CN-122020302-A - Method and system for detecting line loss of distribution network station area

CN122020302ACN 122020302 ACN122020302 ACN 122020302ACN-122020302-A

Abstract

The invention discloses a method and a system for detecting line loss of a distribution network station area, and relates to the technical field of line loss detection, wherein the method comprises the steps of collecting operation data of a main power supply transformer of the station area and branches of all users; the method comprises the steps of carrying out time synchronization and data cleaning on collected operation data to obtain a primary line loss result, establishing a line loss feature library, identifying and correcting the primary line loss result by utilizing a machine learning algorithm to generate a line loss detection result, carrying out classification of the operation state of a platform region, carrying out prediction on the classification result and the line loss trend, and releasing grading early warning information. The invention solves the technical problems of lack of full-flow coverage in the detection of the line loss of the distribution network station area, larger deviation of the detection result of the line loss and untimely recognition of abnormal conditions in the prior art, achieves the technical effects of realizing full-flow management of the detection of the line loss of the distribution network station area and improving the detection accuracy of the line loss and the recognition efficiency of the abnormal conditions.

Inventors

ZHAO YANMING
YAO LINLIN
WANG LIN
WANG FUMIN
ZHAO JIE

Assignees

北京致鸿信息技术有限公司

Dates

Publication Date: 20260512
Application Date: 20260129

Claims (10)

1. The method for detecting the line loss of the distribution network station area is characterized by comprising the following steps of: collecting operation data of a main power supply transformer of a transformer area and branches of each user, wherein the operation data at least comprise voltage, current and power factors; Performing time synchronization and data cleaning on the collected operation data, and calculating the difference value between the theoretical power supply energy and the actual metering energy based on the electric energy balance principle to obtain a primary line loss result; Establishing a line loss feature library by combining historical operation data, meteorological data and time period feature variables; Based on the line loss feature library, identifying and correcting a preliminary line loss result by using a machine learning algorithm to generate a line loss detection result; and classifying the running states of the areas according to the line loss detection results, predicting classification results and line loss trend, and issuing grading early warning information.
2. The method for detecting line loss of a power distribution network station according to claim 1, wherein the time synchronization and the data cleaning are performed on the collected operation data, and the difference between the theoretical power supply energy and the actual measured energy is calculated based on the electric energy balance principle, so as to obtain a preliminary line loss result, and the method comprises the following steps: The method comprises the steps of aligning collected operation data of a main power supply transformer of a transformer area and branches of each user, carrying out collection time stamp alignment, carrying out transmission simulation based on a power distribution network topological structure, and establishing a time alignment relation of each topological structure; Performing time synchronization on the collected operation data based on the time alignment relation, performing abnormal screening and missing and filling on the collected operation data, and establishing an electric energy balance supply relation between a main power supply transformer of a transformer area and a user branch based on an electric energy converging and branching relation represented by a power distribution network topological structure; And calculating an energy difference value according to the electric energy balance supply relation to obtain the primary line loss result.
3. The method for detecting line loss of a distribution network station according to claim 2, wherein the step of performing transmission simulation based on the topology of the distribution network to establish a time alignment relationship of each topology comprises: Based on a topological structure of the power distribution network, abstracting a platform area into a tree network formed by nodes and edges, wherein a root node is a main power supply transformer of the platform area, an intermediate node is branch points of all levels, and leaf nodes are user branches; performing transmission time difference analysis based on transmission paths among nodes, and determining a node transmission synchronization window; and aligning the time synchronization relationship of each acquisition node according to the node transmission synchronization window, and establishing the time alignment relationship of each topological structure.
4. The method for detecting line loss of a power distribution network station according to claim 3, wherein the step of calculating an energy difference value according to the electric energy balance supply relation to obtain the preliminary line loss result comprises the steps of: based on the topological structure of the power distribution network, configuring multi-line loss monitoring nodes, wherein key transmission path nodes of a main power supply transformer and all user branches of a transformer area are covered; according to the electric energy balance supply relation, energy difference value calculation is sequentially carried out on the multi-line loss monitoring nodes, and difference value distribution is obtained; and performing multi-node path line loss result standardization setting according to the difference distribution to obtain the multi-level primary line loss result.
5. The method for detecting line loss of a power distribution network station according to claim 2, wherein the step of establishing a line loss feature library by combining historical operation data, meteorological data and time period feature variables comprises the steps of: respectively carrying out line loss relation fitting on the historical operation data, the meteorological data and the time period characteristic variable, and establishing a line loss influence relation of each dimension; Based on the line loss influence relation of each dimension and the topological structure of the power distribution network, extracting influence variable characteristics, and establishing a mapping relation between the influence variable characteristics and the line loss quantization coefficients; and based on the mapping relation, carrying out association index according to the corresponding data time and the topology node, and establishing the line loss feature library.
6. The method for detecting line loss of a power distribution network station according to claim 5, wherein establishing the mapping relationship between the influencing variable characteristic and the line loss quantization coefficient comprises: Extracting electrical state characteristics, including voltage deviation rate, three-phase imbalance degree and total harmonic distortion rate of a platform area outlet and key nodes of each level; Extracting load behavior characteristics, including daily load rate, peak Gu Chalv and curve shape similarity with a typical power consumption mode calculated based on user branch data; extracting network topology characteristics, including a power supply radius calculated based on a topology structure, line equivalent impedance and the hierarchical depth of nodes in a tree network; extracting space-time environmental characteristics, including a temperature-humidity influence coefficient generated by converting meteorological data, a date type factor generated by converting a time period characteristic variable and a holiday mode label; and carrying out quantitative mapping on the electrical state characteristics, the load behavior characteristics, the network topology characteristics and the space-time environment characteristics and the corresponding line loss influence generation quantity, and constructing the mapping relation.
7. The method for detecting line loss in a power distribution network station according to claim 5, wherein based on the line loss feature library, the machine learning algorithm is used to identify and correct the preliminary line loss result, and the generating of the line loss detection result includes: based on the line loss feature library, learning theoretical line loss change through a machine learning algorithm, and constructing a reference line loss model; And judging the preliminary line loss result through the reference line loss model, determining whether the preliminary line loss is normal theoretical line loss, identifying and marking abnormal theoretical line loss, and generating the line loss detection result.
8. The method for detecting line loss of a distribution network station according to claim 7, further comprising: Performing deviation calculation according to the primary line loss results of each level and the corresponding reference value to obtain a deviation rate; and based on the deviation rate and a prestored association pattern in an abnormal feature library, identifying an abnormal line loss pattern and classifying, and generating the line loss detection result containing at least one abnormal line loss pattern.
9. The method for detecting line loss in a power distribution network station according to claim 8, wherein the abnormal line loss pattern includes suspected power theft loss, equipment failure loss, and abnormal data quality, the method further comprising: Tracing and positioning the identified abnormal line loss mode by combining electrical correlation analysis with a power distribution network topological structure, determining a level with the highest abnormal contribution degree in the primary line loss result, and calculating the correlation coefficient of each user branch current and the total residual current of the system or calculating the equivalent loss power of each line section in the abnormal level; and judging the object with the highest correlation coefficient or the maximum equivalent loss power as a high-probability abnormal source and outputting positioning information.
10. A power distribution network station line loss detection system, wherein the system is configured to implement a power distribution network station line loss detection method according to any one of claims 1-9, the system comprising: The operation data acquisition module is used for acquiring operation data of the main power supply transformer of the transformer area and all user branches, and at least comprises voltage, current and power factors; The primary line loss result acquisition module is used for carrying out time synchronization and data cleaning on the collected operation data, calculating the difference value between the theoretical power supply energy and the actual metering energy based on the electric energy balance principle, and obtaining a primary line loss result; the line loss feature library building module is used for building a line loss feature library by combining historical operation data, meteorological data and time period feature variables; the line loss detection result generation module is used for identifying and correcting the preliminary line loss result by utilizing a machine learning algorithm based on the line loss feature library to generate a line loss detection result; and the early warning information issuing module is used for classifying the running states of the areas according to the line loss detection results, predicting the classification results and the line loss trend and issuing grading early warning information.

