A method and system for identifying the location of a user in a district based on massive voltage data

Abstract

The application discloses a transformer area user position identification method and system based on massive voltage data, and belongs to the technical field of smart power grid management. The application obtains voltage time sequence data of all users in a power distribution transformer area under the condition of the same time period and the same sampling frequency, and pre-processes the voltage time sequence data. According to the pre-processed voltage time sequence data, all branch users in each power grid branch in the power distribution transformer area are determined, and the voltage daily fluctuation of each branch user is determined. According to the voltage daily fluctuation, the position index of each branch user is calculated. According to the preset position index threshold interval and the position index, the position of each branch user in each power grid branch in the power distribution transformer area is determined, and each branch user is divided into a head-end user, an intermediate user or an end user, namely, the voltage daily fluctuation and the position index are calculated, the relative position relationship of the user in the feeder is accurately identified, and the low-voltage transformer area state perception ability and the voltage management level are improved.

Description

Technical Field

[0001] This application relates to the field of smart grid management technology, and in particular to a method and system for identifying the location of transformer substation users based on massive voltage data. Background Technology

[0002] In recent years, with the acceleration of the clean energy transition, the penetration rate of distributed energy devices such as distributed photovoltaics and energy storage in low-voltage distribution networks has been increasing, leading to increasingly complex operation patterns in these networks. The integration of distributed energy has altered the power flow distribution and voltage characteristics of traditional distribution networks, posing new challenges to the safe and stable operation of the grid. To achieve the rational allocation and optimized control of distributed energy, user location information is crucial foundational data for distributed energy planning and configuration. Users in different locations exhibit varying voltage levels, power losses, and impacts on system stability due to differences in their electrical distance within the distribution network.

[0003] The large-scale deployment of smart meters provides a rich data foundation for identifying the location of users in low-voltage distribution areas. However, traditional user location identification methods have many limitations when facing complex distributed energy access scenarios. Typically, scalar information such as voltage and current provided by smart meters is used to solve difficult problems by establishing line parameter equations. An iterative solution method is employed to obtain line impedance values, and appropriate algorithms are combined to perform distribution area topology identification. By calculating the correlation between the user and the distribution area voltage, the distribution area and feeder to which the user belongs can be identified.

[0004] Although the above solutions solve the problems of line parameter identification and user's substation and feeder identification, they do not address the precise location of the user's specific position within the feeder, and still cannot achieve the effect of reasonable configuration of distributed energy equipment based on the user's specific location. Summary of the Invention

[0005] The main purpose of this application is to provide a method and system for identifying the location of users in low-voltage distribution areas based on massive voltage data, aiming to solve the technical problem that the lack of accurate user location information in low-voltage distribution areas leads to the inability to properly configure distributed energy equipment.

[0006] To achieve the above objectives, this application provides a method for identifying the location of transformer substation users based on massive voltage data, the method comprising the following steps:

[0007] Obtain voltage timing data of all users within the distribution area under the same time period and the same sampling frequency, and preprocess the voltage timing data;

[0008] Based on the preprocessed voltage time series data, all branch users in each power grid branch within the distribution substation area are identified, and the daily voltage fluctuation of each branch user is determined.

[0009] Calculate the location index for each branch user based on the daily voltage fluctuation.

[0010] Based on the preset location index threshold range and the location index, the location of each branch user in each power grid branch within the distribution area is determined, and each branch user is correspondingly classified as a head user, intermediate user, or end user.

[0011] In one embodiment, the step of determining all branch users within each power grid branch in the distribution substation area based on preprocessed voltage time-series data, and determining the daily voltage fluctuation of each branch user, includes:

[0012] The voltage time series similarity measurement method based on SBD calculates the similarity between each preprocessed voltage time series data, and generates an SBD distance matrix based on the similarity.

[0013] Using the DBSCAN clustering method, based on the SBD distance matrix, all branch users within each power grid branch in the distribution substation are identified, and the daily voltage fluctuation of each branch user is determined.

[0014] In one embodiment, the step of determining the daily voltage fluctuation of each branch user includes:

[0015] Determine the initial values ​​of daily voltage fluctuations for each branch user during the daytime and nighttime periods respectively;

[0016] Historical voltage fluctuation values ​​for daytime and nighttime periods are obtained separately, and the historical voltage fluctuation values ​​and the initial daily voltage fluctuation values ​​are fused together to obtain the daily voltage fluctuation amount for each branch user.

[0017] In one embodiment, the step of determining the initial values ​​of daily voltage fluctuations for each branch user during the daytime and nighttime periods includes:

[0018] The initial value of daily voltage fluctuation for each branch user during the daytime period is calculated using a calculation method based on voltage range or the difference between high and low quantiles, in order to capture voltage fluctuations caused by drastic changes in photovoltaic power during the daytime period.

[0019] The initial value of the daily voltage fluctuation of each branch user during the nighttime period is calculated using a calculation method based on voltage standard deviation, so as to reflect the continuous voltage deviation caused by changes in base load during the nighttime period.

[0020] In one embodiment, the step of calculating the location index of each branch user based on the daily voltage fluctuation includes:

[0021] Based on the daily voltage fluctuation, the minimum and maximum daily fluctuation of each branch user during the daytime and nighttime periods are determined respectively.

[0022] Based on the minimum and maximum daily fluctuations during the daytime and nighttime periods, the normalized values ​​for each branch user during the daytime and nighttime periods are calculated respectively.

[0023] Dynamic weights are set based on the photovoltaic penetration rate level or seasonal characteristics of the transformer area, and the normalized values ​​of the daytime and nighttime periods are weighted and fused to calculate the location index of each branch user.

[0024] In one embodiment, after determining the location of each branch user within each power grid branch in the distribution substation based on a preset location index threshold range and the location index, and classifying each branch user into first-end users, intermediate users, or last-end users, the method further includes:

[0025] The consistency of the daytime and nighttime location indices for users in the same branch is calculated.

