltaic development on the roof of a whole county as the simulation system. This distribution network system contains 50 nodes and 49 branches, which is a typical radial distribution network system. In order to verify the effectiveness of the cluster division method of the present invention, the present invention lists the location and capacity information of distributed power sources based on the actual feeder topology and distributed photovoltaic access ledger, as shown in Table 1.

[0129] Table 1 Capacity and location of distributed power generation in distribution network system

[0130]

[0131] In the example of the present invention, the voltage level is 10kV, and the installation positions of the distributed power supply nodes are nodes 11, 13, 14, 15, 17, 18, 19, 20, 28, 29, 30, 31, 32, 33, 35, 41, and 46. The present invention adopts an empirical method to determine the number of partitions, that is, for a set containing n points, the number of partitions can be set to Therefore, the number of partitions is set to 5. The electrical distance, reserve degree and cluster strong and weak coupling index are calculated using formulas (2)-(13). The Kmeans-DPSO optimization algorithm is used to optimize the partitions. The maximum number of iterations is set to 500. Iterative optimization is performed in the python3.7.11 environment. The system environment is Intel(R) Core(TM) i5-8300H CPU@2.30GHz 8.00GB.

[0132] In order to verify the scientific effectiveness of the K-means-based discrete particle swarm clustering algorithm in solving the multi-objective clustering model, the traditional multi-objective clustering algorithm based on electrical distance is selected for comparison, where the electrical distance can be calculated by formulas (1)-(6). The clustering results of the two algorithms are shown in Table 2 and Figure 3 , Figure 4 shown.

[0133] Table 2 Comparison between the method of the present invention and the traditional cluster division method based on electrical distance

[0134]

[0135] It can be seen from the results that in the cluster division process, the traditional cluster division based on electrical distance only performs cluster division based on the relative electrical distance between each node, without considering the impact of the distributed photovoltaic power regulation capability within the cluster on the regional division. The partition result separates nodes 2 and 3 as a partition, and the partition does not contain photovoltaic nodes, resulting in the partition without a dominant node and uncontrollable partition. Such a partition result is obviously not conducive to partition control and cross-region coordination.

[0136] The effectiveness and superiority of the clustering algorithm proposed in the present invention are further verified by calculating the comprehensive reserve degree, strong coupling within clusters, and weak coupling between clusters between the clustering results of the method of the present invention and the traditional clustering results based on electrical distance. The comparison of clustering indicators of the two methods is shown in Table 3.

[0137] Table 3 Cluster division index values ​​of two algorithms

[0138]

[0139] It can be seen from the results that the comprehensive reserve degree, strong coupling within the cluster, and weak coupling between clusters of the discrete particle swarm clustering algorithm based on K-means in the present invention are all lower than the index values ​​of the traditional clustering algorithm based on electrical distance. This shows that when solving the optimal clustering scheme, the present invention fully considers the regulation effect of distributed power sources, and through multi-objective optimization, each cluster has sufficient reactive and active reserves, the coupling relationship within the cluster is relatively strong, and the coupling relationship with other clusters is relatively weak, which is convenient for partition control and cluster regulation.

[0140] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A multi-objective clustering method for distribution network suitable for risk prevention and control, characterized in that: include: Setting an objective function of the multi-objective cluster partitioning model, wherein the objective function is a weighted sum of multiple objectives including a comprehensive electrical distance based on voltage sensitivity, a comprehensive reserve degree, strong coupling within a cluster, and weak coupling between clusters; Establishing constraint conditions of the multi-objective cluster partitioning model, the constraint conditions including a minimum number of nodes in the cluster and a number of adjustable resource nodes in the cluster; Based on the objective function and constraint conditions of the multi-objective cluster partitioning model, a multi-objective cluster partitioning model is constructed, a discrete particle swarm cluster partitioning algorithm based on K-means is adopted, the distribution network parameters and adjustable resource information in the multi-objective cluster partitioning model are input, the multi-objective cluster partitioning model is solved, and the partitioning result is obtained.

2. The distribution network multi-objective clustering method suitable for risk prevention and control as claimed in claim 1, characterized in that: The objective function is a weighted sum of indicators such as comprehensive electrical distance based on voltage sensitivity, comprehensive reserve, strong coupling within the cluster, and weak coupling between clusters, and is expressed as: Among them, F is the overall evaluation index, D k , R k , C h , C l are the normalized maximum values ​​of comprehensive electrical distance, comprehensive reserve, strong coupling within clusters, weak coupling between clusters, etc.; ζ d , r , ch and cl are the weight coefficients of the four evaluation indicators, which are determined by combining subjective weighting and objective weighting.

