Method and apparatus for large-scale distributed photovoltaic simplified clustering for weak measurement scenarios
By employing Morse-Smale complex and affine propagation clustering methods, the accuracy problem of distributed photovoltaic clustering under weak measurement scenarios was solved, enabling efficient distributed photovoltaic cluster partitioning and grid regulation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- ELECTRIC POWER RESEARCH INSTITUTE OF STATE GRID SHANDONG ELECTRIC POWER COMPANY
- Filing Date
- 2022-11-29
- Publication Date
- 2026-06-05
AI Technical Summary
In scenarios with weak measurement capabilities, clustering methods for distributed photovoltaic systems rely on communication systems, and common clustering algorithms such as K-means require the number of clusters to be predetermined, leading to inaccurate clustering results and increasing the difficulty of distribution network regulation.
The Morse-Smale complex and affine propagation clustering methods are adopted to construct a topologically simplified model of the distributed photovoltaic power output characteristics by mining multi-source measurement data of key nodes. Sub-clusters are generated using the voltage sensitivity matrix to achieve simplified clustering of distributed photovoltaic power.
It improves the accuracy and computational performance of clustering results, eliminates redundant topological structures, enhances the control capability of distributed photovoltaic clusters, and reduces dependence on historical data.
Smart Images
Figure CN115718876B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of distributed photovoltaic technology, and in particular to a simplified clustering method and apparatus for large-scale distributed photovoltaic systems in weak measurement scenarios. Background Technology
[0002] Distributed photovoltaic (PV) systems offer advantages such as abundant exploitable resources, low development and construction difficulty, and significant energy-saving and environmental benefits. However, large-scale integration can impact the distribution network to some extent, potentially leading to issues like voltage exceeding limits, power backflow, and equipment overload, thus increasing operational risks. Therefore, simplifying the clustering of distributed PV systems to ensure that the resulting sub-clusters exhibit characteristics similar to individual power sources on the external grid, demonstrating grid-friendly characteristics and control capabilities similar to traditional generators, is crucial for grid dispatch and stable operation. The power characteristics of distributed PV are coupled with temporal and spatial information, resulting in complex topological relationships. Currently, due to insufficient coverage of distributed PV monitoring points in the distribution network, while installed devices such as 10kV smart boundary switches, substation fusion terminals, and low-voltage PV smart switches can monitor the basic operational status of distributed PV users, data quality and data transmission links are unstable. This leads to a lack of basic historical data for power prediction at distributed PV monitoring points, severely impacting prediction accuracy and increasing the difficulty of distribution network control and operation. Therefore, it is urgent to address the problem of simplified clustering of distributed PV systems under weak measurement scenarios.
[0003] Currently, most research on clustering methods for distributed photovoltaic (PV) systems with weak measurement capabilities focuses on eliminating unmeasurable node information in the distribution network through network simplification methods. This involves using measurement and communication data from key nodes to transform incompletely measured regional distribution networks into simplified networks with complete measurement information. Existing methods are highly dependent on communication systems, and common clustering processes for large-scale distributed PV, such as the K-means algorithm, require pre-determining the number of clusters, which is often difficult to determine. Furthermore, K-means yields a locally optimal solution, and the clustering results depend on the initially given center values, often requiring multiple initial values to be tried, which presents significant technical limitations in practical applications. Summary of the Invention
[0004] This invention provides a simplified clustering method and apparatus for large-scale distributed photovoltaic (PV) systems in scenarios with weak measurement capabilities. It fully leverages the effective information from multi-source measurement data at key nodes and employs affine propagation clustering to enhance the rationality of distributed PV cluster partitioning. To provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. This summary is not intended as a general commentary, nor is it intended to identify key / important components or describe the scope of protection of these embodiments. Its sole purpose is to present some concepts in a simple form as a prelude to the detailed description that follows.
[0005] According to a first aspect of the present invention, a simplified clustering method for large-scale distributed photovoltaic systems in weak measurement scenarios is provided, comprising:
[0006] Using the similarity of the power output characteristics at the grid connection point of distributed photovoltaic in the distribution network as the Morse feature index value, the key points of the Morse feature index value are searched to form a Morse-Smale complex that reflects the power output characteristics of distributed photovoltaic.
[0007] Define the Morse eigenvalue function, use the average index value of the Morse-Smale complex to back-calculate the eigenvalue index of the key points contained in the Morse-Smale complex, iteratively update the two-stage importance index of the Morse-Smale complex, extract the key points that are most similar to the output characteristics at the grid connection point of distributed photovoltaic, and complete the topology simplification of the distribution network containing large-scale distributed photovoltaic.
[0008] For key grid connection points in a simplified topology containing large-scale distributed photovoltaic (PV) distribution networks, multi-source measurement data is acquired, and typical output scenarios are constructed that take into account the probability distribution characteristics of distributed PV power and the correlation with typical meteorological information. The sensitivity of typical output scenarios to the voltage of nodes in the distribution network is calculated, and the voltage sensitivity expectation matrix under each typical output scenario is obtained. Based on the voltage sensitivity expectation matrix, sub-clusters are generated by affine propagation clustering, and simplified clustering of large-scale distributed PV for weak measurement scenarios is completed.
[0009] In one embodiment, the method uses the similarity of power output characteristics at distributed photovoltaic grid connection points in the distribution network as the Morse feature index value, and the step of searching for key points of the Morse feature index value further includes:
[0010] Using the grid connection points of distributed photovoltaic power in the regional power grid as the basic data structure, and the similarity of the output characteristics of distributed photovoltaic power as the Morse feature index value, key points are searched through the neighbor comparison method.
[0011] In one embodiment, the step of searching for key points using the neighbor comparison method further includes:
[0012] For any distributed photovoltaic grid-connected point n, compare its Morse characteristic index value with that of its neighboring distributed photovoltaic grid-connected point ni;
[0013] If all eigenvalues of n are greater than the eigenvalues of ni, then n is a maximum point; if all eigenvalues of n are less than the eigenvalues of ni, then n is a minimum point; if the difference between the eigenvalues of n and ni is 5, then it is determined to be a saddle point.