[0026] Calculate the standard deviation of the location index of the same branch user over multiple days, and determine the stability of the same branch user over multiple days based on the standard deviation;

[0027] Based on the consistency of the time period and the stability of the multi-day time period, the first-end users, intermediate users, or last-end users obtained from the division are verified.

[0028] In one embodiment, after the step of classifying the branch users into first-end users, intermediate users, or last-end users, the method further includes:

[0029] Based on the relative positions between the first-end users, intermediate users, and last-end users obtained from the division, a dynamic voltage state vector is constructed for each user. The dynamic voltage state vector includes the voltage and current correlation analysis of the user, as well as the equivalent impedance of the distribution transformer from the user to the distribution substation.

[0030] Based on the dynamic voltage state vector, the dominant voltage fluctuation mode and key propagation path of the distribution substation during operation are extracted;

[0031] Based on the dominant voltage fluctuation pattern and key propagation path, a directed graph of user voltage interaction is constructed, wherein the nodes in the directed graph of user voltage interaction are users, and the edge weights between the nodes are the relationship strength of voltage fluctuations between users.

[0032] In one embodiment, after the step of constructing a directed graph of user voltage interaction based on the dominant voltage fluctuation pattern and the critical propagation path, the method further includes:

[0033] If a voltage disturbance source user is identified in the directed graph of user voltage interaction, its subsequent behavior is predicted based on the historical disturbance pattern of the voltage disturbance source user, and corresponding load adjustment suggestions are sent to the power grid dispatching system in advance based on the prediction results.

[0034] If a voltage-vulnerable user is identified within the directed graph of user voltage interactions, the voltage stability margin index of the voltage-vulnerable user is determined. When the voltage stability margin index is lower than a threshold, adjustable resources adjacent to the voltage-vulnerable user are determined and activated according to the directed graph of user voltage interactions. The voltage stability margin index is calculated from the rate of change norm of the dynamic voltage state vector of the voltage-vulnerable user.

[0035] In one embodiment, before the steps of determining all branch users within each grid branch in the distribution substation area based on preprocessed voltage time-series data, and determining the daily voltage fluctuation of each branch user, the method further includes:

[0036] Obtain the self-voltage time series of the distribution transformer in the distribution area, sort the self-voltage time series according to the voltage value, and take the voltage value in the 20th percentile as the voltage threshold.

[0037] The preprocessed voltage time series data of the branch users corresponding to the time periods below the voltage threshold are filtered.

[0038] Furthermore, to achieve the above objectives, this application also provides a transformer substation user location identification system based on massive voltage data, the system comprising:

[0039] The data processing module is used to acquire voltage timing data of all users in the distribution area under the same time period and the same sampling frequency, and to preprocess the voltage timing data.

[0040] The fluctuation determination module is used to determine all branch users in each power grid branch within the distribution substation area based on the preprocessed voltage time series data, and to determine the daily voltage fluctuation of each branch user.

[0041] The index calculation module is used to calculate the location index of each branch user based on the daily voltage fluctuation.

[0042] The location division module is used to determine the location of each branch user in each power grid branch within the distribution area based on a preset location index threshold range and the location index, and to classify each branch user into a head user, intermediate user, or end user.

[0043] One or more technical solutions proposed in this application have at least the following technical effects: By acquiring voltage time-series data of all users within a distribution transformer area under the same time period and sampling frequency, and preprocessing the voltage time-series data; based on the preprocessed voltage time-series data, determining the branch to which each user belongs within the distribution transformer area, and determining the daily voltage fluctuation of each user within each branch; calculating the location index of each user based on the daily voltage fluctuation; determining the location of each user within the distribution transformer area based on a preset location index threshold range and the location index, and correspondingly classifying each user into the distribution transformer area. By deeply analyzing the transmission mechanism of voltage fluctuations in the distribution network and making full use of the existing voltage monitoring data of smart meters, the specific location of each user in the distribution transformer area is determined based on the daily voltage fluctuation and location index of each user in the transformer area, which is based on massive voltage time series data. This allows for accurate identification of the relative position of users in the feeder without increasing additional hardware costs. This solves the technical problem of the lack of accurate user location information in low-voltage transformer areas, which leads to the inability to rationally configure distributed energy equipment. It provides technical support for the optimized configuration, operation control, and safe and stable operation of distributed energy in smart distribution networks. Attached Figure Description

[0044] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0045] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a flowchart illustrating the first embodiment of the method for identifying the location of transformer users based on massive voltage data in this application;

[0047] Figure 2 This is a flowchart illustrating the second embodiment of the method for identifying the location of transformer users based on massive voltage data in this application.

[0048] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0049] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0050] Reference Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the method for identifying the location of transformer users based on massive voltage data in this application.

[0051] In this embodiment, a method for identifying the location of transformer substation users based on massive voltage data is proposed. The method includes the following steps:

[0052] S10: Obtain voltage timing data of all users in the distribution area under the same time period and the same sampling frequency, and preprocess the voltage timing data.

[0053] S20, Based on the preprocessed voltage time series data, determine all branch users in each power grid branch within the distribution substation area, and determine the daily voltage fluctuation of each branch user;

[0054] S30, calculate the location index of each branch user based on the daily voltage fluctuation;

[0055] S40, based on the preset location index threshold range and the location index, determine the location of each branch user in each power grid branch within the distribution area, and classify each branch user into first-end users, intermediate users, or end users.

