10-kilovolt line loss segmentation monitoring method and system based on multi-element data

Abstract

This invention provides a method and system for segmented monitoring of 10 kV line losses based on multi-source data. It relates to the field of distribution network line loss monitoring technology. The method includes: acquiring multi-source operational data of a target distribution network that includes at least target measurement data and topological characteristics; verifying the reliability of the target measurement data based on a preset system reference benchmark to select a reliable measurement dataset; constructing a line segmentation model based on topological characteristics to divide the network into several monitoring segments; and calculating the dynamic line loss index of each monitoring segment based on the model and the reliable measurement dataset. This invention significantly improves the accuracy and reliability of line loss calculation results from the data source.

Description

Technical Field

[0001] This invention relates to the field of power distribution network line loss monitoring technology, specifically to a 10 kV line loss segmentation monitoring method and system based on multivariate data. Background Technology

[0002] Line loss rate is a core indicator reflecting the operational efficiency and lean management level of a distribution network. With the continuous and in-depth construction of smart grids, a large number of measurement devices and information systems have been deployed in 10 kV distribution networks, accumulating massive amounts of multi-source operational data. However, traditional distribution network line loss monitoring methods mainly perform coarse-grained statistics on the entire feeder or over a long time period, making it difficult to accurately characterize the local loss characteristics and dynamic evolution patterns within the distribution network.

[0003] Currently, existing line loss monitoring and analysis technologies have revealed the following significant shortcomings in practical applications: First, due to the independent operation of each business system, multi-source data is often isolated, and the end-side measurement data lacks a reliability verification mechanism based on system-level benchmarks when it is accessed. Distortion or deviation of underlying data often leads to inaccurate line loss calculation results. Second, the distribution network topology is complex and dynamic topology changes such as interconnection and power transfer occur frequently. Existing static monitoring methods cannot adapt to the dynamic reconstruction of network boundaries, which easily leads to errors in the calculation of line loss indicators during topology switching. Finally, after identifying high-loss anomalies, traditional investigation methods mostly rely on manual experience or simple threshold alarms, lacking in-depth mining and correlation analysis of the time-series characteristics of load consumption and line loss fluctuations, making it difficult to accurately locate potential abnormal power consumption behaviors in complex network structures.

[0004] Currently, there is a lack of a good technology that can effectively solve the above problems.

[0005] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a 10 kV line loss segmentation monitoring method and system based on multi-source data, so as to solve the technical problems in the prior art, such as coarse granularity of distribution network line loss calculation, difficulty in guaranteeing the quality of multi-source data, difficulty in adapting to dynamic topology changes, and low efficiency in abnormal user investigation.

[0007] The technical solution of the present invention is as follows:

[0008] On one hand, the present invention provides a 10 kV line loss segmentation monitoring method based on multi-source data, comprising: acquiring multi-source operational data of a target distribution network, wherein the multi-source operational data includes at least target measurement data and topological features; verifying the reliability of the target measurement data based on a preset system reference benchmark to select a reliable measurement dataset; constructing a line segmentation model of the target distribution network based on the topological features to divide it into several monitoring segments; and calculating the dynamic line loss index of each monitoring segment based on the line segmentation model and the reliable measurement dataset.

[0009] Optionally, the step of verifying the reliability of the target measurement data based on a preset system reference benchmark to select a reliable measurement dataset includes: acquiring the target measurement data and the corresponding system reference benchmark within a preset time window based on each distribution node in the target distribution network; evaluating the time series difference between the target measurement data and the system reference benchmark; and when the time series difference meets a preset reliability condition, including the corresponding target measurement data in the reliable measurement dataset.

[0010] Optionally, evaluating the time series difference between the target measurement data and the system reference benchmark includes: extracting a data deviation index between the target measurement data and the system reference benchmark under time-domain alignment; using the system reference benchmark as a reference surface, analyzing the distribution characteristics of the data deviation index, and using the distribution characteristics as the time series difference.

[0011] Optionally, the step of constructing a line segmentation model of the target distribution network based on the topology features to divide it into several monitoring segments includes: identifying node devices with measurement capabilities in the target distribution network based on the topology features; parsing the topology features with the node devices as network boundaries to divide the target distribution network into several independent topology subdomains; establishing a mapping relationship between each topology subdomain and its internal distribution nodes to generate each monitoring segment.