3. The distribution network multi-objective clustering method suitable for risk prevention and control as claimed in claim 1, characterized in that: The comprehensive electrical distance based on voltage sensitivity specifically includes: The changes in voltage amplitude V and phase angle θ (ΔV, Δθ) and the changes in active power P and reactive power Q (ΔP, ΔQ) can be related through equation (2). The corresponding sensitivity calculation formula is shown in equation (3), which is the corresponding item of the Jacobian matrix of the system power flow equation: In the formula, S θP and S θQ Represent the phase angle sensitivity matrix related to active power and reactive power respectively; S VP and S VQ Represent the voltage amplitude sensitivity matrix related to active and reactive power respectively; Under the influence of reactive power, the voltage change relationship between any two points defined by voltage sensitivity is: Where, ΔV i Q , ΔV j Q are the voltage changes of node i and node j under the influence of reactive power; ΔQ j is the reactive variable of node j; is the sensitivity of node i to the reactive voltage change of node j; similarly, the sensitivity of node i to the active voltage change of node j is derived As shown in formula (5): Based on reactive voltage sensitivity and active voltage sensitivity The expression of the electrical distance between nodes can be constructed to define the comprehensive electrical distance d between any two nodes. ij for: Where, d ij is the electrical distance between nodes i and j. The greater the voltage sensitivity between nodes i and j, the smaller the electrical distance. According to formula (6), the comprehensive electrical distance D of cluster k can be obtained: k for:

4. The distribution network multi-objective clustering method suitable for risk prevention and control according to claim 1, characterized in that: The comprehensive reserve ratio specifically includes: The comprehensive reserve ratio is defined as follows: Where U MK Control voltage setpoint for the leading node of cluster k; are the voltage values ​​of the dominant nodes when all distributed generation nodes in cluster k operate at the upper limit of feasible active / reactive power and the lower limit of feasible active / reactive power respectively; are the upward and downward adjustable capacities of the leading node of cluster k, respectively; Comprehensive reserve ratio R k The smaller the value, the stronger the adjustable capability of the cluster. For a cluster without adjustable resources, since it has no adjustable capability, the comprehensive reserve degree is 1.

5. The distribution network multi-objective clustering method suitable for risk prevention and control according to claim 1, characterized in that: The strong coupling within the cluster and the weak coupling between clusters specifically include: Introducing strong and weak coupling indicators within the cluster To characterize the degree of coupling between nodes in the region, the calculation method is shown in equations (11)-(13): In the formula, is the average electrical distance between the adjustable resource nodes and the load nodes in cluster k; N kg 、N kl are the sets of adjustable resources and pure load nodes in cluster k; It is a strong coupling index within the cluster. Its physical meaning is the coupling relationship between the adjustable resource nodes within cluster k and the load nodes within its own cluster. The smaller the value, the stronger the coupling of nodes in the cluster and the closer the connection. is the weak coupling index between clusters, and its physical meaning is the coupling relationship between the adjustable resource nodes in cluster k and the load nodes outside the cluster. The smaller it is, the weaker the coupling between clusters.

6. The distribution network multi-objective clustering method suitable for risk prevention and control according to claim 1, characterized in that: The establishment of the minimum number of nodes in the cluster specifically includes: N kg +N kl ≥2 (14) Formula (14) indicates that a single node cannot form a cluster; The establishing of the constraint on the number of adjustable resource nodes in the cluster specifically includes: N kl ≥1 (15) Formula (15) indicates that a single node cannot form a cluster.

7. A distribution network multi-objective clustering system suitable for risk prevention and control, based on the distribution network multi-objective clustering method suitable for risk prevention and control according to any one of claims 1 to 6, characterized in that: include: The objective function building module of the multi-objective cluster partitioning model is used to obtain the weighted sum of the comprehensive electrical distance, comprehensive reserve degree, strong coupling within the cluster, and weak coupling between clusters based on voltage sensitivity; The constraint condition establishment module of the multi-objective cluster partitioning model is used to establish the minimum number of nodes in the cluster and the number of adjustable resource nodes in the cluster; The multi-objective clustering model building module is used to establish a discrete particle swarm clustering algorithm based on K-means, integrate the objective function and constraint conditions of the multi-objective clustering model, and solve the multi-objective clustering model.

8. The distribution network multi-objective cluster division system suitable for risk prevention and control according to claim 7, characterized in that: In the multi-objective clustering model building module, the particle positions are first initialized based on the K-means clustering algorithm, and then the discrete particle swarm optimization algorithm is used to perform global iterative optimization.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for dividing distribution network multi-objective clusters suitable for risk prevention and control as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for dividing distribution network multi-objective clusters suitable for risk prevention and control as described in any one of claims 1 to 7 are implemented.

Citation Information

Cited By

  • Power distribution network dynamic partitioning method and system considering network topology and resource time-varying characteristics

    CN120087849A