[0056] Understandably, the large-scale deployment of smart meters provides a rich data foundation for identifying the location of users in low-voltage distribution areas. However, traditional user location identification methods have many limitations when facing complex distributed energy access scenarios. Typically, scalar information such as voltage and current provided by smart meters is used to solve difficult problems by establishing line parameter equations. Iterative solutions are then used to obtain line impedance values, and corresponding algorithms are combined to perform distribution area topology identification. By calculating the correlation between the user and the distribution area voltage, the distribution area and feeder to which the user belongs are identified. However, the above scheme cannot accurately identify the specific location of each user, and therefore cannot achieve the effect of reasonable configuration of distributed energy equipment based on the user's specific location. In this embodiment, a complete system for identifying the detailed electrical location of users in distribution areas is constructed. It mainly utilizes voltage time-series data collected within the same time period, analyzes the correlation between user voltage fluctuation characteristics and the grid branch structure, and establishes a mapping model from voltage fluctuation patterns to physical location, thereby achieving accurate identification of the user's electrical location.

[0057] Understandably, in order to establish a standardized data acquisition process and ensure the accuracy and reliability of subsequent analysis, this embodiment acquires voltage timing data of all users within the distribution area under the same time period and sampling frequency. For example, from 00:00 to 24:00 on June 1, 2024, voltage data of all users are synchronously acquired at a frequency of once per minute.

[0058] The data needs to be processed accordingly. The voltage time series data needs to be preprocessed, including data cleaning (removing obvious outliers), missing value imputation (using linear interpolation), data alignment (unifying timestamps), and normalization.

[0059] For example, when detecting whether there is missing data in voltage time series data, if there is, Lagrange interpolation is used to complete the data so that the voltage time series maintains the continuity of the trend and the same length.

[0060] Among them, Lagrange interpolation is a classic numerical interpolation method used to construct a polynomial function based on known data points, so that the function can pass through all known data points accurately. This method can be found in existing technologies and will not be elaborated here.

[0061] Understandably, the similarity of voltage fluctuation patterns can be used to identify user groups on the same branch line. Based on preprocessed voltage time-series data, all branch users within each power grid branch in the distribution substation can be determined. By using clustering algorithms to group users with similar voltage fluctuation patterns into the same branch, the daily voltage fluctuation of each branch user can be determined. This indicator reflects the amplitude and pattern of voltage changes within a day.

[0062] Voltage time series data refers to a sequence of voltage measurements arranged in chronological order, with a sampling frequency typically ranging from 1 to 15 minutes per sampling, and includes information such as the effective voltage value and phase angle.

[0063] Among them, daily voltage fluctuation refers to a quantitative indicator that characterizes the degree of change in a user's voltage within a day. It can be calculated using the range method, standard deviation method, or quantile difference method, depending on the characteristics of the data.

[0064] It should be noted that, based on the electricity consumption patterns of users in the low-voltage distribution area at different times, voltage data for the period from 23:00 to 7:00, where voltage changes slowly, can be deleted, while data for periods with high load connections and large voltage changes can be retained. The remaining time series data are then connected sequentially to form new voltage time series data, thereby improving the reliability of branch relationship identification.

[0065] In this embodiment, to determine the precise location of a user on different branches, a corresponding calculation logic is designed based on the daily voltage fluctuations to calculate the user's location index within the line. By constructing a comprehensive index characterizing the user's electrical location within the line, the location index for each branch user is calculated based on the daily voltage fluctuations. This index comprehensively considers the voltage fluctuation characteristics at different times of day and night, and achieves accurate classification of the user's electrical location based on the location index.

[0066] Specifically, based on the preset location index threshold range and the location index, the location of each branch user in each power grid branch within the distribution area is determined, and each branch user is correspondingly classified as a head user, intermediate user, or end user.

[0067] The location index is a normalized index that comprehensively reflects the electrical location of a user on the line. It is a continuous value between 0 and 1, with a larger value indicating that the user is closer to the end of the line.

[0068] Among them, the classification of users at the beginning / middle / end is based on the electrical distance. The voltage quality of users at the beginning is the best, while the voltage fluctuation of users at the end is the most obvious.

[0069] Specifically, in this embodiment, the location index is set to a continuous number from 0 to 1. Assuming that the corresponding preset location index threshold range is divided as follows, the range [0, 0.3] is for first-end users, [0.3, 0.7] is for intermediate users, and [0.7, 1.0] is for end users.

[0070] In this embodiment, after the step of classifying the branch users into first-end users, intermediate users, or last-end users, the method further includes:

[0071] Based on the relative positions between the first-end users, intermediate users, and last-end users obtained from the division, a dynamic voltage state vector is constructed for each user. The dynamic voltage state vector includes the voltage and current correlation analysis of the user, as well as the equivalent impedance of the distribution transformer from the user to the distribution substation.

[0072] Based on the dynamic voltage state vector, the dominant voltage fluctuation mode and key propagation path of the distribution substation during operation are extracted;

[0073] Based on the dominant voltage fluctuation pattern and key propagation path, a directed graph of user voltage interaction is constructed, wherein the nodes in the directed graph of user voltage interaction are users, and the edge weights between the nodes are the relationship strength of voltage fluctuations between users.

[0074] In this embodiment, based on user location identification, in order to further explore the voltage interaction relationship between users, the dynamic voltage state vector of each user is determined, the dominant voltage fluctuation mode and key propagation path of the transformer area are identified, and a directed graph of user voltage interaction is formed, providing a clear and detailed framework for voltage control.

[0075] It is understandable that when establishing a feature vector that comprehensively describes the voltage state of users, a dynamic voltage state vector is constructed for each user based on the relative positions between the first-end users, intermediate users, and last-end users obtained from the division. The dynamic voltage state vector includes the voltage and current correlation analysis of the user, as well as the equivalent impedance of the distribution transformer from the user to the distribution substation.

[0076] Among them, the dynamic voltage state vector is a multi-dimensional feature vector used to describe the dynamic characteristics of user voltage. Its corresponding vector dimensions include multiple dimensions such as voltage amplitude, phase, harmonic content, and fluctuation characteristics.