[0012] Optionally, the step of calculating the dynamic line loss index of each monitoring segment based on the line segmentation model and the reliable measurement dataset includes: extracting the boundary current flow characteristics and the internal power consumption characteristics of each monitoring segment at the node equipment from the reliable measurement dataset; and generating the dynamic line loss index of each monitoring segment based on the power balance mapping relationship between the boundary current flow characteristics and the internal power consumption characteristics.

[0013] Optionally, after calculating the dynamic line loss index of each of the monitoring segments, the method further includes:

[0014] Monitor the topology connection status of the interconnection equipment in the target distribution network; when a change in the topology connection status is detected, obtain the bidirectional power flow characteristics at the corresponding interconnection equipment; dynamically update the line segmentation model based on the bidirectional power flow characteristics, and synchronously adjust the dynamic line loss index based on the updated line segmentation model.

[0015] Optionally, after calculating the dynamic line loss index of each monitoring segment, the method further includes: comparing the dynamic line loss index of each monitoring segment with a preset theoretical loss benchmark; when the deviation exceeds a preset safety threshold range, marking the corresponding monitoring segment as a high-loss abnormal segment; calculating the trend correlation index between the load consumption time-series characteristics of users in the high-loss abnormal segment and the line loss fluctuation time-series characteristics of the high-loss abnormal segment; and locating suspected abnormal electricity users in the high-loss abnormal segment based on the trend correlation index.

[0016] Optionally, the step of locating suspected abnormal electricity users within the high-loss abnormal segment based on the trend correlation index includes: when the trend correlation index shows a significant negative correlation, and the corresponding user's load consumption decrease interval coincides with the line loss fluctuation increase interval in time, the user is determined to be the suspected abnormal electricity user.

[0017] On the other hand, the present invention also provides a 10 kV line loss segmentation monitoring system based on multi-source data, comprising: a data acquisition module for acquiring multi-source operational data of a target distribution network, wherein the multi-source operational data includes at least target measurement data and topological features; a quality verification module for performing reliability verification on the target measurement data based on a preset system reference benchmark to select a reliable measurement dataset; a segmentation modeling module for constructing a line segmentation model of the target distribution network based on the topological features to divide it into several monitoring segments; and a line loss calculation module for calculating the dynamic line loss index of each monitoring segment based on the line segmentation model and the reliable measurement dataset.

[0018] Optionally, the system further includes: a dynamic topology processing module: used to monitor the topology connection status of interconnecting equipment in the target distribution network; when a change in the topology connection status is detected, the module obtains the bidirectional power flow characteristics at the corresponding interconnecting equipment and dynamically updates the line segment model based on the bidirectional power flow characteristics so that the line loss calculation module can synchronously adjust the dynamic line loss index; and an anomaly diagnosis module: used to compare the dynamic line loss index of each monitored segment with the preset theoretical loss benchmark to mark high-loss anomaly segments, and locate suspected abnormal electricity users by calculating the trend correlation index between the load consumption time-series characteristics of users in the high-loss anomaly segments and the line loss fluctuation time-series characteristics of the high-loss anomaly segments.

[0019] The beneficial effects of this invention are as follows:

[0020] This invention evaluates the deviation and verifies the reliability of end-side target measurement data in the time series dimension based on a system reference benchmark. This effectively eliminates distorted and abnormal underlying data, thereby building a highly reliable data foundation for subsequent power feature extraction and significantly improving the accuracy and reliability of line loss calculation results from the data source.

[0021] Meanwhile, by using node devices with measurement capabilities as the network boundary to deeply analyze the topological structure characteristics, this invention accurately divides the complex distribution network into several independent topological subdomains and calculates their dynamic line loss indicators, realizing a dimensional leap from "extensive overall calculation" to "refined local assessment" in line loss management, thereby clearly exposing the real high-loss nodes hidden in the local network structure.

[0022] Furthermore, by real-time monitoring of the topological connection status of the interconnection equipment and extracting the bidirectional power flow characteristics during power transfer changes, this invention achieves adaptive updating of the line segmentation model and synchronous adjustment of boundary power attributes. This effectively overcomes the computational failure problem that traditional static monitoring models are prone to cause during dynamic reconfiguration of the grid structure, thereby greatly enhancing the robustness of the line loss monitoring system under complex power grid operating conditions.

