Power distribution network dispatching method based on voltage support strength clustering

Abstract

The invention discloses a power distribution network scheduling method based on voltage support strength clustering, and belongs to the technical field of power distribution network voltage scheduling, and the method comprises the steps: S1, collecting parameters of each node of a power distribution network as the input of a power flow equation, and carrying out the calculation to obtain a power flow data set; s2, performing sensitivity analysis on the data in the power flow data set to determine a sensitivity factor; s3, differential features in the sensitivity factors are extracted to serve as input of a voltage supporting capacity index formula, and voltage supporting capacity indexes of all nodes of the power distribution network are obtained through calculation; s4, performing clustering analysis according to the voltage support capability indexes to determine the voltage support strength of each node of the power distribution network; s5, the power distribution network executes a voltage scheduling action in response to a voltage scheduling strategy established according to the voltage support strength; the problem that the voltage scheduling strategy is not suitable for the power distribution network due to the fact that the action relation of photovoltaic power generation system power of different nodes of the power distribution network on the voltage of the power distribution network is not considered in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network voltage dispatching, and in particular to a distribution network dispatching method based on voltage support strength clustering. Background Art

[0002] With the rapid development of renewable energy, photovoltaic power generation, as an important clean energy, is increasingly connected to the distribution network. However, the power generation capacity and output characteristics of photovoltaic power generation systems fluctuate greatly under different weather conditions, resulting in voltage fluctuations, over-limit and other problems in the distribution network connected to photovoltaic power generation.

[0003] When solving the above problems, the more economical and effective solution of the existing technology is generally to control the voltage of the distribution network through the reactive power of the photovoltaic power generation system to solve the problems of voltage fluctuation and over-limit of the distribution network. However, the operation scenario of the distribution network is complex and the distribution network has many nodes connected to the photovoltaic power generation system, which makes it difficult for the existing technology to analyze the deep relationship between the power of the photovoltaic power generation system and the voltage of the distribution network in the distribution network, and further difficult to achieve stable operation of the distribution network by controlling the voltage of the distribution network through the power of the photovoltaic power generation system.

[0004] Chinese patent, publication number: CN112366752B, publication date: September 9, 2022, discloses a real-time voltage control method based on photovoltaic inverter control, including: S1: calculating the maximum reactive power capacity and the average line voltage; S2: comparing the line voltage of the photovoltaic inverter and the average line voltage, and calculating the reactive reference value of the inverter according to the maximum reactive power capacity; S3: taking the grid voltage Vabc and current Iabc at the grid connection point, the active reference value Pref set by the grid operator, and the reactive reference value Qref calculated in step S2 as inputs of the measurement controller, and using coordinate transformation to generate symmetrical components; S4: inputting the symmetrical components generated by S3 into the reference generator module to generate a current reference value; S5: inputting the current reference value generated in S4 into the current controller module, and the current controller module controls the output current according to the reference value; However, when analyzing the relationship between the power of the photovoltaic power generation system and the voltage of the distribution network, the invention did not deeply analyze the relationship between the power of the photovoltaic power generation system corresponding to different nodes of the distribution network and the voltage of the distribution network, resulting in poor stability of the distribution network when operating according to the voltage scheduling strategy obtained therefrom. Summary of the invention

[0005] The purpose of the present invention is to address the problem that the existing technology does not consider the effect of the power of photovoltaic power generation systems at different nodes of the distribution network on the voltage of the distribution network, resulting in the problem that the voltage dispatching strategy is not suitable for the distribution network; a distribution network dispatching method based on voltage support strength clustering is proposed, and a sensitivity factor characterizing the degree of correlation between power data and voltage data is obtained by performing sensitivity analysis on the calculation results of the power flow equation, and then a cluster analysis is performed on each node of the distribution network according to the voltage support capability index calculated based on the sensitivity factor to determine the voltage support strength of each node of the distribution network. The distribution network executes a voltage dispatching action in response to the voltage dispatching strategy established according to the voltage support strength, thereby significantly improving the stability of the distribution network.