[0077] Among them, the dominant voltage fluctuation mode refers to the dominant voltage fluctuation pattern and propagation law in the distribution substation area, which can be identified by spectrum analysis, mode decomposition or machine learning methods.

[0078] Specifically, the identification of the dominant voltage mode requires extracting the dominant voltage change mode of the distribution area from the user voltage data, extracting the dominant voltage fluctuation mode and key propagation path of the distribution area in operation based on the dynamic voltage state vector, and using principal component analysis or mode decomposition methods to identify the main voltage fluctuation mode and its propagation characteristics.

[0079] Furthermore, based on the data obtained after the above processing, a directed voltage interaction graph can be constructed to visualize and quantify the voltage influence relationship between users. In this process, the dominant voltage fluctuation pattern and key propagation path are mainly used to determine the connection relationship and voltage change between users in the power grid, and a directed voltage interaction graph of users is constructed. In this graph, the nodes of the directed voltage interaction graph of users are users, and the edge weights between the nodes are the relationship strength of voltage fluctuations between users.

[0080] For example, in a certain distribution area, the voltage fluctuations of user H (a major photovoltaic user) will propagate along a specific path (from the first user through intermediate users to the last user), affecting downstream users I and J. In the directed graph of voltage interaction, the edge weight from H to I is 0.8, and the edge weight from H to J is 0.6, indicating that H has a stronger influence on I.

[0081] In this embodiment, after the step of constructing a directed graph of user voltage interaction based on the dominant voltage fluctuation mode and the key propagation path, the method further includes:

[0082] If a voltage disturbance source user is identified in the directed graph of user voltage interaction, its subsequent behavior is predicted based on the historical disturbance pattern of the voltage disturbance source user, and corresponding load adjustment suggestions are sent to the power grid dispatching system in advance based on the prediction results.

[0083] If a voltage-vulnerable user is identified within the directed graph of user voltage interactions, the voltage stability margin index of the voltage-vulnerable user is determined. When the voltage stability margin index is lower than a threshold, adjustable resources adjacent to the voltage-vulnerable user are determined and activated according to the directed graph of user voltage interactions. The voltage stability margin index is calculated from the rate of change norm of the dynamic voltage state vector of the voltage-vulnerable user.

[0084] In this embodiment, a control scheme that enables proactive early warning and corresponding resource allocation can be implemented based on the voltage interaction graph, transforming from situational awareness to proactive control. The system identifies disturbance source users and vulnerable users based on the directed voltage interaction graph, and takes control measures in advance based on the prediction results, thereby achieving the effects of early detection, early warning, and early handling of voltage problems.

[0085] Among them, voltage disturbance source users refer to users whose electricity consumption behavior has a significant impact on the voltage quality of the distribution area, such as large-capacity distributed photovoltaic users and impulsive load users. In the directed graph of user voltage interaction, the out-degree (the sum of the weights of the edges pointing to other nodes) and in-degree (the sum of the weights of the edges pointed to by other nodes) of each node are calculated. Voltage disturbance source users usually have a high out-degree, that is, they have a greater impact on other users. Specifically, an appropriate threshold can be set to select the top few users in terms of out-degree or users with a high ratio of out-degree to in-degree as voltage disturbance source users.

[0086] Among them, voltage-vulnerable users refer to users who are easily affected by voltage fluctuations of other users and have poor voltage stability. In the directed graph of user voltage interaction, voltage-vulnerable users usually have a high in-degree, that is, they are greatly affected by other users. Combined with the voltage stability margin index for identification, users with high in-degree and low voltage stability margin index are selected as voltage-vulnerable users.

[0087] The voltage stability margin index is an indicator that quantifies the degree of voltage stability of a user. It can be calculated based on the norm of the rate of change of the state vector. The smaller the value, the closer it is to the stability boundary. The voltage stability margin index is calculated from the norm of the rate of change of the dynamic voltage state vector of the voltage vulnerable user. Specifically, the Euclidean norm or Mahalanobis distance can be used to calculate the degree of change of the state vector. The specific calculation content can be referred to the existing technology, and will not be elaborated here.

[0088] The process involves identifying disturbance sources and assessing vulnerability separately. When identifying and predicting disturbance sources, voltage disturbance sources are identified and their behavioral trends are predicted. If a voltage disturbance source user is identified within the directed graph of user voltage interactions, the user is marked and its disturbance characteristics are analyzed. Furthermore, the user's subsequent behavior is predicted based on its historical disturbance patterns.

[0089] Among them, time series forecasting algorithms are used to predict future disturbance trends, and based on the forecast results, corresponding load adjustment suggestions are sent to the power grid dispatching system in advance, such as suggesting adjusting the output of distributed power sources or switching capacitors.

[0090] The process involves separate identification of disturbance sources and vulnerability assessment. During vulnerability assessment, voltage-vulnerable users are identified and their stability margins are evaluated. If a voltage-vulnerable user is identified within the directed voltage interaction graph, its voltage state is closely monitored to determine its voltage stability margin index. When the voltage stability margin index falls below a threshold, an early warning signal is issued. Based on the directed voltage interaction graph, adjustable resources adjacent to the voltage-vulnerable user are identified and activated, such as adjusting the controllable loads or energy storage devices of adjacent users.

[0091] For example, user K is identified as a voltage-vulnerable user with a voltage stability margin index of 0.35 (threshold 0.4). The energy storage device of neighboring user L is automatically activated to provide reactive power support to node K, raising the margin index to 0.45.

[0092] In this embodiment, before the steps of determining all branch users within each power grid branch in the distribution substation area based on the preprocessed voltage time-series data, and determining the daily voltage fluctuation of each branch user, the method further includes:

[0093] Obtain the self-voltage time series of the distribution transformer in the distribution area, sort the self-voltage time series according to the voltage value, and take the voltage value in the 20th percentile as the voltage threshold.

[0094] The preprocessed voltage time series data of the branch users corresponding to the time periods below the voltage threshold are filtered.