Description

Method and system for detecting line loss of distribution network station area Technical Field The invention relates to the technical field of line loss detection, in particular to a method and a system for detecting line loss of a power distribution network station area. Background In the operation management of a power distribution network, the line loss of a transformer area is a core index for measuring the power supply efficiency and the management level, and the accurate detection and the abnormal management of the line loss directly affect the economic benefit and the power supply reliability of the power grid. At present, the traditional method for detecting the line loss of the transformer area is mostly dependent on manual accounting or single dimension data calculation, and has obvious technical limitations that on one hand, part of the method only collects operation data of main transformers or few nodes, the whole coverage of user branch data is not realized, a data processing link lacks a strict time synchronization and cleaning mechanism, the line loss calculation result is easily distorted due to data deviation, on the other hand, the traditional method is mostly based on fixed formulas to estimate the theoretical line loss, and is difficult to establish an accurate model by combining dynamic factors such as historical operation rules, weather changes, time period characteristics and the like, so that misjudgment or missed judgment is easily caused when the line loss is abnormally identified, abnormal response is delayed, and the problem source can not be rapidly positioned. In the prior art, the line loss detection of a distribution network station area lacks of full-flow coverage, and the technical problems of large deviation of line loss detection results and untimely abnormal condition identification are solved. Disclosure of Invention The application provides a method and a system for detecting line loss of a power distribution network station, which are used for solving the technical problems that in the prior art, the line loss detection of the power distribution network station lacks full-flow coverage, the deviation of the line loss detection result is large, and abnormal situation identification is not timely. In view of the above problems, the present application provides a method and a system for detecting line loss in a power distribution network station. The first aspect of the application provides a method for detecting line loss of a distribution network station area, which comprises the following steps: The method comprises the steps of collecting operation data of a main power supply transformer of a transformer area and branches of each user, at least comprising voltage, current and power factors, carrying out time synchronization and data cleaning on the collected operation data, calculating a difference value between theoretical power supply energy and actual metering energy based on an electric energy balance principle to obtain a primary line loss result, combining historical operation data, meteorological data and time period characteristic variables to establish a line loss characteristic library, carrying out identification and correction on the primary line loss result by using a machine learning algorithm based on the line loss characteristic library to generate a line loss detection result, carrying out classification of the operation state of the transformer area according to the line loss detection result, carrying out prediction on the classification result and line loss trend, and issuing grading early warning information. In a second aspect of the present application, there is provided a line loss detection system for a distribution network station, the system comprising: The system comprises an operation data acquisition module, a line loss feature library creation module, a line loss detection result generation module and an early warning information issuing module, wherein the operation data acquisition module is used for acquiring operation data of a main power supply transformer of a transformer area and all user branches and at least comprises voltage, current and power factors, the initial line loss result acquisition module is used for carrying out time synchronization and data cleaning on the acquired operation data, calculating the difference value between theoretical power supply energy and actual metering energy based on an electric energy balance principle to obtain an initial line loss result, the line loss feature library creation module is used for creating a line loss feature library by combining historical operation data, meteorological data and time period feature variables, the line loss detection result generation module is used for carrying out recognition and correction on the initial line loss result by utilizing a machine learning algorithm based on the line loss feature library to generate a line loss detection result, and the early