[0006] In a first aspect, a technical solution provided in an embodiment of the present invention is a distribution network dispatching method based on voltage support strength clustering, comprising the following steps: S1, collect the parameters of each node of the distribution network as the input of the power flow equation to calculate and obtain the power flow data set; S2. Performing sensitivity analysis on the data in the power flow data set to determine a sensitivity factor; S3, extracting the differential characteristics in the sensitivity factor as the input of the voltage support capability index formula to calculate the voltage support capability index of each node of the distribution network; S4. Performing cluster analysis based on the voltage support capability index to determine the voltage support strength of each node in the distribution network; S5. The distribution network performs a voltage dispatching action in response to a voltage dispatching strategy established according to the voltage support strength.

[0007] In this scheme, the parameters of the distribution network are input into the power flow equation to calculate the power flow state of the distribution network, and the state of each node of the distribution network is obtained. A sensitivity analysis is performed on the data in the power flow data set obtained based on the calculation results of the power flow equation to determine the sensitivity factor that characterizes the degree of correlation between power data and voltage data, and to clarify the relationship between power and voltage in each node in the distribution network. Then, the voltage support capability index of each node in the distribution network is calculated based on the relationship between power and voltage, and a cluster analysis is performed on each node in the distribution network based on the voltage support capability index to digitize the support strength of different nodes in the distribution network for the distribution network voltage. Finally, the distribution network executes a voltage dispatching action in response to the voltage dispatching strategy established according to the voltage support strength, and dispatches the distribution network voltage according to the real-time changes of the photovoltaic power generation system connected to the distribution network, thereby significantly improving the stability and power supply quality of the distribution network.

[0008] Preferably, the power flow equation in S1 is specifically as follows: Where Pi in represents the active power injected at node i, represents the reactive power injected at node i, U i represents the voltage amplitude at node i, U j represents the voltage amplitude at node j, θ ij represents the voltage phase difference between node i and node j, G ij represents the conductance between node i and node j in the distribution network node admittance matrix, B ij represents the susceptance between node i and node j in the node admittance matrix of the distribution network, Ω N represents the set of all nodes in the distribution network, Ω PQ represents the set of all PQ nodes in the distribution network, N bus Represents the total number of nodes in the distribution network.

[0009] Preferably, in S1, the specific process of constructing the power flow data set is: S11, establishing an active power change matrix based on the active power of each node in the distribution network; Synchronously, a reactive power variation matrix is ​​established based on the reactive power of each node in the distribution network; Synchronously, a voltage amplitude variation matrix is ​​established based on the voltage amplitudes of each node of the distribution network; Synchronously, a voltage phase angle difference variation matrix is ​​established based on the voltage phase angle differences of different nodes in the distribution network; S12. Based on the distribution network nodes, the active power change matrix, reactive power change matrix, voltage amplitude change matrix and voltage phase angle difference change matrix are sorted to obtain a power flow data set.

[0010] In this scheme, the active power and reactive power of each node in the distribution network are respectively established with corresponding change matrices, and the voltage amplitude and voltage phase difference of each node in the distribution network are respectively established with corresponding change matrices, so as to facilitate the analysis of the active resource characteristics and reactive resource characteristics of different nodes in the distribution network and provide support for the subsequent determination of the sensitivity factor.

[0011] Preferably, in S2, the specific process of performing sensitivity analysis on the data in the power flow data set to determine the sensitivity factor is: S21, extracting the active power change matrix, reactive power change matrix, voltage amplitude change matrix and voltage phase angle difference change matrix corresponding to the power flow data set according to the steady-state solution of the power flow equation; S22. Construct a linearized equation based on the active power change matrix, the reactive power change matrix, the voltage amplitude change matrix and the voltage phase angle difference change matrix, and invert the linearized equation to determine a sensitivity factor.