[0095] It is understandable that when the voltage of the distribution transformers in each branch of the distribution area is low, the line voltage drop effect is obvious, and the voltage difference between users at different locations is amplified (especially the difference between end users and beginning users). Therefore, the voltage time series data under low voltage conditions is more conducive to location identification. When obtaining the voltage time series data after prediction processing, the data where the transformer voltage is lower than the corresponding threshold should be retained.

[0096] Specifically, in this embodiment, the key data in the voltage time series data is screened mainly by analyzing the voltage time series corresponding to the distribution transformer voltage, identifying low voltage periods, and based on this, setting corresponding voltage thresholds for screening data. The user voltage data in the time period corresponding to the screened data is then used to identify the location, thereby improving the sensitivity and accuracy of the analysis results.

[0097] The voltage threshold determination process mainly relies on the actual distribution characteristics of the distribution transformer voltage within the distribution substation area to determine the low voltage threshold. Specifically, it involves obtaining the self-voltage time series of the distribution transformers within the distribution substation area and sorting the self-voltage time series according to the voltage value, so that the series reflects the overall voltage level of the substation area, the distribution of voltage values, and the data that are actually in the low voltage range.

[0098] In this embodiment, the voltage value in the 20th percentile can be used as the voltage threshold to divide the dataset into two parts. After sorting the data in ascending order, the first 20% of the data should be less than or equal to this value. That is, the voltage is below this value 20% of the time and above this value 80% of the time. Similarly, the voltage threshold can be adjusted according to actual needs, for example, 15% or 30%. The voltage threshold is not specifically limited.

[0099] Specifically, the preprocessed voltage time series data of branch users corresponding to the time periods below the voltage threshold are selected, and the user voltage data of these low voltage periods are used as the key input for subsequent analysis.

[0100] For example, if the 20th percentile of the voltage time series of a certain distribution transformer area is 215V, all time periods with voltages below 215V (such as peak load periods in the morning and evening) are selected, and the voltage data of all users in these time periods are extracted for location identification analysis.

[0101] This embodiment acquires voltage time-series data of all users within a distribution transformer area under the same time period and sampling frequency, and preprocesses the voltage time-series data. Based on the preprocessed voltage time-series data, it determines the branch to which each user belongs within the distribution transformer area and the daily voltage fluctuation of each user within each branch. Based on the daily voltage fluctuation, it calculates the location index of each user. Based on a preset location index threshold range and the location index, it determines the location of each user within the distribution transformer area and classifies each user as a head user, intermediate user, or end user within the distribution transformer area. In other words, by deeply analyzing the transmission mechanism of voltage fluctuations in the distribution network and fully utilizing the existing voltage monitoring data of smart meters, based on the daily voltage fluctuation and location index of each user in the transformer area from massive voltage time-series data, it determines the specific location of each user in the distribution transformer area. Without increasing additional hardware costs, it achieves accurate identification of the relative position of users in the feeder, thereby solving the technical problem of the lack of accurate user location information in low-voltage transformer areas, which leads to the inability to rationally configure distributed energy equipment. This provides technical support for the optimized configuration, operation control, and safe and stable operation of distributed energy in smart distribution networks.

[0102] like Figure 2 As shown, based on the first embodiment, a second embodiment of the method for identifying the location of transformer substation users based on massive voltage data is proposed. In this embodiment, the method further includes:

[0103] S21, Based on the SBD-based voltage time series similarity measurement method, calculate the similarity between each preprocessed voltage time series data, and generate an SBD distance matrix based on the similarity;

[0104] S22. Using the DBSCAN clustering method, based on the SBD distance matrix, identify all branch users within each power grid branch in the distribution substation area, and determine the daily voltage fluctuation of each branch user.

[0105] In this embodiment, the voltage time series similarity measurement method of SBD and the branch user clustering method of DBSCAN, along with the shape-based distance measurement method and density clustering algorithm, are used to accurately identify user groups with similar voltage fluctuation patterns, laying the foundation for subsequent location identification.

[0106] The voltage time series similarity measurement method based on Shape-Based Distance (SBD) calculates the similarity between preprocessed voltage time series data. SBD calculates the optimal distance between two sequences under different phase alignments using a sliding window. By considering the amplitude and phase differences of the sequences, it reflects the essential similarity of voltage fluctuations more effectively than Euclidean distance. Based on this similarity, an SBD distance matrix is ​​generated. Furthermore, the mathematical principles of the SBD algorithm can be found in relevant existing literature and will not be elaborated upon here.

[0107] When SBD=0, it indicates that the two voltage sequences are completely identical; when SBD is close to 1, it indicates that the two voltage sequences are very different; when SBD<0.5, the two sequences are generally considered to have high similarity.

[0108] Among them, the DBSCAN algorithm (Density-Based Spatial Clustering of Applications with Noise) does not require pre-specifying the number of clusters, can identify clusters of arbitrary shapes, and effectively handle noise points. It requires setting two key parameters: neighborhood radius and minimum number of samples. Depending on the size of the distribution area, the neighborhood radius is usually set to 0.1-0.3, and the minimum number of samples is set to 3-5. Using the DBSCAN clustering method, based on the SBD distance matrix, it identifies all branch users in each power grid branch within the distribution area, and determines users in the same cluster as users on the same branch line.

[0109] For example, in a certain distribution area, the SBD distances of the voltage sequences of users A, B, and C are as follows: the distance between A and B is 0.12, the distance between AC is 0.15, and the distance between BC is 0.11. When the neighborhood radius is 0.2 and the minimum number of samples is 2, these three users are clustered into the same branch because the SBD distance between them is smaller than the neighborhood radius and meets the density requirement.

