On-site electricity consumption abnormity monitoring and detecting method and system based on Beidou space-time reference

Abstract

The invention discloses an on-site electricity consumption abnormity monitoring and detecting method and system based on a Beidou space-time reference. The method comprises the steps that all sampling terminals collect electricity consumption speed detection monitoring data on an electric power transmission line; preliminarily judging a distribution transformer area with abnormal power utilization based on the power utilization speed detection monitoring data; calculating the electricity larceny suspicion degree of each distribution transformer area which is preliminarily judged to have the electricity utilization abnormity, and determining the distribution transformer area with the electricity utilization abnormity based on a calculation result; and determining the users with abnormal electricity utilization in the distribution transformer area with abnormal electricity utilization and the abnormal electricity utilization types of the users based on each set abnormal electricity utilization type criterion. The method has the outstanding advantages of uniform space-time reference, abundant detection dimensions, low abnormal false alarm rate and the like.

Description

Technical Field

[0001] The present invention belongs to the field of power consumption anomaly monitoring, and specifically relates to a field power consumption anomaly monitoring and detection method and system based on Beidou time and space reference. Background Art

[0002] Abnormal electricity usage by users is becoming increasingly serious. This not only disrupts the normal power supply and consumption order and causes significant economic losses to power companies, but also threatens the power safety of other users, seriously affecting economic development and social stability. Monitoring and detecting abnormal user electricity usage is time-consuming and complex. According to incomplete statistics, since 2020, Jiangsu Power Company has handled over 15,000 cases of abnormal electricity usage due to metering anomalies and electricity theft each year, resulting in over 100 million kWh of incorrect electricity usage.

[0003] Traditional on-site electricity inspections rely primarily on on-site personnel. As abnormal electricity usage becomes increasingly concealed (such as high-frequency interference theft and remote diversion theft), traditional technologies are unable to meet the requirements of rapid positioning (accuracy), real-time response (timeliness), and portable screening (convenience). As a result, existing automated user electricity inspection methods still have room for technical improvement:

[0004] From a data collection perspective, existing automation methods often focus on simple electrical parameters, failing to comprehensively consider multiple environmental and equipment operating factors (such as electromagnetic fields and temperature). In severe weather or complex electromagnetic environments, electrical equipment performance can be affected, leading to abnormal power usage. However, existing automation methods lack the ability to consider temperature effects, monitor electromagnetic signals, and perform correlation analysis. This results in incomplete capture of abnormal characteristics, making it impossible to detect potential risks in a timely manner and providing comprehensive assurance for the safe and stable operation of the power system.

[0005] From the perspective of anomaly detection methods, existing automated methods rely on simple thresholds or single-dimensional data to judge electricity usage anomalies. However, new electricity theft methods (such as high-frequency interference theft, remote control theft, etc.) often escape detection because they do not deeply mine the characteristics of multi-source data, resulting in increased economic losses for power companies and disrupted electricity management order.

[0006] From the perspective of analytical capabilities, the existing automation methods have the problem that rule judgment relies on manual experience and a dynamic correlation model based on the power grid topology (such as Kirchhoff's law) has not been established.

[0007] In terms of response speed, the existing automation method has a high delay in system global anomaly detection (>5 minutes), making it difficult to respond to instantaneous electricity theft in a timely manner.

[0008] In terms of inspection accuracy, existing automated methods have a false alarm rate of >15% (the missed detection rate in complex electromagnetic interference scenarios is as high as 30%).

[0009] From the perspective of service personalization, the existing automation methods have the problem of rigid inspection strategies and cannot adapt to the electricity consumption patterns of different users (such as differentiated monitoring of industry, commerce and residents).

[0010] From the perspective of on-site tools, existing automated methods have the problem that the detection equipment has a single function (only measuring electrical parameters) and it is difficult to locate complex power theft points (such as high-frequency interference sources).

[0011] Furthermore, in existing automation methods, electricity monitoring / detection equipment typically lacks a unified time and space benchmark, resulting in chaotic timestamps and ambiguous location information for data collected by different terminals. For example, in regional power grids, the monitoring equipment at substations and distributed energy access points is out of sync, making it impossible to effectively match data when analyzing power fluctuations and fault correlations. This results in slow grid fault location and significant deviations in line loss calculations, seriously interfering with stable grid operation and energy efficiency management decisions. While existing automation methods have established a unified time method, these methods suffer from the following deficiencies:

[0012] On the one hand, existing technologies have insufficient timing accuracy in the process of multi-terminal data time alignment, resulting in large time deviations between data collected by different devices, affecting the accuracy of data analysis results;

[0013] On the other hand, the existing technology has a relatively complex processing process for the integration of time information, resulting in low efficiency of data synchronization and failure to meet the needs of real-time monitoring and rapid decision-making.