[0012] In this scheme, the corresponding active power change matrix, reactive power change matrix, voltage amplitude change matrix and voltage phase angle difference change matrix are extracted at the steady-state solution of the power flow equation to construct a linearized equation. The nonlinear power flow equation can be linearized first and then the sensitivity factor can be calculated to determine the sensitivity factor. While simplifying and improving the calculation efficiency of the power flow equation, it can better reflect the data characteristics of power and voltage and effectively improve the accuracy of the sensitivity factor.

[0013] Preferably, the linearization equation in S22 is specifically as follows: In the formula, ΔP is the active power change matrix, ΔQ is the reactive power change matrix, Δθ is the voltage phase angle difference change matrix, ΔU is the voltage amplitude change matrix, J is the Jacobian matrix, is the first-order partial derivative of active power with respect to voltage phase angle difference, is the first-order partial derivative of active power with respect to voltage amplitude, is the first-order partial derivative of reactive power with respect to voltage phase angle difference, is the first-order partial derivative of reactive power with respect to voltage amplitude.

[0014] Preferably, in S22, the inversion result of the linearized equation is specifically: Where Δθ is the voltage phase angle change matrix, ΔU is the voltage amplitude change matrix, ΔP is the active power change matrix, ΔQ is the reactive power change matrix, and J -1 is the inverse matrix of the Jacobian matrix, S is the sensitivity factor matrix, S Pθ is the sensitivity factor of the voltage phase angle difference corresponding to the active power, S PU is the sensitivity factor of active power to voltage amplitude, S Qθ is the sensitivity factor of reactive power corresponding to voltage phase angle difference, S QU is the sensitivity factor of reactive power to voltage amplitude.

[0015] Preferably, in S3, the specific process of extracting the differential characteristics in the sensitivity factor as the input of the voltage support capability index formula to calculate the voltage support capability index of each node of the distribution network is: S31, extracting the corresponding differential features in the sensitivity factor based on the corresponding relationship between the power data and the voltage data in the power flow data set, and obtaining the rated capacity of the photovoltaic power station corresponding to the node where the photovoltaic power station is installed in the distribution network; S32. Calculate the voltage support capability index by using the differential characteristic and the rated capacity of the photovoltaic power station as an input voltage support capability index formula.

[0016] In this scheme, the differential characteristics in the sensitivity factor are extracted to represent the corresponding relationship between the power data and the voltage data in the power flow data set, so as to facilitate the calculation based on the differential characteristics, and the rated capacity of the photovoltaic power station corresponding to the node where the photovoltaic power station is installed in the distribution network is added in the calculation process, so as to clarify the interaction relationship between the photovoltaic power generation power of different nodes and the distribution network voltage, and determine the voltage support capacity index of the node.

[0017] Preferably, in S32, the voltage support capability index formula is specifically: Where WVSAI i is the voltage support capability index of node i, Ω PQ is the set of all PQ nodes in the distribution network, is the differential characteristic of the voltage amplitude at node i corresponding to the reactive power at node j, Ω C It is a collection of nodes that install photovoltaic power stations in the distribution network. is the rated capacity of the PV power station installed at node j.

[0018] Preferably, in S4, the specific process of performing cluster analysis according to the voltage support capability index to determine the voltage support strength of each node in the distribution network is: S41, establishing a sample set based on the voltage support capability index and the photovoltaic power generation power and node voltage in the node parameters; S42, randomly selecting a sample in the sample set as a sample to be tested, and selecting a companion test sample according to a preset test threshold; S43, calculating the Euclidean distance between the sample to be tested and the companion test sample according to the Euclidean distance formula; S44, clustering the samples in the sample set according to the Euclidean distance to determine a strong support sample set, a normal support sample set, and a weak support sample set; S45. Determine the voltage support strength of nodes corresponding to different samples according to the strong support sample set, the normal support sample set, and the weak support sample set.