[0023] Finally, by introducing trend correlation indicators, this invention deeply explores the inherent negative correlation mapping relationship between the load consumption time-series characteristics of specific users and the fluctuation time-series characteristics of high-loss abnormal segmented line losses. This enables the automated tracking of line loss anomalies in the macro network structure to the abnormal electricity consumption behavior of micro users, completely breaking the inefficient mode of relying on manual experience for investigation. As a result, it significantly improves the efficiency of managing stubborn high-loss lines and the accuracy of locating abnormal users. Attached Figure Description

[0024] Figure 1 This is a flowchart of a 10 kV line loss segmentation monitoring method based on multivariate data in an embodiment of the present invention.

[0025] Figure 2 A flowchart of a 10 kV line loss segmentation monitoring method based on multivariate data, provided as another preferred embodiment of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention; that is, the described embodiments are merely some embodiments of the invention, and not all embodiments. The components of the embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0027] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0028] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus 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 apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0029] As mentioned above, current line loss monitoring and analysis technologies have revealed the following significant shortcomings in practical applications: First, due to the independent operation of each business system, multi-source data is often isolated, and the end-side measurement data lacks a reliability verification mechanism based on system-level benchmarks when accessing the system. Distortion or deviation of underlying data often leads to inaccurate line loss calculation results. Second, the distribution network topology is complex and dynamic topology changes such as interconnection and power transfer occur frequently. Existing static monitoring methods cannot adapt to the dynamic reconstruction of network boundaries, which easily leads to errors in the calculation of line loss indicators during topology switching. Finally, after identifying high-loss anomalies, traditional investigation methods mostly rely on manual experience or simple threshold alarms, lacking in-depth mining and correlation analysis of the time-series characteristics of load consumption and line loss fluctuations, making it difficult to accurately locate potential abnormal power consumption behaviors in complex network structures.

[0030] To address this issue, the present invention provides a method and system for segmented monitoring of 10 kV line loss based on multi-source data, which solves the above problems in the following manner.

[0031] Example 1:

[0032] like Figures 1 to 2 As shown in the figure, this embodiment provides a method for segmented monitoring of 10 kV line loss based on multivariate data. The method specifically includes the following steps:

[0033] S100. Obtain multi-source operation data of the target distribution network, wherein the multi-source operation data includes at least target measurement data and topology characteristics.

[0034] To ensure the comprehensiveness and accuracy of the data base, the aforementioned multi-source operational data is mainly collected from multiple channels, including distribution automation systems (such as FTU, DTU, TTU, and other measurement terminals), electricity consumption information collection systems, dispatching SCADA systems, and PMS systems. After acquiring the heterogeneous data, the system performs data cleaning and missing value imputation. The data cleaning includes removing outliers and duplicate data, while the missing value imputation specifically employs linear interpolation or adjacent mean methods. A unified time window (e.g., 15 minutes) is used for time-series alignment to eliminate clock differences between systems. In addition, the equipment codes of different systems are mapped to a unified identifier, and preprocessing operations such as data conversion and storage are performed to establish a data warehouse, providing high-quality data support for subsequent analysis.

[0035] To synchronize the high-frequency, second-level measurement data of the distribution automation system with the low-frequency measurement data of the data acquisition system, the system employs a time-domain frequency reduction and alignment mechanism. Specifically, the average active power is calculated by definite integration of the high-frequency power curve within a unified time slice, and then the average active power is reduced and aligned to the time section of the low-frequency system, thereby eliminating spurious biases caused by clock pulse differences.

[0036] Furthermore, when processing multi-source operational data, a convolutional neural network (CNN) is introduced to extract local features of the network's spatial distribution, a long short-term memory network (LSTM) is used to process long-term dependencies in the time series, and the extracted spatiotemporal features are weighted and fused using the SDPA (ScaledDot-ProductAttention) attention mechanism, thereby deeply mining the implicit correlations between data.

[0037] S200. Based on a preset system reference benchmark, the target measurement data is subjected to reliability verification in order to select a reliable measurement dataset.

[0038] The execution logic for data reliability verification is implemented through the following steps:

[0039] S210. Based on each distribution node in the target distribution network, obtain the target measurement data and the corresponding system reference within a preset time window.

[0040] S220. Evaluate the time series difference between the target measurement data and the system reference benchmark.