[0014] In one embodiment, the step of constructing a Morse-Smale complex reflecting the characteristics of distributed photovoltaic power output further includes:
[0015] Starting from each saddle point, proceed along the direction of the minimum or maximum gradient of the eigenvalues until the next extreme point is found; when a saddle point corresponds to a pair of maxima, a descending Morse simple complex is formed; when a saddle point corresponds to a pair of minima, an ascending Morse simple complex is formed; according to the duality of the Morse-Smale complex, the descending Morse simple complex and the ascending Morse simple complex are pairwise orthogonal and form the Morse-Smale complex of the distributed photovoltaic power output characteristics.
[0016] In one embodiment, the method defines a Morse eigenvalue function, uses the average index value of the Morse-Smale complex to back-calculate the characteristic index values of the key points contained in the Morse-Smale complex, iteratively updates the two-stage importance index of the Morse-Smale complex, and extracts the key points most similar to the output characteristics at the distributed photovoltaic grid connection point. The steps of completing the topology simplification of the large-scale distributed photovoltaic distribution network further include:
[0017] Introducing an eigenvalue function, we calculate the feature importance index for all distributed photovoltaic grid-connected points extracted from the descending Morse simple complex and the ascending Morse simple complex, expressed as:
[0018]
[0019] In the formula, F l (n l f(v) is the difference of the Morse characteristic functions between all points in the descending Morse simple complex and the extreme points of the simple complex containing distributed photovoltaic distribution networks; i ) is the Morse characteristic function of all points in the descending Morse simple complex and the ascending Morse simple complex; max[f u [(a)] is the maximum value of the Morse characteristic function of the minimum point connected to the saddle point in the ascending Morse complex; min[f d (b)] is the minimum value of the Morse characteristic function of the minimum point connected to the saddle point in the descending Morse complex; N is the number of points on the descending Morse simple complex and the ascending Morse simple complex, that is, the total number of distributed grid-connected points extracted from the Morse complex.
[0020] In one embodiment, the method defines a Morse eigenvalue function, uses the average index value of the Morse-Smale complex to back-calculate the characteristic index values of the key points contained in the Morse-Smale complex, iteratively updates the two-stage importance index of the Morse-Smale complex, and extracts the key points most similar to the output characteristics at the distributed photovoltaic grid connection point. The steps of completing the topology simplification of the large-scale distributed photovoltaic distribution network further include:
[0021] Based on the calculation of the importance index of all distributed photovoltaic grid-connected points extracted from the descending Morse simple complex and the ascending Morse simple complex, the mean of the calculated feature importance is used as the first-stage importance index of the Morse-Smale complex, expressed as:
[0022]
[0023] In the formula, F l 1 It is a first-stage importance indicator of the Morse-Smale complex.
[0024] In one embodiment, the method defines a Morse eigenvalue function, uses the average index value of the Morse-Smale complex to back-calculate the characteristic index values of the key points contained in the Morse-Smale complex, iteratively updates the two-stage importance index of the Morse-Smale complex, and extracts the key points most similar to the output characteristics at the distributed photovoltaic grid connection point. The steps of completing the topology simplification of the large-scale distributed photovoltaic distribution network further include:
[0025] The characteristic importance index of all distributed photovoltaic grid-connected points extracted from the inverse calculation of the descending Morse simple complex and the ascending Morse simple complex is expressed as:
[0026]
[0027] In the formula, F l '(n l This involves incorporating the first-stage importance index of the Morse-Smale complex into F. l (n l After that, the difference between the Morse characteristic function of all points in the descending Morse simple complex and the ascending Morse simple complex containing the distributed photovoltaic distribution network and the extreme point of the simple complex is recalculated, which represents the characteristic importance index of all the distributed photovoltaic grid-connected points extracted from the descending Morse simple complex and the ascending Morse simple complex.
[0028] In one embodiment, the method defines a Morse eigenvalue function, uses the average index value of the Morse-Smale complex to back-calculate the characteristic index values of the key points contained in the Morse-Smale complex, iteratively updates the two-stage importance index of the Morse-Smale complex, and extracts the key points most similar to the output characteristics at the distributed photovoltaic grid connection point. The steps of completing the topology simplification of the large-scale distributed photovoltaic distribution network further include:
[0029] The importance index of the Morse-Smale complex feature is updated for distributed photovoltaic distribution networks, and a two-stage importance index of the Morse-Smale complex is calculated. This index is then used for topology simplification, and is expressed as follows:
[0030]
[0031] In the formula, F l 2 It is the updated two-stage importance index for the Morse-Smale complex.
[0032] In one embodiment, the method further includes the following steps for acquiring multi-source measurement data and constructing a typical output scenario that takes into account the probability distribution characteristics of distributed photovoltaic power and the correlation with typical meteorological information at key grid connection points in a simplified topology of a large-scale distributed photovoltaic power distribution network:
[0033] When selecting typical power output scenarios, a frequency histogram is established based on the historical power statistics of distributed photovoltaic power at key grid connection points. Data at the midpoint of each interval of the frequency histogram constitutes a set of typical power output scenarios.
[0034] In one embodiment, the step of calculating the sensitivity of a typical power output scenario to the node voltage in the distribution network further includes:
[0035] The sensitivity of the voltage amplitude at node i in the distribution network to the injected power under a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation is calculated and expressed as:
[0036]
[0037] In the formula, u ij It represents the sensitivity of the voltage amplitude at node i in the distribution network to the injected power in a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation. It is the voltage amplitude at node i responding to the voltage change of the injected power in a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation. It is the change in injected power at a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation;
[0038] The voltage sensitivity of injected power under typical power output scenarios at key grid-connected points of distributed photovoltaic systems is defined as follows:
[0039] c ij =u ii +u jj -u ij -u ji i,j=1,2,...,n
[0040]
[0041] In the formula, c ij It is the comprehensive voltage sensitivity of the voltage amplitude at each node i in the distribution network to the injected power under a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation. It is composed of the sensitivity of node i and node j themselves, as well as the mutual sensitivity between node i and node j. C ij It is c ij The matrix formed; u ii It is the sensitivity of the voltage amplitude at node i to the injected power in a typical power output scenario at the key grid connection point i of distributed photovoltaic power generation; u jj It is the sensitivity of the voltage amplitude at node j to the injected power in a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation; u ij It is the sensitivity of the voltage amplitude at node i to the injected power in a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation; u ji It represents the sensitivity of the voltage amplitude at node j to the injected power in a typical power output scenario at the critical grid-connected point i of a distributed photovoltaic system.