[0019] In this scheme, a sample set is established based on the voltage support capability index and the photovoltaic power generation power and node voltage in the node parameters, and the samples in the sample set are divided based on the calculated Euclidean distance between different samples. The interaction relationship between different nodes in the distribution network and the distribution network voltage is determined through the correspondence between the samples and the nodes in the distribution network, and the voltage support strength of each node is clarified to support the construction of the voltage dispatching strategy of the distribution network.

[0020] Preferably, in S43, the Euclidean distance formula is specifically: In the formula, yi is the i-th sample, y j is the jth sample, d(y i ,y j ) is the Euclidean distance between the i-th sample and the j-th sample, y ik is the kth feature attribute value of the i-th sample, y jk is the kth feature attribute value of the jth sample, and m is the test threshold.

[0021] The present invention has at least the following substantial effects: This solution inputs the parameters of the distribution network into the power flow equation to calculate the power flow state of the distribution network, obtains the state of each node in the distribution network, and performs sensitivity analysis on the data in the power flow data set obtained based on the calculation results of the power flow equation to determine the sensitivity factor that characterizes the degree of correlation between power data and voltage data, clarifies the relationship between power and voltage in each node in the distribution network, and then calculates the voltage support capability index of each node in the distribution network based on the relationship between power and voltage, and performs cluster analysis on each node in the distribution network based on the voltage support capability index to digitize the support strength of different nodes in the distribution network for the distribution network voltage. Finally, the distribution network executes voltage dispatching actions in response to the voltage dispatching strategy established according to the voltage support strength, and dispatches the distribution network voltage according to the real-time changes of the photovoltaic power generation system connected to the distribution network, significantly improving the stability and power supply quality of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Other features, objects and advantages of the present invention will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings. The drawings are only for the purpose of illustrating preferred embodiments and are not to be considered as limiting the present invention. Also, the same reference symbols are used throughout the drawings to represent the same parts.

[0023] Figure 1 This is a flow chart of a distribution network dispatching method based on voltage support strength clustering according to the present invention; Figure 2 This is a schematic diagram of the distribution network structure connected to the photovoltaic power station; Figure 3 Schematic diagram of the voltage support intensity division corresponding to photovoltaic power generation power at each node in the distribution network. DETAILED DESCRIPTION

[0024] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific implementation method described herein is only an optimal embodiment of the present invention, which is only used to explain the present invention and does not limit the scope of protection of the present invention. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0025] Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe the operations (or steps) as sequential processes, many of the operations (or steps) therein can be implemented in parallel, concurrently, or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the drawings; the process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0026] Embodiment 1: like Figure 1 As shown, this embodiment provides a distribution network scheduling method based on voltage support strength clustering, comprising the following steps: S1. Collect the parameters of each node of the distribution network as the input of the power flow equation to calculate the power flow data set.

[0027] In one embodiment, the power flow equation is specifically as follows: Where P i in represents the active power injected at node i, represents the reactive power injected at node i, U i represents the voltage amplitude at node i, U j represents the voltage amplitude at node j, θ ij represents the voltage phase difference between node i and node j, G ij represents the conductance between node i and node j in the distribution network node admittance matrix, B ij represents the susceptance between node i and node j in the node admittance matrix of the distribution network, Ω N represents the set of all nodes in the distribution network, Ω PQ represents the set of all PQ nodes in the distribution network, N bus Represents the total number of nodes in the distribution network.

[0028] In one embodiment, the specific process of constructing the power flow data set is: S11, establishing an active power change matrix based on the active power of each node in the distribution network; Synchronously, a reactive power variation matrix is ​​established based on the reactive power of each node in the distribution network; Synchronously, a voltage amplitude variation matrix is ​​established based on the voltage amplitudes of each node of the distribution network; Synchronously, a voltage phase angle difference variation matrix is ​​established based on the voltage phase angle differences of different nodes in the distribution network; S12. Based on the distribution network nodes, the active power change matrix, reactive power change matrix, voltage amplitude change matrix and voltage phase angle difference change matrix are sorted to obtain a power flow data set.