[0110] In this embodiment, the step of determining the daily voltage fluctuation of each branch user includes:

[0111] Determine the initial values ​​of daily voltage fluctuations for each branch user during the daytime and nighttime periods respectively;

[0112] Historical voltage fluctuation values ​​for daytime and nighttime periods are obtained separately, and the historical voltage fluctuation values ​​and the initial daily voltage fluctuation values ​​are fused together to obtain the daily voltage fluctuation amount for each branch user.

[0113] Understandably, when calculating daily voltage fluctuations, it is possible to distinguish between different electricity consumption periods during the day and night, adopt fluctuation calculation strategies suitable for the characteristics of each period, and improve the stability and reliability of the results through multi-day data fusion.

[0114] Based on the differences in electricity consumption characteristics, a day is divided into two typical periods: daytime and nighttime. The daytime period is defined as 7:00-18:00 (the active period of photovoltaic output), and the nighttime period is defined as 18:00-7:00 the next day (the period dominated by base load, where the data in 23:00-7:00 can be deleted and not considered). The initial values ​​of daily voltage fluctuations for each branch user during the daytime and nighttime periods are determined respectively.

[0115] In this embodiment, historical data can be used to smooth daily fluctuations and improve the robustness of the results. Historical voltage fluctuation values ​​corresponding to daytime and nighttime periods are obtained separately. Typically, the historical data of the most recent 7 days are taken, and the historical voltage fluctuation values ​​and the initial daily voltage fluctuation values ​​are fused together using a weighted average method, with recent data having a higher weight.

[0116] For example, user D's initial daily voltage fluctuation on June 1st was 0.035, and the historical daily voltage fluctuation values ​​for the past 7 days were 0.028, 0.031, 0.033, 0.030, 0.032, 0.029, and 0.034, respectively. Specifically, an exponentially weighted average (with a decay factor of 0.8) can be used to fuse these values, resulting in a daily voltage fluctuation of 0.032.

[0117] In this embodiment, the step of determining the initial values ​​of daily voltage fluctuations for each branch user during the daytime and nighttime periods includes:

[0118] The initial value of daily voltage fluctuation for each branch user during the daytime period is calculated using a calculation method based on voltage range or the difference between high and low quantiles, in order to capture voltage fluctuations caused by drastic changes in photovoltaic power during the daytime period.

[0119] The initial value of the daily voltage fluctuation of each branch user during the nighttime period is calculated using a calculation method based on voltage standard deviation, so as to reflect the continuous voltage deviation caused by changes in base load during the nighttime period.

[0120] In this embodiment, the calculation of daily voltage fluctuation is divided into daytime fluctuation calculation and nighttime fluctuation calculation. In the calculation of daily voltage fluctuation, a calculation method based on voltage range or the difference between high and low quantiles is used to calculate the initial value of daily voltage fluctuation for each branch user during the daytime period. Specifically, the range method calculates the difference between the maximum and minimum voltage of the day; the quantile difference method calculates the difference between the 95th quantile and the 5th quantile voltage to capture voltage fluctuations caused by drastic changes in photovoltaic power during the daytime period. In the calculation of nighttime fluctuation, a calculation method based on voltage standard deviation is used to calculate the initial value of daily voltage fluctuation for each branch user during the nighttime period. By calculating the standard deviation of all voltage sampling values ​​during the nighttime period, the continuous voltage deviation caused by changes in base load during the nighttime period is reflected.

[0121] Among them, voltage range refers to the difference between the maximum and minimum voltage values ​​over a period of time, which is suitable for capturing drastic but short-lived voltage fluctuations.

[0122] Among them, voltage standard deviation refers to the statistical measure of the degree to which voltage values ​​deviate from the average value, and is suitable for assessing the continuous fluctuation characteristics of voltage.

[0123] For example, user E's voltage value fluctuates between 218V and 225V during the daytime, with a range of 7V. The normalized initial value for daytime fluctuation is 0.032. During the nighttime, the voltage value fluctuates between 219V and 221V, with a standard deviation of 0.8V. The normalized initial value for nighttime fluctuation is 0.015.

[0124] In this embodiment, the step of calculating the location index of each branch user based on the daily voltage fluctuation includes:

[0125] Based on the daily voltage fluctuation, the minimum and maximum daily fluctuation of each branch user during the daytime and nighttime periods are determined respectively.

[0126] Based on the minimum and maximum daily fluctuations during the daytime and nighttime periods, the normalized values ​​for each branch user during the daytime and nighttime periods are calculated respectively.

[0127] Dynamic weights are set based on the photovoltaic penetration rate level or seasonal characteristics of the transformer area, and the normalized values ​​of the daytime and nighttime periods are weighted and fused to calculate the location index of each branch user.

[0128] In this embodiment, the initial values ​​of daily fluctuations corresponding to daytime and nighttime periods are calculated by dynamic weighting. The influence of dimensions is eliminated by normalization. The weights of daytime and nighttime characteristics are dynamically adjusted according to the actual operating status of the transformer area, so that the location index can better reflect the user's true electrical location.

[0129] Specifically, based on the daily voltage fluctuation, the minimum and maximum daily fluctuation of each branch user during the daytime and nighttime periods are determined respectively. Based on the minimum and maximum daily fluctuation of the daytime and nighttime periods, the normalized values ​​of each branch user during the daytime and nighttime periods are calculated respectively. The min-max normalization method can be used to map the original values ​​to the [0,1] interval, thereby calculating the location index of each branch user.

[0130] The dynamic weights used in calculating the location index can be dynamically adjusted according to the season and photovoltaic penetration rate. The dynamic weights are set according to the photovoltaic penetration rate level or seasonal characteristics of the transformer area. For example, in summer or in transformer areas with high photovoltaic penetration, the weights are higher during the daytime (e.g., 0.7); in winter or in transformer areas with low photovoltaic penetration, the weights are higher at night (e.g., 0.6).