[0042] In one embodiment, the step of obtaining the voltage sensitivity expectation matrix for each typical output scenario further includes:
[0043] The voltage sensitivity expectation matrix is defined as:
[0044]
[0045] In the formula, E ij This is the voltage sensitivity expectation matrix, where M is the number of typical power output scenarios; p m It represents the probability of a typical power output scenario m occurring. It is the expected voltage sensitivity of the voltage amplitude at node i to the injected power in a typical power output scenario m at the key grid connection point j of distributed photovoltaic power generation.
[0046] In one embodiment, the method further includes the step of generating sub-clusters based on affine propagation clustering according to the voltage sensitivity expectation matrix to complete the simplified clustering of large-scale distributed photovoltaic systems for weak measurement scenarios:
[0047] The algorithm is initialized based on the affine propagation clustering algorithm. The set of real-valued similarities between data points in the voltage sensitivity expectation matrix is used as the initial similarity matrix, which is used as the input of the affine propagation clustering algorithm. The similarity matrix s(i,k) represents the degree to which data point k is suitable as the cluster center of data point i, and is calculated using Euclidean distance.
[0048] The similarity matrix s(i,k) is represented as:
[0049] s(i,k)=-d 2 (x(i),x(k))=-||x(i)-x(k)|| 2
[0050] In the formula, d 2 (x(i),x(k)) is the square of the Euclidean distance between data point k and data point i.
[0051] In one embodiment, the method further includes the step of generating sub-clusters based on affine propagation clustering according to the voltage sensitivity expectation matrix to complete the simplified clustering of large-scale distributed photovoltaic systems for weak measurement scenarios:
[0052] The attraction information matrix r(i,k) and the belonging information matrix a(i,k) are transmitted between each data point. The affine propagation clustering algorithm updates the attraction information matrix r(i,k) and the belonging information matrix a(i,k) according to the data points to perform iterations until the cluster center is determined and the belonging relationship between the data points and the cluster center is determined.
[0053] The attraction information matrix r(i,k) is represented as:
[0054]
[0055] In the formula, This indicates the degree of advantage of the strongest competitor among the data points other than data point k as the cluster center;
[0056] The attribution information matrix a(i,k) is represented as:
[0057]
[0058] In the formula, r(i',k) represents the similarity of data point k as the cluster center of other points besides data point i. It takes all attraction values greater than or equal to 0 and adds the similarity r(k,k) of data point k as the cluster center. That is, the degree to which data point k is suitable to be selected as the cluster center by data point i with the support of these data points with attraction values greater than 0; a(k,k) is the degree of belonging of data point k itself.
[0059] According to a second aspect of the present invention, a simplified clustering device for large-scale distributed photovoltaic systems oriented towards weak measurement scenarios is provided.
[0060] In one embodiment, the device includes a Morse-Smale complex building module, a topology simplification module, and a simplified clustering module; wherein,
[0061] The Morse-Smale complex construction module uses the similarity of the output characteristics at the grid connection point of distributed photovoltaic in the distribution network as the Morse feature index value, searches for the key points of the Morse feature index value, and constructs a Morse-Smale complex that reflects the output characteristics of distributed photovoltaic.
[0062] The topology simplification module defines the Morse eigenvalue function, uses the average index value of the Morse-Smale complex to back-calculate the characteristic index value of the key points contained in the Morse-Smale complex, iteratively updates the two-stage importance index of the Morse-Smale complex, extracts the key points that are most similar to the output characteristics at the distributed photovoltaic grid connection point, and completes the topology simplification of the large-scale distributed photovoltaic distribution network.
[0063] The simplified clustering module acquires multi-source measurement data for key grid-connected points in the simplified topology of large-scale distributed photovoltaic distribution networks. It constructs typical output scenarios that take into account the probability distribution characteristics of distributed photovoltaic power and the correlation with typical meteorological information. It calculates the sensitivity of typical output scenarios to the voltage of nodes in the distribution network, obtains the voltage sensitivity expectation matrix under each typical output scenario, and generates sub-clusters based on affine propagation clustering according to the voltage sensitivity expectation matrix, thus completing the simplified clustering of large-scale distributed photovoltaics for weak measurement scenarios.
[0064] According to a third aspect of the present invention, a computer device is provided.
[0065] In some embodiments, the computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method as described in the first aspect.
[0066] According to a fourth aspect of the present invention, a computer-readable storage medium is provided.
[0067] In some embodiments, a computer program is stored on a computer-readable storage medium; the computer program is executed by a processor to implement the steps of the method as described in the first aspect.
[0068] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:
[0069] 1. Compared with the traditional topology simplification method that divides distributed photovoltaics according to geographical location or electrical distance, this invention uses the similarity of the output characteristics of distributed photovoltaics as the Morse feature index value to realize the extraction and simplification of key features of large-scale distributed photovoltaic topology under weak measurement scenarios, and generates a simplified topology of large-scale distributed photovoltaic distribution network. This effectively eliminates redundant topology structures while avoiding excessive extraction of historical data features.
[0070] 2. This invention proposes to use the average index value of the Morse-Smale complex to back-calculate the feature index value of its key points, and iteratively update the two-stage importance index of the Morse-Smale complex, which effectively improves the accuracy of the Morse-Smale complex in describing the power output characteristics of each grid connection point. Based on the measurement data at some key grid connection points, a data foundation can be provided for the expression and application of distributed photovoltaic clustering in the future.
[0071] 3. This invention defines typical output scenarios that cover meteorological information and the probability distribution characteristics of distributed photovoltaic power. It calculates the voltage sensitivity expectation matrix of the node voltage in the distribution network under each scenario and uses it as the basis for dividing sub-clusters. Based on the affine propagation clustering method, it adaptively generates distributed photovoltaic sub-clusters, realizes rapid clustering of high-dimensional and multi-type measurement information, and improves the clustering effect and computing performance simultaneously.