[0029] In this embodiment, while collecting the parameters of each node in the distribution network, the structural characteristics of the distribution network are obtained, and the distribution and order of each node in the distribution network are obtained according to the structural characteristics of the distribution network.

[0030] S2. Perform sensitivity analysis on the data in the power flow data set to determine a sensitivity factor.

[0031] In one embodiment, the specific process of sensitivity analysis includes: S21, extracting the active power change matrix, reactive power change matrix, voltage amplitude change matrix and voltage phase angle difference change matrix corresponding to the power flow data set according to the steady-state solution of the power flow equation; S22. Construct a linearized equation based on the active power change matrix, the reactive power change matrix, the voltage amplitude change matrix and the voltage phase angle difference change matrix, and invert the linearized equation to determine a sensitivity factor.

[0032] In one embodiment, the linearization equation in S22 is specifically as follows: In the formula, ΔP is the active power change matrix, ΔQ is the reactive power change matrix, Δθ is the voltage phase angle difference change matrix, ΔU is the voltage amplitude change matrix, J is the Jacobian matrix, is the first-order partial derivative of active power with respect to voltage phase angle difference, is the first-order partial derivative of active power with respect to voltage amplitude, is the first-order partial derivative of reactive power with respect to voltage phase angle difference, is the first-order partial derivative of reactive power with respect to voltage amplitude.

[0033] In one embodiment, the inverse result of the linearized equation in S22 is specifically: Where Δθ is the voltage phase angle change matrix, ΔU is the voltage amplitude change matrix, ΔP is the active power change matrix, ΔQ is the reactive power change matrix, and J -1 is the inverse matrix of the Jacobian matrix, S is the sensitivity factor matrix, S Pθ is the sensitivity factor of the voltage phase angle difference corresponding to the active power, S PU is the sensitivity factor of active power to voltage amplitude, S Qθ is the sensitivity factor of reactive power corresponding to voltage phase angle difference, S QU is the sensitivity factor of reactive power to voltage amplitude.

[0034] In this embodiment, the nonlinear power flow equation is first linearized at the steady-state solution of the power flow equation, that is, at a set of voltage amplitudes and voltage phase angle values ​​that meet the power balance conditions of all nodes under the conditions of the determined distribution network structure and the determined parameters of each node of the distribution network, and then the Jacobian matrix is ​​used to perform sensitivity analysis on the calculation results of the linearized power flow equation to obtain the reactive resource adaptability of the sag mitigation scenario, that is, the sensitivity factor determined by inverting the linearized equation; The sensitivity factor S of reactive power corresponding to voltage amplitude QU For example, the sensitivity factor S QU Each value in corresponds to a set of differential relations when The larger the value, the greater the j Caused by U i The more sensitive the change, that is, the more obvious the increase in reactive power injection at node j is, the more obvious the increase in voltage at node i is. On the contrary, when The smaller the value of is, the smaller the effect of increasing reactive power injection at node j on the voltage at node i is; if If the value of is negative, increasing the reactive power at node j and injecting it into node i will reduce the voltage at node i.

[0035] S3. Extract the differential characteristics in the sensitivity factor as the input of the voltage support capability index formula to calculate the voltage support capability index of each node in the distribution network.

[0036] In one embodiment, the differential characteristics in the sensitivity factor are extracted to calculate the voltage support capability index, specifically including: S31, extracting the corresponding differential characteristics in the sensitivity factor based on the corresponding relationship between the power data and the voltage data in the power flow data set, and obtaining the rated capacity of the photovoltaic power station corresponding to the node where the photovoltaic power station is installed in the distribution network; S32. Calculate the voltage support capability index by using the differential characteristic and the rated capacity of the photovoltaic power station as an input voltage support capability index formula.