[0014] In summary, the existing automated user electricity inspection method is difficult to meet the needs of on-site rapid inspection and screening, and is difficult to adapt to electricity safety and new businesses.

[0015] CN113570002A discloses a method, system, storage medium, and device for establishing an electricity theft user prediction model. The method includes: obtaining electricity usage data of multiple users and grid operation data of the substations to which the users belong, extracting all features, performing correlation analysis on all features and whether the users have committed electricity theft, and selecting correlation features with high correlation with whether the users have committed electricity theft from all features. A training table for each user is created based on the correlation features and their corresponding data values. Each training table is categorized and labeled based on the theft results of each user to obtain a training table labeled with either a theft code or a non-theft code. A training set is created based on the categorized and labeled training tables for all users, and the training set is input into an initial prediction model for training to obtain a pre-trained prediction model. However, this patent relies on historical electricity usage data to train a static model, does not consider temperature effects, and cannot perceive electromagnetic field parameters. This results in a high underdetection rate (>35%) for new types of electricity theft, such as high-frequency interference, and a significant increase in false alarm rate with fluctuations in user electricity usage (reaching 30%-40% during peak and valley periods).

[0016] CN118097213A discloses a method and device for identifying and collecting evidence of electricity theft, comprising: a detection unit for identifying electricity theft, including miswiring identification, error verification, and transformer accuracy and change detection; a collection unit for collecting data information, which serves as evidence of the theft; a processing unit for processing data, including data analysis, data storage, and image recognition; and a communication and display unit for communicating and visually displaying the data processing process and results, including uplink and downlink communication and display functions. However, this patent uses fixed rules to detect electricity theft and is unable to identify dynamic methods such as remote control theft (missed detection rate > 60%). Furthermore, the evidence data lacks a temporal and spatial reference, resulting in insufficient judicial validity. Summary of the Invention

[0017] In order to solve the shortcomings of the existing technology such as inconsistent time and space references, single detection dimension, and high abnormal false alarm rate, the present invention provides a field power consumption anomaly monitoring and detection method and system based on the Beidou time and space reference.

[0018] The present invention adopts the following technical solutions.

[0019] The present invention discloses a method for monitoring and detecting abnormal power consumption on site based on the Beidou space-time reference, comprising:

[0020] The sampling terminals are aligned in time and space through the BeiDou satellite navigation system;

[0021] Each sampling terminal collects power consumption rapid detection monitoring data on the power transmission line; the power consumption rapid detection monitoring data includes the current and voltage at the incoming cable on the primary side of the transformer, as well as the current, voltage, magnetic field strength, power load and power consumption on the user side;

[0022] Based on the power consumption rapid inspection monitoring data, it is preliminarily determined that there are distribution transformer substations with abnormal power consumption;

[0023] Calculate the suspicion of electricity theft for each distribution transformer substation that is initially judged to have abnormal electricity usage, and determine the distribution transformer substation with abnormal electricity usage based on the calculation results;

[0024] Based on the set power consumption abnormality type judgment criteria, the users with power consumption abnormality and their power consumption abnormality types in the distribution transformer substation with power consumption abnormality are determined; the power consumption abnormality type judgment criteria are set based on the power consumption rapid detection monitoring data collected on the user side.

[0025] More preferably,

[0026] The said performing time and space alignment of each sampling terminal through the Beidou satellite navigation system means achieving synchronization of the time references of the built-in clocks of different sampling terminals and unification of the spatial coordinates through Beidou satellite signals;

[0027] The terminals include devices with built-in Beidou modules and devices without built-in Beidou modules.

[0028] More preferably,

[0029] Devices with built-in Beidou modules achieve spatiotemporal alignment based on Beidou satellite time and device positioning signals; devices without built-in Beidou modules achieve spatiotemporal alignment based on Beidou satellite time and device positioning signals received by devices with built-in Beidou modules.