[0072] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description
[0073] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.
[0074] Figure 1 This is a flowchart of a simplified clustering method for large-scale distributed photovoltaic systems in weak measurement scenarios provided in an embodiment of this application;
[0075] Figure 2 This is a schematic diagram of the adaptive generation process of distributed photovoltaic sub-clusters based on the affine propagation clustering method provided in the embodiments of this application;
[0076] Figure 3 This is a simplified topology diagram of a large-scale distributed photovoltaic power distribution network provided in the embodiments of this application;
[0077] Figure 4 This is a clustering effect diagram of a large-scale distributed photovoltaic power distribution network provided in the embodiments of this application;
[0078] Figure 5 This is a structural diagram of a large-scale distributed photovoltaic simplified clustering device for weak measurement scenarios provided in the embodiments of this application;
[0079] Figure 6 This is a schematic diagram of the structure of a computer device according to an exemplary embodiment. Detailed Implementation
[0080] The following description and accompanying drawings fully illustrate specific embodiments described herein to enable those skilled in the art to practice them. Some embodiments may include or substitute parts and features of other embodiments. The scope of the embodiments herein encompasses the entire scope of the claims and all available equivalents thereof. Throughout this document, the terms “first,” “second,” etc., are used only to distinguish one element from another without requiring or implying any actual relationship or order between the elements. Indeed, a first element can also be referred to as a second element, and vice versa. Furthermore, the terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a structure, apparatus, or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a structure, apparatus, or device. Without further limitation, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the structure, apparatus, or device that includes said element. The various embodiments described herein are presented in a progressive manner, with each embodiment focusing on its differences from other embodiments; similar or identical parts between embodiments can be referred to interchangeably.
[0081] In this document, unless otherwise stated, the term "multiple" means two or more.
[0082] This invention provides a simplified clustering method for large-scale distributed photovoltaic (PV) systems with incomplete measurement information. Based on Morse theory, it extracts key points with the most similar output characteristics at the grid connection points of distributed PV systems, establishing a reliable simplified topology for the distributed PV distribution network. Furthermore, it fully mines the effective information from multi-source measurement data of key nodes and employs affine propagation clustering to enhance the rationality of distributed PV cluster partitioning.
[0083] Figure 1 The flowchart of the simplified clustering method for large-scale distributed photovoltaic systems in weak measurement scenarios of the present invention is shown. Figure 2 This illustration shows a schematic diagram of the adaptive generation process of distributed photovoltaic sub-clusters based on the affine propagation clustering method provided in an embodiment of this application. Figure 1 and Figure 2 As shown:
[0084] S100: Using the similarity of the output characteristics at the grid connection point of distributed photovoltaic in the distribution network as the Morse feature index value, the key points of the Morse feature index value are searched to form a Morse-Smale complex that reflects the output characteristics of distributed photovoltaic.
[0085] In practical implementation, the grid connection points of distributed photovoltaic (PV) systems in the regional power grid are used as the basic data structure. The similarity of the output characteristics of distributed PV systems is used as the Morse feature index value, and key points are searched using the neighbor comparison method. For any distributed PV grid connection point n, it is compared with its neighboring distributed PV grid connection points n. i The Morse characteristic index value, when all the characteristic values of n are greater than n i If the eigenvalues of n are all less than n, then n is a local maximum; when all the eigenvalues of n are less than n... i If the eigenvalues of n are eigenvalues, then n is a local point; when the eigenvalues of n are eigenvalues of n, then n is a local point; i If the magnitude of the eigenvalue comparison value changes by 5, it is determined to be a saddle point. Starting from each saddle point, proceed along the direction of the minimum or maximum gradient of the eigenvalues until the next extreme point is found. When a saddle point corresponds to a pair of maxima, it forms a descending Morse simple complex; when a saddle point corresponds to a pair of minima, it forms an ascending Morse simple complex. According to the duality of the Morse-Smale complex, the descending Morse simple complex and the ascending Morse simple complex are pairwise orthogonal, forming a Morse-Smale complex of the distributed photovoltaic power output characteristics.
[0086] S200: Define the Morse eigenvalue function, use the average index value of the Morse-Smale complex to back-calculate the characteristic index value of the key points contained in the Morse-Smale complex, iteratively update the two-stage importance index of the Morse-Smale complex, extract the key points that are most similar to the output characteristics at the distributed photovoltaic grid connection point, and complete the topology simplification of the large-scale distributed photovoltaic distribution network.
[0087] In practical implementation, to better distinguish the subtle differences in output characteristics among various distributed photovoltaic grid-connected points and to make the Morse-Smale complex more accurate in describing feature points, an eigenvalue function is introduced. This function calculates the feature importance index for all distributed photovoltaic grid-connected points extracted from the descending and ascending Morse simple complexes, expressed as:
[0088]
[0089] In the formula, F l (n l f(v) is the difference of the Morse characteristic functions between all points in the descending Morse simple complex and the extreme points of the simple complex containing distributed photovoltaic distribution networks; i) is the Morse characteristic function of all points in the descending Morse simple complex and the ascending Morse simple complex; max[f u [(a)] is the maximum value of the Morse characteristic function of the minimum point connected to the saddle point in the ascending Morse complex; min[f d (b) is the minimum value of the Morse characteristic function of the minimum point connected to the saddle point in the descending Morse complex.
[0090] Furthermore, in some embodiments of this application, in order to distinguish between secondary features and erroneous features and stabilize the output results of distributed photovoltaic power output feature extraction, based on the above calculation of the Morse-Smale complex feature importance index for distributed photovoltaic grid-connected points, the mean value of feature importance is calculated as the first-stage importance index of the Morse-Smale complex, expressed as:
[0091]
[0092] In the formula, F l 1 It is a first-stage importance indicator of the Morse-Smale complex; N is the number of points on the descending Morse simple complex and the ascending Morse simple complex, that is, the total number of distributed grid connection points extracted from the Morse complex.