[0037] In one embodiment, the voltage support capability indicator formula in S32 is specifically: Where WVSAI i is the voltage support capability index of node i, Ω PQ is the set of all PQ nodes in the distribution network, is the differential characteristic of the voltage amplitude at node i corresponding to the reactive power at node j, Ω C It is a collection of nodes that install photovoltaic power stations in the distribution network. is the rated capacity of the PV power station installed at node j.

[0038] In this embodiment, since the differential relationship corresponding to the sensitivity factor has been revealed in the process of inverting the linearized equation to determine the sensitivity factor in the aforementioned step, it is only necessary to extract the differential relationship contained in the sensitivity factor as a differential feature. At this time, the voltage support capacity of each node can be calculated according to the differential feature. The formula for calculating the voltage support capacity of each node is as follows: Where, VSAI i Represents the voltage support capability of node i, Ω I represents the set of nodes with installed photovoltaic power stations, Ω PQ represents the set of all PQ nodes in the distribution network, Represents the differential characteristics of the voltage amplitude at node i corresponding to the reactive power at node j; however, the voltage support capability VSAI i The rated capacity of the PV power station corresponding to different nodes is not considered, so the voltage support capacity VSAI i On this basis, the installed capacity of the photovoltaic power station is used as the weight coefficient to weight the voltage support capacity, and the influence of the installed capacity of the photovoltaic power station on the node load can be obtained, that is, the voltage support capacity index. The specific calculation formula is the above voltage support capacity index formula.

[0039] S4. Perform cluster analysis based on the voltage support capability index to determine the voltage support strength of each node in the distribution network.

[0040] In one embodiment, the specific process of performing cluster analysis according to the voltage support capability indicator includes: S41, establishing a sample set based on the voltage support capability index and the photovoltaic power generation power and node voltage in the node parameters; S42, randomly selecting a sample in the sample set as a sample to be tested, and selecting a companion test sample according to a preset test threshold; S43, calculating the Euclidean distance between the sample to be tested and the companion test sample according to the Euclidean distance formula.

[0041] The Euclidean distance formula is specifically: In the formula, y i is the i-th sample, y j is the jth sample, d(y i ,y j ) is the Euclidean distance between the i-th sample and the j-th sample, y ik is the kth feature attribute value of the i-th sample, y jk is the kth feature attribute value of the jth sample, and m is the test threshold.

[0042] S44, clustering the samples in the sample set according to the Euclidean distance to determine a strong support sample set, a normal support sample set, and a weak support sample set; S45. Determine the voltage support strength of nodes corresponding to different samples according to the strong support sample set, the normal support sample set, and the weak support sample set.

[0043] In this embodiment, the KNN clustering algorithm is used to analyze the data information in the constructed sample set. Before the analysis, the K value needs to be set, that is, the size of the test threshold m. However, there is no unified method for determining the K value. Generally, an initial value is determined first, and then the optimal K value is obtained by continuous debugging based on experimental results. In this embodiment, based on past experience, the K value is set to 3, and the companion test sample is the K samples closest to the sample to be tested in the sample set. Then, the K samples closest to the sample to be tested are found according to the calculated Euclidean distance and classified. If Figure 2 The neighboring nodes of each node in the distribution network shown are shown in Table 1: Table 1. Statistics of neighboring nodes of each node node The three nearest nodes node The three nearest nodes 2 4、21、23 20 2、23、26 4 2、21、23 21 2、4、23 7 26、28、31 23 2、4、21 9 4、17、33 26 7、12、28 12 7、26、28 28 7、26、31 14 23、28、33 31 7、28、33 17 4、9、33 33 9、17、31 The corresponding cluster analysis is shown in Table 2. The specific division method is as follows Figure 3 As shown: Table 2. Cluster analysis results of each node Clustering node Strong support 2、4、20、21、23 Ordinary support 7、12、26、28、31 Weak support 9、14、17、33 So far, the support strength of different nodes in the distribution network to the distribution network voltage has been obtained through the voltage support capability index.