[0041] The specific data extraction and feature analysis methods inherent in the evaluation process of this time series variance include:

[0042] S221. Extract the deviation index between the target measurement data and the system reference benchmark in the time domain alignment state.

[0043] S222. Using the system reference benchmark as the reference surface, analyze the distribution characteristics of the data deviation index, and use the distribution characteristics as the time series difference.

[0044] In actual operation, the following formula is used for quantitative calculation of the data deviation extraction and difference assessment mentioned above.

[0045]

[0046] in, Indicates the degree of time series difference (i.e., the percentage of data difference); This refers to the time-series data of active power of the distribution automation terminal, which serves as the target measurement data. This refers to the active power time-series data of the electricity consumption information acquisition system, which serves as the reference benchmark for the system.

[0047] S230. When the time series difference meets the preset reliability conditions, the corresponding target measurement data is included in the reliable measurement dataset.

[0048] Based on the aforementioned calculation formula, the preset reliability condition is set as a percentage of the difference. The difference is less than a preset threshold (e.g., 10%). When the difference meets this requirement, it is determined that the measurement data has not experienced severe drift or distortion, and is thus included in the trust set.

[0049] S300. Based on the topological characteristics, construct a line segmentation model of the target distribution network to divide it into several monitoring segments.

[0050] Based on the deployment locations of the underlying devices and their connection relationships within the network structure, the process of constructing the above model and dividing it into segments specifically includes:

[0051] S310. Based on the topological characteristics, identify the node devices with measurement capabilities in the target distribution network.

[0052] S320. Using the node device as the network boundary, analyze the topology features and divide the target distribution network into several independent topological subdomains.

[0053] S330. Establish the mapping relationship between each of the topological subdomains and the internal power distribution nodes under their jurisdiction, and generate each of the monitoring segments.

[0054] S400. Based on the line segmentation model and the reliable measurement dataset, calculate the dynamic line loss index of each monitoring segment.

[0055] The specific calculation steps for mapping and calculating the electrical characteristics at different levels are as follows:

[0056] S410. Extract the boundary current characteristics of each monitoring segment at the node device and the internal power consumption characteristics at the internal power distribution node from the trusted measurement dataset.

[0057] S420. Based on the power balance mapping relationship between the boundary flow power characteristics and the internal power consumption characteristics, generate the dynamic line loss index for each monitoring segment.

[0058] In the underlying logic, the aforementioned characteristics of boundary current flow, internal power consumption, and the final dynamic line loss index are rigorously calculated using the following mathematical formula model:

[0059] The boundary flow of electricity in the inflow segment:

[0060]

[0061] Outflow segment boundary current:

[0062]

[0063] Internal power consumption (segmented power sales):

[0064]

[0065] The final dynamic line loss indicator (segmented line loss rate):

[0066]

[0067] in, This represents the power operation curve of the node equipment in the inflow monitoring segment; This represents the power operation curve of the node equipment in the outflow monitoring segment; This indicates the actual load power curve of each internal power distribution node within the monitoring segment; This indicates the specified time window for line loss analysis; This represents the obtained target dynamic line loss index.

[0068] To accommodate the frequent load shifting phenomena in smart grids, a mechanism for handling network topology changes is also included after the basic line loss calculation logic described above.

[0069] S500: Monitor the topology connection status of the interconnection equipment in the target distribution network.

[0070] S510. When a change in the power transfer status of the topology connection is detected, the bidirectional power flow characteristics at the corresponding communication device are obtained. Specifically, this is achieved by collecting forward and reverse power data from the metering module at the communication switch that has bidirectional metering function.

[0071] S520. The line segmentation model is dynamically updated based on the bidirectional power flow characteristics, and the dynamic line loss index is adjusted synchronously based on the updated line segmentation model.

[0072] To achieve intelligent diagnosis of abnormal end-users, after calculating the dynamic line loss index of each monitoring segment, the following diagnostic steps are also included:

[0073] S600. The dynamic line loss index of each monitoring segment is compared with the preset theoretical loss benchmark for deviation. In this process, a theoretical line loss optimization model and an anomaly diagnosis model can be established in advance.

[0074] S610. Extract the average deviation of the monitoring segment over historical normal operating cycles. with standard deviation The lower bound of the safety threshold for this segment is dynamically generated based on the Raida criterion. When the deviation exceeds the dynamically generated safety threshold range (e.g., ±5%), the corresponding monitoring segment is marked as a high-loss abnormal segment.