[0093] Substituting equation (2) into equation (1) for correction, the characteristic importance index of all distributed photovoltaic grid-connected points extracted from the descending Morse simple complex and the ascending Morse simple complex is calculated and expressed as:
[0094]
[0095] In the formula, F l '(n l This involves incorporating the first-stage importance index of the Morse-Smale complex into F. l (n l After that, the difference between the Morse characteristic function of all points in the descending Morse simple complex and the ascending Morse simple complex containing the distributed photovoltaic distribution network and the extreme point of the simple complex is recalculated, which represents the characteristic importance index of all the distributed photovoltaic grid-connected points extracted from the descending Morse simple complex and the ascending Morse simple complex.
[0096] Finally, the importance index of the Morse-Smale complex feature in the distributed photovoltaic distribution network is updated again, and the two-stage importance index of the Morse-Smale complex is calculated. Based on this, the topology is simplified and expressed as:
[0097]
[0098] In the formula, F l 1 It is the two-stage importance indicator of the updated Morse-Smale complex; N is the number of points on the descending Morse simple complex and the ascending Morse simple complex, that is, the total number of distributed grid connection points extracted from the Morse complex.
[0099] Figure 3 The diagram shows the effect of topology simplification in a large-scale distributed photovoltaic power distribution network. The black curve in the diagram illustrates the effect of topology simplification.
[0100] Please see further. Figure 1 , Figure 2 and Figure 4 :
[0101] S300: For key grid connection points in a simplified topology of a large-scale distributed photovoltaic (PV) distribution network, multi-source measurement data is acquired, typical output scenarios are constructed considering the probability distribution characteristics of distributed PV power and the correlation with typical meteorological information, the sensitivity of typical output scenarios to the voltage of nodes in the distribution network is calculated, and the voltage sensitivity expectation matrix under each typical output scenario is obtained. Based on the voltage sensitivity expectation matrix, sub-clusters are generated by affine propagation clustering, and simplified clustering of large-scale distributed PV for weak measurement scenarios is completed.
[0102] In the specific implementation, in terms of selecting typical power output scenarios, four meteorological conditions are mainly considered: sunny days, rainy days, cloudy days, and strong winds. Based on the historical power statistics of distributed photovoltaic power at key grid connection points in a certain region, a frequency histogram is established, and the data at the midpoint of each interval of the histogram constitutes a set of typical power output scenarios.
[0103] like Figure 2 As shown, based on this, the sensitivity of the voltage amplitude at node i in the distribution network to the injected power under a typical power output scenario at the key grid connection point j of distributed photovoltaic power is calculated, and expressed as:
[0104]
[0105] In the formula, It is the voltage amplitude at node i responding to the voltage change of the injected power in a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation. It is the change in injected power at a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation.
[0106] The voltage sensitivity of injected power under typical power output scenarios at key grid-connected points of distributed photovoltaic systems is defined as follows:
[0107] c ij =u ii +u jj-u ij -u ji i,j=1,2,...,n (6)
[0108]
[0109] In the formula, c ij It is the comprehensive voltage sensitivity of the voltage amplitude at each node i in the distribution network to the injected power under a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation. It is composed of the sensitivity of node i and node j themselves, as well as the mutual sensitivity between node i and node j. C ij It is c ij The matrix formed; u ii It is the sensitivity of the voltage amplitude at node i to the injected power in a typical power output scenario at the key grid connection point i of distributed photovoltaic power generation; u jj It is the sensitivity of the voltage amplitude at node j to the injected power in a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation; u ij It is the sensitivity of the voltage amplitude at node i to the injected power in a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation; u ji It is the sensitivity of the voltage amplitude at node j to the injected power in a typical power output scenario at the key grid connection point i of distributed photovoltaic power generation.
[0110] The voltage sensitivity expectation matrix is constructed as the basis for dividing large-scale distributed photovoltaic sub-clusters. The voltage sensitivity expectation matrix is defined as follows:
[0111]
[0112] In the formula, M represents the number of typical power output scenarios; p m It represents the probability of a typical power output scenario m occurring. It is the expected voltage sensitivity of the voltage amplitude at node i to the injected power in a typical power output scenario m at the key grid connection point j of distributed photovoltaic power generation.
[0113] Furthermore, in some embodiments of this application, the algorithm is initialized based on the affine propagation clustering algorithm. The set of real-valued similarities between data points in the voltage sensitivity expectation matrix is used as the initial similarity matrix as the algorithm input. The similarity matrix s(i,k) represents the degree to which data point k is suitable as the cluster center of data point i, and is calculated using Euclidean distance.
[0114] The similarity matrix s(i,k) is represented as:
[0115] s(i,k)=-d 2 (x(i),x(k))=-||x(i)-x(k)|| 2 (9)
[0116] In the formula, s(i,k) is a similarity matrix representing the degree to which data point k is suitable as the cluster center of data point i; d 2 (x(i), x(k)) is the square of the Euclidean distance between data point k and data point i, which can also be written as ||x(i)-x(k)|| 2 In the form of , both have the same mathematical meaning.
[0117] Two types of information are transmitted between the data points: the attraction information matrix r(i,k) and the belonging information matrix a(i,k). To select suitable cluster centers, the algorithm continuously updates these two information matrices based on the data points.
[0118] The attraction information matrix r(i,k) of candidate cluster center x(k) to any data point x(i) represents the degree to which k is suitable as the cluster center of data point i. The larger the value, the greater the probability that k will become the cluster center.
[0119] The attraction information matrix r(i,k) is represented as:
[0120]
[0121] In the formula, r(i,k) represents the attractiveness of data point k to data point i, indicating the degree of advantage of data point k as a cluster center; s(i,k) represents the similarity of the degree to which data point k is suitable as a cluster center of data point i. This represents the degree of advantage of the strongest competitor among the data points other than data point k as the cluster center. Here, a(i,k') represents the degree of belonging of the data points other than data point k to data point i, and s(i,k') represents the similarity of the degree of the data points other than data point k as the cluster center of data point i, reflecting the competition of the data points other than data point i for data point i.
[0122] The degree of belonging of data point x(i) to candidate cluster center x(k) is a(i,k), with an initial value of 0, which indicates the suitability of i to choose k as its data point. The larger the value, the greater the probability that i belongs to the sub-cluster centered at k.