[0131] For example, user F's daytime normalized value is 0.75, and its nighttime normalized value is 0.45. Currently, it is summer, and photovoltaic penetration is high; therefore, a daytime weight is set. α The weight is 0.7, with a nighttime weight of (1- α If the value is 0.3, then the position index = 0.75 × 0.7 + 0.45 × 0.3 = 0.66.

[0132] In determining the initial daily voltage fluctuations for each user within the distribution transformer area during the day and night, the location index of each user within the transformer area is further calculated, as follows:

[0133] ;

[0134] ;

[0135] ;

[0136] Among them, based on the location index of user i User location segmentation, first-end users: Intermediate users: End users: .

[0137] in, Let be the normalized value for user i during the daytime period. Let this be the initial daily fluctuation for user i during the daytime period. This represents the maximum daily fluctuation for all users during the daytime period. This represents the minimum daily fluctuation during the daytime period. Let be the normalized value for user i during the nighttime period. Let this be the initial daily fluctuation for user i during the nighttime period. This represents the maximum daily fluctuation during the nighttime period. This represents the minimum daily fluctuation during the nighttime period.

[0138] In this embodiment, after determining the location of each branch user in each power grid branch within the distribution substation based on a preset location index threshold range and the location index, and classifying each branch user into first-end users, intermediate users, or last-end users, the method further includes:

[0139] The consistency of the daytime and nighttime location indices for users in the same branch is calculated.

[0140] Calculate the standard deviation of the location index of the same branch user over multiple days, and determine the stability of the same branch user over multiple days based on the standard deviation;

[0141] Based on the consistency of the time period and the stability of the multi-day time period, the first-end users, intermediate users, or last-end users obtained from the division are verified.

[0142] It is understandable that after the user location is divided, the corresponding division structure may be affected by abnormal or incorrect voltage data in the power grid, which may lead to deviations in the calculation results. Therefore, after the user location division results are determined, it is necessary to use appropriate verification methods to verify the division results.

[0143] Specifically, the verification mechanism for user location segmentation results assesses the reliability of the segmentation results by analyzing the consistency of user location indices at different time periods and on different dates, and marks and reviews suspicious results.

[0144] The verification scheme includes a comprehensive verification and result confirmation of time period consistency test and multi-day stability assessment. It verifies whether the electrical location of users is consistent in different time periods and verifies the stability of the user location index over multiple consecutive days. Then, the division results are corrected based on the consistency and stability assessment results.

[0145] Among them, time period consistency refers to the degree of consistency of the location index of users at different time periods, which can be expressed by correlation coefficient, cosine similarity or difference index.

[0146] Among them, multi-day stability refers to the stability of the location index during continuous observation over multiple days, which can be calculated using standard deviation, coefficient of variation, or trend stability index.

[0147] In this embodiment, consistency calculation is performed by calculating the time period consistency between the daytime location index and the nighttime location index of the same branch user. The correlation coefficient or the reciprocal of the absolute difference is used as the consistency index. When the time period consistency is lower than the threshold (e.g., 0.8), the user's location division result is marked as suspicious.

[0148] In this embodiment, stability calculation is performed by calculating the standard deviation of the location index of the same branch user over multiple days, and determining the stability of the same branch user over multiple days based on the standard deviation. When the standard deviation is greater than a threshold (e.g., 0.1), the user's location division result is marked as unstable.

[0149] In summary, based on the consistency of the time period and the stability of the multi-day time period, the first-end users, intermediate users, or last-end users obtained from the division are verified. For users who fail the verification, a manual review process is initiated or a backup algorithm is used to recalculate.

[0150] For example, suppose user G is classified as an end user, but his daytime location index is 0.82, his nighttime location index is 0.45, his time period consistency is poor (0.54), and his multi-day location index standard deviation is 0.12, which is unstable. Therefore, the user can be automatically marked as needing manual review.

[0151] Specifically, after the initial segmentation, the consistency of users' time periods is verified. With long-term stability To ensure the accuracy of user location identification in the transformer area, the following measures are taken:

[0152] ;

[0153] in, The closer the value is to 1, the more reliable the user's location identification result is.

[0154] ;

[0155] in, Let the standard deviation of the multi-day location index of user i over a long time scale be denoted as . This is the average of the location index over multiple days.

[0156] It should be noted that end-users play a significant role in distributed renewable energy planning and energy storage configuration during the identification of user locations within a distribution area. Therefore, special criteria are added for the identification of end-users within the distribution area, requiring simultaneous fulfillment of... , , If so, the user is considered a reliable end user.

[0157] This embodiment uses a voltage time series similarity measurement method based on SBD to calculate the similarity between preprocessed voltage time series data and generate an SBD distance matrix based on the similarity. Using the DBSCAN clustering method, based on the SBD distance matrix, it identifies all branch users within each power grid branch in the distribution substation area and determines the daily voltage fluctuation of each branch user. This enables the division of users within each power grid branch in the distribution substation area and the determination of the daily voltage fluctuation of each user, thereby further determining the approximate location of each user.

[0158] Furthermore, to achieve the above objectives, this embodiment also proposes a transformer substation user location identification system, the system comprising:

[0159] The data processing module is used to acquire the voltage timing data of all users within the distribution area under the same time period and the same sampling frequency, and to preprocess the voltage timing data.

[0160] The fluctuation determination module is used to determine the branch to which each user belongs in the distribution area based on the preprocessed voltage time series data, and to determine the daily voltage fluctuation of each user in each branch.

[0161] The index calculation module calculates the location index for each user based on the daily voltage fluctuation.

[0162] The location segmentation module is used to determine the location of each user in the distribution radio area based on a preset location index threshold range and the location index, and to classify each user into the first user, middle user or last user in the distribution radio area.