[0075] S620. Calculate the trend correlation index between the load consumption time-series characteristics of users in the high-loss anomaly segment and the line loss fluctuation time-series characteristics of the high-loss anomaly segment.

[0076] In actual calculations, this trend correlation index is quantified using the Pearson correlation coefficient to accurately characterize the linkage between user behavior and sudden changes in line loss.

[0077] Meanwhile, to address the limited applicability of the Pearson coefficient under massive high-dimensional data, the system introduces stochastic matrix theory and analyzes the distribution characteristics of line loss based on spatiotemporal correlation. Specifically, anomaly indicators are constructed through the following process:

[0078] Obtain measurement data of each node within the high-loss anomaly segment on a continuous time cross-section, and construct the original measurement state matrix. ; for the original measurement state matrix Standardization is performed to obtain the standard state matrix. And the asymmetric empirical correlation matrix is ​​calculated using the following formula. :

[0079]

[0080] The distribution of its eigenvalues ​​is solved using the single-ring theorem, and the average spectral radius of the eigenvalues ​​in the complex plane is calculated. :

[0081]

[0082] in, The first of the empirical correlation matrix 1 eigenvalue, The dimension of the matrix; This represents the number of time sections.

[0083] When the power grid operation status is abnormal, the average spectral radius The decrease, its attenuation, is physically strictly equivalent to a surge in local correlation. The system will reduce the average spectral radius. As an anomaly indicator, further extract features from multi-source data to assist in multidimensional and accurate diagnosis.

[0084] S630. Based on the trend correlation index, locate suspected abnormal electricity users within the high-loss abnormal segment.

[0085] The strict criteria for determining the location process described above specifically include:

[0086] S631. When the trend correlation index is characterized as a significant negative correlation (i.e., the absolute value of the Pearson correlation coefficient is greater than 0.6), and the corresponding user's load consumption decrease interval coincides with the line loss fluctuation increase interval in time, the "K-value method" in the anti-electricity theft location method is further introduced to calculate the ratio of the difference in user electricity consumption on the inflection point day before and after to the difference in power loss, so as to accurately characterize abnormal electricity consumption behavior and thus determine the user as the suspected abnormal electricity consumption user.

[0087] S700. The system's diagnostic and calculation results are presented graphically, specifically including: dynamically displaying the line loss status of each monitoring segment on a GIS map, highlighting and warning abnormal segments that exceed a preset threshold, and pushing the list of suspected abnormal electricity users to the terminal.

[0088] Example 2:

[0089] Based on the same general inventive concept, this second embodiment provides a 10 kV line loss segmentation monitoring system based on multi-source data. This system corresponds to the method in the first embodiment and can perform corresponding monitoring and analysis tasks on the underlying network and hardware architecture. Specifically, the 10 kV line loss segmentation monitoring system based on multi-source data includes the following modules:

[0090] Data Acquisition Module: The data acquisition module is used to acquire multi-source operational data of the target distribution network. The multi-source operational data includes at least target measurement data and topological features. To achieve deep fusion and latent feature mining of multi-source heterogeneous data, the data acquisition module further includes:

[0091] The multi-source acquisition submodule is used to collect the above heterogeneous data from multiple channels such as power distribution automation system, electricity consumption information acquisition system, dispatch SCADA system and PMS system, and perform data cleaning and time sequence alignment operations.

[0092] The time-domain frequency reduction and alignment submodule is used to synchronize the clock between the high-frequency second-level measurement data of the power distribution automation system and the low-frequency measurement data of the data acquisition system. It uses a time-domain frequency reduction and alignment mechanism to calculate the average active power of the high-frequency power curve within a unified time slice, and then reduces and aligns it to the time section of the low-frequency system to eliminate the spurious deviation caused by the difference in clock pulses.

[0093] Quality verification module: The quality verification module is used to perform reliability verification on the target measurement data based on a preset system reference benchmark, so as to screen out reliable measurement datasets. Regarding the quantitative evaluation logic for data reliability, the quality verification module further includes:

[0094] The node acquisition submodule is used to acquire the target measurement data and the corresponding system reference benchmark within a preset time window based on each distribution node in the target distribution network.