[0123] The attribution information matrix a(i,k) is represented as:
[0124]
[0125] In the formula, a(i,k) is the degree of belonging of data point k to data point i, reflecting the suitability of data point i in choosing data point k as its cluster center. Here, r(i',k) represents the similarity of data point k as the cluster center of other points besides data point i. It takes all attraction values greater than or equal to 0 and adds the similarity r(k,k) of data point k as the cluster center. That is, the suitability of data point k being chosen by data point i as its cluster center with the support of these data points with attraction values greater than 0. Similarly, a(k,k) is the degree of belonging of data point k itself.
[0126] If both of these pieces of information are relatively large, it indicates that the data point x(k) is more likely to become a cluster center. Through continuous iteration, these two types of information are transferred between data points until the cluster center and the relationship between the data point and the cluster center are determined.
[0127] It should be understood that although the steps in the flowchart are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order constraint on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the diagram may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.
[0128] Please see Figure 5 One embodiment of this application provides a large-scale distributed photovoltaic simplified clustering device for weak measurement scenarios, including a Morse-Smale complex construction module 10, a topology simplification module 20, and a simplified clustering module 30; wherein,
[0129] The Morse-Smale complex construction module 10 uses the similarity of the output characteristics at the grid connection point of distributed photovoltaic in the distribution network as the Morse feature index value, searches for the key points of the Morse feature index value, and constructs a Morse-Smale complex that reflects the output characteristics of distributed photovoltaic.
[0130] The topology simplification module 20 defines the Morse eigenvalue function, uses the average index value of the Morse-Smale complex to back-calculate the characteristic index value of the key points contained in the Morse-Smale complex, iteratively updates the two-stage importance index of the Morse-Smale complex, extracts the key points that are most similar to the output characteristics at the grid connection point of the distributed photovoltaic, and completes the topology simplification of the large-scale distributed photovoltaic distribution network.
[0131] The simplified clustering module 30 acquires multi-source measurement data for key grid-connected points in the simplified topology of a large-scale distributed photovoltaic distribution network, constructs typical output scenarios that take into account the probability distribution characteristics of distributed photovoltaic power and the correlation with typical meteorological information, calculates the sensitivity of typical output scenarios to the voltage of nodes in the distribution network, obtains the voltage sensitivity expectation matrix under each typical output scenario, and generates sub-clusters based on affine propagation clustering according to the voltage sensitivity expectation matrix, thus completing the large-scale distributed photovoltaic simplified clustering for weak measurement scenarios.
[0132] Specific limitations regarding the aforementioned simplified clustering device for large-scale distributed photovoltaic systems in weak measurement scenarios can be found in the limitations of the simplified clustering method for large-scale distributed photovoltaic systems in weak measurement scenarios described above, and will not be repeated here. Each module in the aforementioned simplified clustering device for large-scale distributed photovoltaic systems in weak measurement scenarios can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.
[0133] In another embodiment of this application, a computer device is provided, which may be a server, and its internal structure diagram may be as follows. Figure 6 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores static and dynamic information data. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements the steps in the above method embodiments.
[0134] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the computer device to which the present invention is applied. A specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0135] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the method embodiments described above.
[0136] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.
[0137] This invention is not limited to the structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this invention is limited only by the appended claims.
Claims
1. A simplified clustering method for large-scale distributed photovoltaic systems in weak measurement scenarios, characterized in that, include: Using the similarity of the power output characteristics at the grid connection point of distributed photovoltaic power in the distribution network as the Morse feature index value, the key points of the Morse feature index value are searched to form a Morse-Smale complex that reflects the power output characteristics of distributed photovoltaic power. Define the Morse eigenvalue function, use the average index value of the Morse-Smale complex to back-calculate the eigenvalue index value of the key points contained in the Morse-Smale complex, iteratively update the two-stage importance index of the Morse-Smale complex, extract the key points that are most similar to the output characteristics at the distributed photovoltaic grid connection point, and complete the topology simplification of the large-scale distributed photovoltaic distribution network. For key grid connection points in a simplified topology of a large-scale distributed photovoltaic (PV) distribution network, multi-source measurement data is acquired, and typical output scenarios that take into account the probability distribution characteristics of distributed PV power and the correlation with typical meteorological information are constructed. The sensitivity of the typical output scenarios to the voltage of nodes in the distribution network is calculated, and the voltage sensitivity expectation matrix under each typical output scenario is obtained. Based on the voltage sensitivity expectation matrix, sub-clusters are generated by affine propagation clustering to complete the large-scale distributed PV simplified clustering for weak measurement scenarios. When selecting the typical power output scenario, a frequency histogram is established based on the historical power statistics of the distributed photovoltaic at the key grid connection point, and the data at the midpoint of each interval of the frequency histogram constitutes the typical power output scenario set. The steps for calculating the sensitivity of the typical power output scenario to the node voltage in the distribution network include: The sensitivity of the voltage amplitude at node i in the distribution network to the injected power under a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation is calculated and expressed as: In the formula, It represents the sensitivity of the voltage amplitude at node i in the distribution network to the injected power in a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation. It is the voltage amplitude at node i responding to the voltage change of the injected power in a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation. It is the change in injected power at a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation; The voltage sensitivity of injected power under typical power output scenarios at key grid-connected points of distributed photovoltaic systems is defined as follows: In the formula, It is the comprehensive voltage sensitivity of the voltage amplitude at each node i in the distribution network to the injected power in a typical power output scenario at the key grid-connected point j of distributed photovoltaic power generation. It is composed of the sensitivity of node i and node j themselves, as well as the mutual sensitivity between node i and node j. It is by The matrix formed; It is the sensitivity of the voltage amplitude at node i to the injected power in a typical power output scenario at the key grid connection point i of distributed photovoltaic power generation; It is the sensitivity of the voltage amplitude at node j to the injected power in a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation; It is the sensitivity of the voltage amplitude at node i to the injected power in a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation; It is the sensitivity of the voltage amplitude at node j to the injected power in a typical power output scenario at the key grid connection point i of distributed photovoltaic power generation.