[0044] S5. The distribution network performs a voltage dispatching action in response to a voltage dispatching strategy established according to the voltage support strength.

[0045] This embodiment has at least the following substantial effects: In this embodiment, the parameters of the distribution network are input into the power flow equation to calculate the power flow state of the distribution network, and the state of each node of the distribution network is obtained. A sensitivity analysis is performed on the data in the power flow data set obtained based on the calculation result of the power flow equation to determine the sensitivity factor that characterizes the degree of correlation between power data and voltage data, and the relationship between power and voltage in each node in the distribution network is clarified. Then, the voltage support capability index of each node in the distribution network is calculated based on the relationship between power and voltage, and a cluster analysis is performed on each node in the distribution network based on the voltage support capability index to digitize the support strength of different nodes in the distribution network for the distribution network voltage. Finally, the distribution network executes a voltage dispatching action in response to the voltage dispatching strategy established according to the voltage support strength, and dispatches the distribution network voltage according to the real-time changes of the photovoltaic power generation system connected to the distribution network, thereby significantly improving the stability and power supply quality of the distribution network.

[0046] The above specific embodiments are preferred embodiments of the present invention, and are not intended to limit the specific implementation scope of the present invention. The scope of the present invention includes but is not limited to the present specific embodiments. All equivalent changes made in accordance with the shape, structure, and method of the present invention are within the protection scope of the present invention.

Claims

1. A distribution network dispatching method based on voltage support strength clustering, characterized in that: The method comprises the following steps: S1, collect the parameters of each node of the distribution network as the input of the power flow equation to calculate and obtain the power flow data set; S2. Performing sensitivity analysis on the data in the power flow data set to determine a sensitivity factor; S3, extracting the differential characteristics in the sensitivity factor as the input of the voltage support capability index formula to calculate the voltage support capability index of each node of the distribution network; S4. Performing cluster analysis based on the voltage support capability index to determine the voltage support strength of each node in the distribution network; S5. The distribution network performs a voltage dispatching action in response to a voltage dispatching strategy established according to the voltage support strength.

2. The distribution network dispatching method based on voltage support strength clustering according to claim 1 is characterized in that: The power flow equation in S1 is specifically as follows: In the formula, represents the active power injected at node i, represents the reactive power injected at node i, U i represents the voltage amplitude at node i, U j represents the voltage amplitude at node j, θ ij represents the voltage phase difference between node i and node j, G ij represents the conductance between node i and node j in the distribution network node admittance matrix, B ij represents the susceptance between node i and node j in the node admittance matrix of the distribution network, Ω N represents the set of all nodes in the distribution network, Ω PQ represents the set of all PQ nodes in the distribution network, N bus Represents the total number of nodes in the distribution network.

3. The distribution network dispatching method based on voltage support strength clustering according to claim 2 is characterized in that: In S1, the specific process of constructing the power flow data set is as follows: S11, establishing an active power change matrix based on the active power of each node in the distribution network; Synchronously, a reactive power variation matrix is ​​established based on the reactive power of each node in the distribution network; Synchronously, a voltage amplitude variation matrix is ​​established based on the voltage amplitudes of each node of the distribution network; Synchronously, a voltage phase angle difference variation matrix is ​​established based on the voltage phase angle differences of different nodes in the distribution network; S12. Based on the distribution network nodes, the active power change matrix, reactive power change matrix, voltage amplitude change matrix and voltage phase angle difference change matrix are sorted to obtain a power flow data set.

4. The distribution network dispatching method based on voltage support strength clustering according to claim 1, characterized in that: In S2, the specific process of performing sensitivity analysis on the data in the power flow data set to determine the sensitivity factor is: S21, extracting the active power change matrix, reactive power change matrix, voltage amplitude change matrix and voltage phase angle difference change matrix corresponding to the power flow data set according to the steady-state solution of the power flow equation; S22. Construct a linearized equation based on the active power change matrix, the reactive power change matrix, the voltage amplitude change matrix and the voltage phase angle difference change matrix, and invert the linearized equation to determine a sensitivity factor.