[0095] The difference assessment submodule is used to evaluate the time series difference between the target measurement data and the system reference benchmark. Specifically, it extracts the data deviation index between the target measurement data and the system reference benchmark under time-domain alignment; and analyzes the distribution characteristics of the data deviation index using the system reference benchmark as a reference surface. The calculation process of the time series difference is expressed by the following formula:

[0096]

[0097] in, Indicates the degree of time series difference (i.e., the percentage of data difference); This refers to the time-series data of active power of the distribution automation terminal, which serves as the target measurement data. This refers to the active power time-series data of the electricity consumption information acquisition system, which serves as the reference benchmark for the system.

[0098] The filtering and inclusion submodule is used to include the corresponding target measurement data into the reliable measurement dataset when the time series difference meets the preset reliability conditions.

[0099] Segmentation Modeling Module: This module is used to construct a segmentation model of the target distribution network based on the topological characteristics, thereby dividing it into several monitoring segments. To achieve refined segmentation of the network structure, the segmentation modeling module further includes:

[0100] The node identification submodule is used to identify the node devices with measurement capabilities in the target distribution network based on the topological characteristics.

[0101] The subdomain parsing submodule is used to parse the topology features with the node device as the network boundary and divide the target distribution network into several independent topology subdomains.

[0102] The mapping generation submodule is used to establish the mapping relationship between each of the topological subdomains and the internal power distribution nodes under their jurisdiction, and to generate each of the monitoring segments.

[0103] Line loss calculation module: The line loss calculation module is used to calculate the dynamic line loss index of each monitoring segment based on the line segmentation model and the reliable measurement dataset. To accurately quantify the power characteristics at different levels within each segment, the line loss calculation module further includes:

[0104] The power extraction submodule is used to extract the boundary power flow characteristics of each monitoring segment at the node device and the internal power consumption characteristics at the internal power distribution node from the trusted measurement dataset.

[0105] The index generation submodule is used to generate the dynamic line loss index for each monitoring segment based on the power balance mapping relationship between the boundary current power characteristics and the internal power consumption characteristics. The calculation process of the dynamic line loss index is expressed by the following formula:

[0106]

[0107]

[0108]

[0109]

[0110] in, This indicates the boundary flow of electricity in the inflow segment; This indicates the boundary flow of electricity in the outflow segment; Indicates internal power consumption (segmented power sales). This represents the final dynamic line loss indicator (segmented line loss rate). This represents the power operation curve of the node equipment in the inflow monitoring segment; This represents the power operation curve of the node equipment in the outflow monitoring segment; This indicates the actual load power curve of each internal power distribution node within the monitoring segment; This indicates the specified time window for line loss analysis.

[0111] Dynamic topology processing module: The dynamic topology processing module is used to monitor the topology connection status of interconnection equipment in the target distribution network. To adapt to the frequent load transfer phenomena in smart grids, the dynamic topology processing module further includes:

[0112] The status monitoring submodule is used to monitor the open / closed status of the tie switch in real time through the power distribution automation system;

[0113] The power transfer feature acquisition submodule is used to acquire the bidirectional power flow features at the corresponding communication device when a power transfer change is detected in the topology connection state.

[0114] The model dynamic update submodule is used to dynamically update the line segment model based on the bidirectional power flow characteristics, so that the line loss calculation module can synchronously adjust the dynamic line loss index.

[0115] Anomaly Diagnosis Module: This module compares the dynamic line loss indicators of each monitoring segment with a preset theoretical loss benchmark to determine the deviation. To achieve multi-dimensional and accurate diagnosis and tracing of anomaly users at the end point, the anomaly diagnosis module further includes:

[0116] The deviation comparison submodule is used to compare and calculate the dynamic line loss index of each monitoring segment with the preset theoretical loss benchmark.

[0117] The high-loss marking submodule is used to extract the average deviation of the monitoring segment over its historical normal operating cycles. with standard deviation The lower bound of the safety threshold for this segment is dynamically generated based on the Raida criterion. When the deviation exceeds the dynamically generated safety threshold range (e.g., ±5%), the corresponding monitoring segment is marked as a high-loss abnormal segment.