2. The simplified clustering method for large-scale distributed photovoltaic systems in weak measurement scenarios according to claim 1, characterized in that, The step of searching for key points of the Morse feature index, using the similarity of power output characteristics at distributed photovoltaic grid connection points in the distribution network as the Morse feature index value, further includes: Using the grid connection points of the distributed photovoltaic power generation system in the regional power grid as the basic data structure, and the similarity of the output characteristics of the distributed photovoltaic power generation system as the Morse feature index value, the key points are searched using the neighbor comparison method.
3. The simplified clustering method for large-scale distributed photovoltaic systems in weak measurement scenarios according to claim 2, characterized in that, The step of searching for the key points using the neighbor comparison method further includes: For any one of the distributed photovoltaic grid-connected points n, compare its Morse characteristic index value with that of the adjacent distributed photovoltaic grid-connected points ni; If all eigenvalues of n are greater than the eigenvalues of ni, then n is a maximum point; if all eigenvalues of n are less than the eigenvalues of ni, then n is a minimum point; if the difference between the eigenvalues of n and ni is 5, then it is determined to be a saddle point.
4. The simplified clustering method for large-scale distributed photovoltaic systems in weak measurement scenarios according to claim 3, characterized in that, The steps of constructing the Morse-Smale complex, which reflects the characteristics of distributed photovoltaic power output, further include: Starting from each saddle point, proceed along the direction of the minimum or maximum gradient of the eigenvalue until the next extreme point is found; when the saddle point corresponds to a pair of the maximum points, a descending Morse simple complex is formed; when the saddle point corresponds to a pair of the minimum points, an ascending Morse simple complex is formed; according to the duality of the Morse-Smale complex, the descending Morse simple complex and the ascending Morse simple complex are pairwise orthogonal to form the Morse-Smale complex of the distributed photovoltaic power output characteristics.
5. The simplified clustering method for large-scale distributed photovoltaic systems in weak measurement scenarios according to claim 4, characterized in that, The steps of defining a Morse eigenvalue function, using the average index value of the Morse-Smale complex to back-calculate the feature index values of the key points contained in the Morse-Smale complex, iteratively updating the two-stage importance index of the Morse-Smale complex, and extracting the key points most similar to the output characteristics at the distributed photovoltaic grid connection point to complete the topology simplification of the large-scale distributed photovoltaic distribution network further include: Introducing an eigenvalue function, the feature importance index for all distributed photovoltaic grid-connected points extracted from the descending Morse simple complex and the ascending Morse simple complex is calculated and expressed as: In the formula, It is the difference of the Morse characteristic function between all points in the descending Morse simple complex and the extreme point of the simple complex in the distributed photovoltaic distribution network; It is the Morse characteristic function of all points in the descending Morse simple complex and the ascending Morse simple complex; It is the maximum value of the Morse characteristic function of the minimum point connected to the saddle point in the ascending Morse complex; is the minimum value of the Morse characteristic function of the minimum point connected to the saddle point in the descending Morse complex; N is the number of points on the descending Morse simple complex and the ascending Morse simple complex, that is, the total number of distributed grid connection points extracted from the Morse complex.
6. The simplified clustering method for large-scale distributed photovoltaic systems in weak measurement scenarios according to claim 5, characterized in that, The steps of defining a Morse eigenvalue function, using the average index value of the Morse-Smale complex to back-calculate the feature index values of the key points contained in the Morse-Smale complex, iteratively updating the two-stage importance index of the Morse-Smale complex, and extracting the key points most similar to the output characteristics at the distributed photovoltaic grid connection point to complete the topology simplification of the large-scale distributed photovoltaic distribution network further include: Based on the calculation of the importance indices of all distributed photovoltaic grid-connected points extracted from the descending Morse simple complex and the ascending Morse simple complex, the mean of the importance indices is calculated as the first-stage importance index of the Morse-Smale complex, expressed as: In the formula, It is a first-stage importance indicator of the Morse-Smale complex.
7. The simplified clustering method for large-scale distributed photovoltaic systems in weak measurement scenarios according to claim 6, characterized in that, The steps of defining a Morse eigenvalue function, using the average index value of the Morse-Smale complex to back-calculate the feature index values of the key points contained in the Morse-Smale complex, iteratively updating the two-stage importance index of the Morse-Smale complex, and extracting the key points most similar to the output characteristics at the distributed photovoltaic grid connection point to complete the topology simplification of the large-scale distributed photovoltaic distribution network further include: The feature importance index of all distributed photovoltaic grid-connected points extracted from the descending Morse simple complex and the ascending Morse simple complex is expressed as: In the formula, This involves incorporating the first-stage importance index of the Morse-Smale complex. Then, the difference between the Morse characteristic function of all points in the descending Morse simple complex and the ascending Morse simple complex containing the distributed photovoltaic distribution network and the extreme point of the simple complex is recalculated, which represents the characteristic importance index of all the distributed photovoltaic grid-connected points extracted from the descending Morse simple complex and the ascending Morse simple complex.
8. The simplified clustering method for large-scale distributed photovoltaic systems in weak measurement scenarios according to claim 7, characterized in that, The steps of defining a Morse eigenvalue function, using the average index value of the Morse-Smale complex to back-calculate the feature index values of the key points contained in the Morse-Smale complex, iteratively updating the two-stage importance index of the Morse-Smale complex, and extracting the key points most similar to the output characteristics at the distributed photovoltaic grid connection point to complete the topology simplification of the large-scale distributed photovoltaic distribution network further include: The importance index of the Morse-Smale complex feature in the distributed photovoltaic distribution network is updated, and the two-stage importance index of the Morse-Smale complex is calculated. This index is then used for topology simplification, and is expressed as follows: In the formula, It is the updated two-stage importance index for the Morse-Smale complex.
9. The simplified clustering method for large-scale distributed photovoltaic systems in weak measurement scenarios according to claim 1, characterized in that, The step of obtaining the voltage sensitivity expectation matrix under each of the typical output scenarios further includes: The voltage sensitivity expectation matrix is defined as: In the formula, This is the voltage sensitivity expectation matrix, where M is the number of typical power output scenarios; It represents the probability of a typical power output scenario m occurring. It is the expected voltage sensitivity of the voltage amplitude at node i to the injected power in a typical power output scenario m at the key grid connection point j of distributed photovoltaic power generation.