[0163] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0164] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0165] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0166] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for identifying the location of transformer substation users based on massive voltage data, characterized in that, The method includes the following steps: Obtain voltage timing data of all users within the distribution area under the same time period and the same sampling frequency, and preprocess the voltage timing data; Based on the preprocessed voltage time series data, all branch users in each power grid branch within the distribution substation area are identified, and the daily voltage fluctuation of each branch user is determined. Calculate the location index for each branch user based on the daily voltage fluctuation. Based on the preset location index threshold range and the location index, the location of each branch user in each power grid branch within the distribution area is determined, and each branch user is correspondingly classified as a head user, intermediate user, or end user. The step of calculating the location index of each branch user based on the daily voltage fluctuation includes: Based on the daily voltage fluctuation, the minimum and maximum daily fluctuation of each branch user during the daytime and nighttime periods are determined respectively. Based on the minimum and maximum daily fluctuations during the daytime and nighttime periods, the normalized values ​​for each branch user during the daytime and nighttime periods are calculated respectively. Dynamic weights are set based on the photovoltaic penetration rate level or seasonal characteristics of the transformer area, and the normalized values ​​of the daytime and nighttime periods are weighted and fused to calculate the location index of each branch user. After determining the location of each branch user within each power grid branch in the distribution substation area based on a preset location index threshold range and the location index, and classifying each branch user into first-end users, intermediate users, or last-end users, the method further includes: The consistency of the daytime and nighttime location indices for users in the same branch is calculated. Calculate the standard deviation of the location index of the same branch user over multiple days, and determine the stability of the same branch user over multiple days based on the standard deviation; Based on the consistency of the time period and the stability of the multi-day time period, the first-end users, intermediate users, or last-end users obtained from the division are verified.

2. The method as described in claim 1, characterized in that, The step of determining all branch users within each power grid branch in the distribution substation area based on the preprocessed voltage time-series data, and determining the daily voltage fluctuation of each branch user, includes: The voltage time series similarity measurement method based on SBD calculates the similarity between each preprocessed voltage time series data, and generates an SBD distance matrix based on the similarity. Using the DBSCAN clustering method, based on the SBD distance matrix, all branch users within each power grid branch in the distribution substation are identified, and the daily voltage fluctuation of each branch user is determined.

3. The method as described in claim 2, characterized in that, The step of determining the daily voltage fluctuation of each branch user includes: Determine the initial values ​​of daily voltage fluctuations for each branch user during the daytime and nighttime periods respectively; Historical voltage fluctuation values ​​for daytime and nighttime periods are obtained separately, and the historical voltage fluctuation values ​​and the initial daily voltage fluctuation values ​​are fused together to obtain the daily voltage fluctuation amount for each branch user.

4. The method as described in claim 3, characterized in that, The steps of determining the initial values ​​of daily voltage fluctuations for each branch user during the daytime and nighttime periods include: The initial value of daily voltage fluctuation for each branch user during the daytime period is calculated using a calculation method based on voltage range or the difference between high and low quantiles, in order to capture voltage fluctuations caused by drastic changes in photovoltaic power during the daytime period. The initial value of the daily voltage fluctuation of each branch user during the nighttime period is calculated using a calculation method based on voltage standard deviation, so as to reflect the continuous voltage deviation caused by changes in base load during the nighttime period.

5. The method as described in claim 1, characterized in that, After the step of classifying the branch users into first-end users, intermediate users, or last-end users, the method further includes: Based on the relative positions between the first-end users, intermediate users, and last-end users obtained from the division, a dynamic voltage state vector is constructed for each user. The dynamic voltage state vector includes the voltage and current correlation analysis of the user, as well as the equivalent impedance of the distribution transformer from the user to the distribution substation. Based on the dynamic voltage state vector, the dominant voltage fluctuation mode and key propagation path of the distribution substation during operation are extracted; Based on the dominant voltage fluctuation pattern and key propagation path, a directed graph of user voltage interaction is constructed, wherein the nodes in the directed graph of user voltage interaction are users, and the edge weights between nodes are the relationship strength of voltage fluctuations between users.

6. The method as described in claim 5, characterized in that, After the step of constructing a directed graph of user voltage interaction based on the dominant voltage fluctuation mode and the key propagation path, the method further includes: If a voltage disturbance source user is identified in the directed graph of user voltage interaction, its subsequent behavior is predicted based on the historical disturbance pattern of the voltage disturbance source user, and corresponding load adjustment suggestions are sent to the power grid dispatching system in advance based on the prediction results. If a voltage-vulnerable user is identified within the directed graph of user voltage interactions, the voltage stability margin index of the voltage-vulnerable user is determined. When the voltage stability margin index is lower than a threshold, adjustable resources adjacent to the voltage-vulnerable user are determined and activated according to the directed graph of user voltage interactions. The voltage stability margin index is calculated from the rate of change norm of the dynamic voltage state vector of the voltage-vulnerable user.

7. The method as described in claim 1, characterized in that, Before the step of determining all branch users within each power grid branch in the distribution substation area based on the preprocessed voltage time-series data, and determining the daily voltage fluctuation of each branch user, the method further includes: Obtain the self-voltage time series of the distribution transformers in the distribution area, sort the self-voltage time series according to the voltage value, and use the voltage value in the 20th percentile as the voltage threshold. The preprocessed voltage time series data of the branch users corresponding to the time periods below the voltage threshold are filtered.

8. A transformer substation location identification system based on the method of any one of claims 1-7, characterized in that, The system includes: The data processing module is used to acquire voltage timing data of all users in the distribution area under the same time period and the same sampling frequency, and to preprocess the voltage timing data. The fluctuation determination module is used to determine all branch users in each power grid branch within the distribution substation area based on the preprocessed voltage time series data, and to determine the daily voltage fluctuation of each branch user. The index calculation module is used to calculate the location index of each branch user based on the daily voltage fluctuation. The location division module is used to determine the location of each branch user in each power grid branch within the distribution area based on a preset location index threshold range and the location index, and to classify each branch user into a head user, intermediate user, or end user.

Images