[0118] The correlation analysis submodule is used to calculate the trend correlation index between the load consumption time-series characteristics of users within the high-loss anomaly segment and the line loss fluctuation time-series characteristics of the high-loss anomaly segment. In this calculation process, to address the limited applicability of the Pearson coefficient with massive data, stochastic matrix theory is introduced to construct the original measurement state matrix and perform standardization to obtain the standard state matrix. The asymmetric empirical correlation matrix is ​​calculated using the following formula. :

[0119]

[0120] The eigenvalue distribution is solved using the single-ring theorem and singular value equivalence matrices, and the average spectral radius on the complex plane is calculated. :

[0121]

[0122] in, The first of the empirical correlation matrix 1 eigenvalue, The dimension of the matrix; The number of time sections. The average spectral radius. The degree of reduction in offset is used as an anomaly indicator for feature extraction and correlation analysis across big data dimensions.

[0123] The anomaly location submodule is used to locate suspected abnormal electricity users within the high-loss anomaly segment based on the trend correlation index. Specifically, when the trend correlation index shows a significant negative correlation, and the corresponding user's load consumption decrease interval coincides with the line loss fluctuation increase interval in time, the "K-value method" in the anti-electricity theft location method is used for precise characterization. The determination process of the above behavior characterization is expressed by the following formula:

[0124]

[0125] in, The correlation coefficient representing abnormal electricity consumption behavior; This represents the difference in battery consumption for users on two inflection point days. This represents the difference in power loss between the high-loss abnormal segment on two consecutive inflection point days. The two consecutive inflection point days are defined by the time boundary where the load consumption decrease range coincides with the line loss fluctuation increase range. When the ratio is confirmed to be close to 1 or a preset strong correlation range, i.e., when it is confirmed that the user's power consumption change has a unique and dominant impact on line loss, the user is finally identified as the suspected abnormal power user.

[0126] Visual monitoring module: The visual monitoring module is used to graphically interact with the system's diagnostic and calculation results. Specifically, it is used to dynamically display the line loss status of each monitoring segment on a GIS map, highlight and warn abnormal segments that exceed a preset threshold, and push the list of suspected abnormal electricity users to the terminal.

[0127] It should be understood that, in the embodiments of the present invention, "B corresponding to A" means that B is associated with A, and B can be determined based on A. However, it should also be understood that determining B based on A does not mean that B is determined solely based on A; B can also be determined based on A and / or other information.

[0128] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0129] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0130] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through some interfaces, apparatuses, or units, or they may be electrical, mechanical, or other forms of connection.

[0131] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention, depending on actual needs.

[0132] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0133] From the above description of the embodiments, those skilled in the art will clearly understand that the present invention can be implemented in hardware, firmware, or a combination thereof. When implemented in software, the above-described functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of a computer program from one place to another. Storage media can be any available medium accessible to a computer. For example, but not limited to, computer-readable media can include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code having the form of instructions or data structures and accessible to a computer. Furthermore, any connection can suitably be a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of the medium. As used in this invention, disk and disc include compressed optical discs (CDs), laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray discs, wherein disks typically magnetically copy data, while discs optically copy data using lasers. The combinations described above should also be included within the scope of protection for computer-readable media.

[0134] In summary, the above description is merely a preferred embodiment of the technical solution of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for segmented monitoring of 10 kV line loss based on multivariate data, characterized in that, include: Acquire multi-source operational data of the target distribution network, wherein the multi-source operational data includes at least target measurement data and topological characteristics; Based on a preset system reference benchmark, the target measurement data is subjected to reliability verification in order to select a reliable measurement dataset. Based on the aforementioned topological characteristics, a line segmentation model of the target distribution network is constructed to divide it into several monitoring segments; Based on the line segmentation model and the reliable measurement dataset, the dynamic line loss index of each monitoring segment is calculated.

2. The 10 kV line loss segmentation monitoring method based on multivariate data according to claim 1, characterized in that, The reliability verification of the target measurement data based on a preset system reference benchmark, in order to select a reliable measurement dataset, includes: Based on each distribution node in the target distribution network, acquire the target measurement data and the corresponding system reference within a preset time window; Assess the time series difference between the target measurement data and the system reference baseline; When the time series difference meets the preset reliability conditions, the corresponding target measurement data is included in the reliable measurement dataset.