10. The simplified clustering method for large-scale distributed photovoltaic systems in weak measurement scenarios according to claim 1, characterized in that, The steps for generating sub-clusters based on affine propagation clustering according to the voltage sensitivity expectation matrix, and completing the simplified clustering of large-scale distributed photovoltaic systems for weak measurement scenarios, further include: The algorithm is initialized based on the affine propagation clustering algorithm. The set of real-valued similarities between data points in the voltage sensitivity expectation matrix is used as the initial similarity matrix and is used as the input of the affine propagation clustering algorithm. The similarity matrix s(i,k) represents the degree to which data point k is suitable as the cluster center of data point i, and is calculated using Euclidean distance. The similarity matrix s(i,k) is represented as: In the formula, It is the square of the Euclidean distance between data point k and data point i.
11. The simplified clustering method for large-scale distributed photovoltaic systems in weak measurement scenarios according to claim 1, characterized in that, The steps for generating sub-clusters based on affine propagation clustering according to the voltage sensitivity expectation matrix, and completing the simplified clustering of large-scale distributed photovoltaic systems for weak measurement scenarios, further include: The attraction information matrix r(i,k) and the belonging information matrix a(i,k) are transmitted between each data point. The affine propagation clustering algorithm updates the attraction information matrix r(i,k) and the belonging information matrix a(i,k) according to the data points to perform iterations until the cluster center is determined and the belonging relationship between the data points and the cluster center is determined. The attraction information matrix r(i,k) is represented as: In the formula, This indicates the degree of advantage of the strongest competitor among the data points other than data point k as the cluster center; The affiliation information matrix a(i,k) is represented as: In the formula, This represents the similarity of data point k as the cluster center for all points other than data point i. It is calculated by adding all attraction values greater than or equal to 0 to the similarity score of data point k as the cluster center. That is, the degree to which data point k is suitable to be selected as its cluster center by data point i, given the support of data points with an attraction value greater than 0. It is the degree of belonging to the data point k itself.
12. A simplified clustering device for large-scale distributed photovoltaic systems in weak measurement scenarios, characterized in that, It includes a Morse-Smale complex building module, a topology simplification module, and a simplified clustering module; among which, The Morse-Smale complex construction module uses the similarity of the output characteristics at the grid connection point of distributed photovoltaic power in the distribution network as the Morse feature index value, searches for the key points of the Morse feature index value, and constructs a Morse-Smale complex that reflects the output characteristics of distributed photovoltaic power. The topology simplification module defines the Morse eigenvalue function, uses the average index value of the Morse-Smale complex to back-calculate the characteristic index value of the key points contained in the Morse-Smale complex, iteratively updates the two-stage importance index of the Morse-Smale complex, extracts the key points most similar to the output characteristics at the distributed photovoltaic grid connection point, and completes the topology simplification of the large-scale distributed photovoltaic distribution network. A simplified clustering module acquires multi-source measurement data for key grid-connected points in a simplified topology of a large-scale distributed photovoltaic (PV) distribution network. It constructs typical output scenarios that consider the probability distribution characteristics of distributed PV power and their correlation with typical meteorological information. The module calculates the sensitivity of these typical output scenarios to the voltage of nodes in the distribution network, obtaining the voltage sensitivity expectation matrix for each typical output scenario. Based on this voltage sensitivity expectation matrix, sub-clusters are generated using affine propagation clustering, completing the simplified clustering of large-scale distributed PV for weak measurement scenarios. When selecting typical output scenarios, a frequency histogram is established based on the historical power statistics of the distributed PV at key grid-connected points. Data at the midpoint of each interval of the frequency histogram constitutes a set of typical output scenarios. Calculating the sensitivity of the typical power output scenario to the node voltage in the distribution network includes: The sensitivity of the voltage amplitude at node i in the distribution network to the injected power under a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation is calculated and expressed as: In the formula, It represents the sensitivity of the voltage amplitude at node i in the distribution network to the injected power in a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation. It is the voltage amplitude at node i responding to the voltage change of the injected power in a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation. It is the change in injected power at a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation; The voltage sensitivity of injected power under typical power output scenarios at key grid-connected points of distributed photovoltaic systems is defined as follows: In the formula, It is the comprehensive voltage sensitivity of the voltage amplitude at each node i in the distribution network to the injected power in a typical power output scenario at the key grid-connected point j of distributed photovoltaic power generation. It is composed of the sensitivity of node i and node j themselves, as well as the mutual sensitivity between node i and node j. It is by The matrix formed; It is the sensitivity of the voltage amplitude at node i to the injected power in a typical power output scenario at the key grid connection point i of distributed photovoltaic power generation. It is the sensitivity of the voltage amplitude at node j to the injected power in a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation; It is the sensitivity of the voltage amplitude at node i to the injected power in a typical power output scenario at the key grid connection point j of distributed photovoltaic power generation; It is the sensitivity of the voltage amplitude at node j to the injected power in a typical power output scenario at the key grid connection point i of distributed photovoltaic power generation.
13. The large-scale distributed photovoltaic simplified clustering device for weak measurement scenarios according to claim 12, characterized in that, Using the similarity of power output characteristics at distributed photovoltaic grid connection points in the distribution network as the Morse feature index value, the search for key points of the Morse feature index value further includes: Using the grid connection points of the distributed photovoltaic power generation system in the regional power grid as the basic data structure, and the similarity of the output characteristics of the distributed photovoltaic power generation system as the Morse feature index value, the key points are searched using the neighbor comparison method.
14. The large-scale distributed photovoltaic simplified clustering device for weak measurement scenarios according to claim 13, characterized in that, Searching for the key points using the neighbor comparison method further includes: For any one of the distributed photovoltaic grid-connected points n, compare its Morse characteristic index value with that of the adjacent distributed photovoltaic grid-connected points ni; If all eigenvalues of n are greater than the eigenvalues of ni, then n is a maximum point; if all eigenvalues of n are less than the eigenvalues of ni, then n is a minimum point; if the difference between the eigenvalues of n and ni is 5, then it is determined to be a saddle point.
15. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-11.
16. A computer-readable storage medium, characterized in that, It stores a computer program thereon; the computer program is executed by a processor to implement the method as described in any one of claims 1-11.