5. The distribution network dispatching method based on voltage support strength clustering according to claim 4 is characterized in that: The linearization equation in S22 is specifically as follows: In the formula, ΔP is the active power change matrix, ΔQ is the reactive power change matrix, Δθ is the voltage phase angle difference change matrix, ΔU is the voltage amplitude change matrix, and J is the Jacobian matrix. is the first-order partial derivative of active power with respect to voltage phase angle difference, is the first-order partial derivative of active power with respect to voltage amplitude, is the first-order partial derivative of reactive power with respect to voltage phase angle difference, is the first-order partial derivative of reactive power with respect to voltage amplitude.

6. The distribution network dispatching method based on voltage support strength clustering according to claim 4 is characterized in that: In S22, the inverse result of the linearized equation is specifically: Where Δθ is the voltage phase angle difference change matrix, ΔU is the voltage amplitude change matrix, ΔP is the active power change matrix, ΔQ is the reactive power change matrix, and J -1 is the inverse matrix of the Jacobian matrix, S is the sensitivity factor matrix, S Pθ is the sensitivity factor of the voltage phase difference corresponding to the active power, S PU is the sensitivity factor of active power to voltage amplitude, S Qθ is the sensitivity factor of reactive power corresponding to voltage phase angle difference, S QU is the sensitivity factor of reactive power to voltage amplitude.

7. The distribution network dispatching method based on voltage support strength clustering according to claim 1, characterized in that: In S3, the specific process of extracting the differential characteristics in the sensitivity factor as the input of the voltage support capability index formula to calculate the voltage support capability index of each node of the distribution network is: S31, extracting the corresponding differential features in the sensitivity factor based on the corresponding relationship between the power data and the voltage data in the power flow data set, and obtaining the rated capacity of the photovoltaic power station corresponding to the node where the photovoltaic power station is installed in the distribution network; S32. Calculate the voltage support capability index by using the differential characteristic and the rated capacity of the photovoltaic power station as an input voltage support capability index formula.

8. The distribution network dispatching method based on voltage support strength clustering according to claim 7, characterized in that: In S32, the voltage support capability index formula is specifically: Where WVSAI i is the voltage support capability index of node i, Ω PQ is the set of all PQ nodes in the distribution network, is the differential characteristic of the voltage amplitude at node i corresponding to the reactive power at node j, Ω C It is a collection of nodes that install photovoltaic power stations in the distribution network. is the rated capacity of the PV power station installed at node j.

9. The distribution network dispatching method based on voltage support strength clustering according to claim 1, characterized in that: In S4, the specific process of performing cluster analysis according to the voltage support capability index to determine the voltage support strength of each node in the distribution network is as follows: S41, establishing a sample set based on the voltage support capability index and the photovoltaic power generation power and node voltage in the node parameters; S42, randomly selecting a sample in the sample set as a sample to be tested, and selecting a companion test sample according to a preset test threshold; S43, calculating the Euclidean distance between the sample to be tested and the companion sample according to the Euclidean distance formula; S44, clustering the samples in the sample set according to the Euclidean distance to determine a strong support sample set, a normal support sample set, and a weak support sample set; S45. Determine the voltage support strength of nodes corresponding to different samples according to the strong support sample set, the normal support sample set, and the weak support sample set.

10. The distribution network dispatching method based on voltage support strength clustering according to claim 9, characterized in that: In S43, the Euclidean distance formula is specifically: In the formula, y i is the i-th sample, y j is the jth sample, d(y i ,y j ) is the Euclidean distance between the i-th sample and the j-th sample, y ik is the kth feature attribute value of the i-th sample, y jk is the kth feature attribute value of the jth sample, and m is the test threshold.

Images