3. The 10 kV line loss segmentation monitoring method based on multivariate data according to claim 2, characterized in that, The assessment of the time series difference between the target measurement data and the system reference baseline includes: Extract the deviation index between the target measurement data and the system reference benchmark in the time domain alignment state; Using the system reference benchmark as the reference surface, the distribution characteristics of the data deviation index are analyzed, and the distribution characteristics are used as the time series difference.

4. The 10 kV line loss segmentation monitoring method based on multivariate data according to claim 1, characterized in that, Based on the aforementioned topological characteristics, a line segmentation model of the target distribution network is constructed to divide it into several monitoring segments, including: Based on the aforementioned topological characteristics, identify the node devices with measurement capabilities in the target distribution network; Using the node devices as network boundaries, the topology features are analyzed to divide the target distribution network into several independent topological subdomains. Establish a mapping relationship between each of the topological subdomains and the internal power distribution nodes under their jurisdiction, and generate each of the monitoring segments.

5. The 10 kV line loss segmentation monitoring method based on multivariate data according to claim 4, characterized in that, The calculation of dynamic line loss indicators for each monitoring segment based on the line segmentation model and the reliable measurement dataset includes: From the trusted measurement dataset, extract the boundary current characteristics of each monitoring segment at the node device and the internal power consumption characteristics at the internal power distribution node; Based on the power balance mapping relationship between the boundary flow power characteristics and the internal power consumption characteristics, the dynamic line loss index of each monitoring segment is generated.

6. The 10 kV line loss segmentation monitoring method based on multivariate data according to claim 1, characterized in that, After calculating the dynamic line loss index of each monitoring segment, the method further includes: Monitor the topology connection status of interconnection equipment in the target distribution network; When a change in the power transfer status of the topology is detected, the bidirectional power flow characteristics at the corresponding communication device are obtained; The line segmentation model is dynamically updated based on the bidirectional power flow characteristics, and the dynamic line loss index is adjusted synchronously based on the updated line segmentation model.

7. The 10 kV line loss segmentation monitoring method based on multivariate data according to claim 1, characterized in that, After calculating the dynamic line loss index of each monitoring segment, the method further includes: The dynamic line loss index of each monitoring segment is compared with the preset theoretical loss benchmark for deviation. When the deviation exceeds the preset safety threshold range, the corresponding monitoring segment is marked as a high-loss abnormal segment; Calculate the trend correlation index between the load consumption time-series characteristics of users within the high-loss anomaly segment and the line loss fluctuation time-series characteristics of the high-loss anomaly segment; Based on the trend correlation index, suspected abnormal electricity users are located within the high-loss abnormal segment.

8. The 10 kV line loss segmentation monitoring method based on multivariate data according to claim 7, characterized in that, The step of locating suspected abnormal electricity users within the high-loss abnormal segment based on the trend correlation index includes: When the trend correlation index shows a significant negative correlation, and the corresponding user's load consumption decrease range coincides with the line loss fluctuation increase range in time sequence, the user is determined to be the suspected abnormal electricity user.

9. A 10 kV line loss segmentation monitoring system based on multivariate data, characterized in that, include: Data acquisition module: used to acquire multi-source operation data of the target distribution network, wherein the multi-source operation data includes at least target measurement data and topology characteristics; Quality verification module: used to perform reliability verification on the target measurement data based on a preset system reference benchmark, so as to select reliable measurement datasets; Segmentation modeling module: used to construct a line segmentation model of the target distribution network based on the topological characteristics, so as to divide it into several monitoring segments; Line loss calculation module: used to calculate the dynamic line loss index of each monitoring segment based on the line segmentation model and the reliable measurement dataset.

10. The 10 kV line loss segmentation monitoring system based on multi-source data according to claim 9, characterized in that, The system also includes: Dynamic topology processing module: used to monitor the topology connection status of the interconnection equipment in the target distribution network. When the topology connection status is detected to have changed, the module obtains the bidirectional power flow characteristics at the corresponding interconnection equipment and dynamically updates the line segmentation model based on the bidirectional power flow characteristics so that the line loss calculation module can adjust the dynamic line loss index synchronously. Anomaly diagnosis module: It is used to compare the dynamic line loss index of each monitoring segment with the preset theoretical loss benchmark to mark the high loss anomaly segment, and to locate suspected abnormal electricity users by calculating the trend correlation index between the load consumption time sequence characteristics of users in the high loss anomaly segment and the line loss fluctuation time sequence characteristics of the high loss anomaly segment.

Images