[0030] More preferably,

[0031] Based on the power consumption rapid inspection monitoring data, it is preliminarily determined that there are distribution transformer areas with abnormal power consumption, and the display matrix of the distribution transformer areas with abnormal power consumption is calculated based on the power consumption rapid inspection monitoring data;

[0032] When the value of the element in row k and column k in the display matrix of the abnormal power consumption distribution transformer area is non-zero, it is preliminarily judged that the kth distribution transformer area is the distribution transformer area with abnormal power consumption, where k is an integer, representing the kth distribution transformer area where the power consumption rapid detection monitoring data is collected.

[0033] More preferably,

[0034] The display matrix of the distribution transformer area with abnormal power consumption is as follows:

[0035]

[0036] Among them, ΔY is the display matrix of the distribution transformer area with abnormal power consumption; V m is the distribution transformer area voltage measurement value vector obtained based on the voltage at each user side; ΔI theft is the current vector of the equivalent current of electricity theft; I m is the phasor of the injection current measurement value of the distribution transformer substation based on the current at each user side; Y is the node admittance matrix.

[0037] More preferably,

[0038] Calculate the suspicion of electricity theft for each distribution transformer area that is initially judged to have abnormal electricity usage. Based on the calculation results, determine the distribution transformer area with abnormal electricity usage, including:

[0039] The suspicion of electricity theft is calculated in real time for each distribution transformer substation that is initially judged to have abnormal electricity consumption; the calculation formula for the suspicion of electricity theft is based on the current at the primary side incoming cable of the transformer and the current setting on each user side.

[0040] When the suspicion of electricity theft in a distribution transformer area is greater than a set first threshold, it is determined that abnormal electricity usage exists in the distribution transformer area.

[0041] More preferably,

[0042] The calculation formula for the suspicion of electricity theft is as follows:

[0043]

[0044] Where k is an integer, representing the kth distribution transformer area where the power consumption rapid inspection monitoring data is collected; K theft,t,k is the suspicion of electricity theft in the kth distribution transformer area at time t; I r,t,k is the current at the primary side incoming cable of the kth distribution transformer station at time t; I e,t,k It is the total current on the user side of the kth distribution transformer substation converted at time t.

[0045] More preferably,

[0046] The various power consumption abnormality type judgment criteria include high-frequency power theft abnormality judgment criteria, strong magnetic power theft abnormality judgment criteria, remote control power theft abnormality judgment criteria and leakage fault judgment criteria.

[0047] More preferably,

[0048] When the electricity fast detection monitoring data of a certain user side meets the high-frequency electricity theft abnormality judgment criteria, it is determined that the user has a high-frequency electricity theft abnormality;

[0049] The abnormal judgment criteria for high-frequency electricity theft include:

[0050] Determine whether a user-side current is less than a set first current threshold;

[0051] If the user-side current is less than the first current threshold, determining whether the time during which the user-side current is less than the first current threshold is greater than a set first time threshold T1;

[0052] If it is greater than the first time threshold, and the user has current data less than the first current threshold during the same period within the set first historical period, then it is determined that the first high-frequency electricity theft condition is met;

[0053] If it is greater than the first time threshold, and the average current historical data of other users in the same substation area as the user within [T-T1, T] is greater than the set first current threshold, then it is determined that the second high-frequency electricity theft condition is met; where T represents the current time;

[0054] If the user side current / voltage / power load / power consumption is within [TT L1 -0.5T2,TT L1 ] and the average value in [TT L1 -T2,TT L1 -0.5T2], and the sign of the difference between the average values ​​within [TT L1-0.5T2,TT L1 ] and the average value in [TT L1 -T2,TT L1 -0.5T2], the difference between the average values ​​has the same sign, then it is determined that the third high-frequency electricity theft condition is met; where T L1 T2 is the duration of the user-side current being less than the first current threshold; T3 is the set second time threshold;

[0055] If there is high-frequency current in the user-side current, it is determined that the fourth high-frequency electricity theft condition is met;

[0056] If the user side magnetic field strength is [TT L1 ,T] is greater than the average value in [TT L1 -T3,TT L1 ], it is determined that the fifth high-frequency electricity theft condition is met; wherein T3 is the set third time threshold;

[0057] If a user satisfies the first to fifth high-frequency electricity theft conditions simultaneously, it is determined that high-frequency electricity theft occurs at the user.

[0058] More preferably,

[0059] When the electric speed detection monitoring data of a certain user side meets the strong magnetic power theft abnormality judgment criteria, it is determined that the user has a strong magnetic power theft abnormality;

[0060] Abnormal judgment criteria for strong magnetic power theft include:

[0061] Determine whether a user-side current is less than a set second current threshold;

[0062] If the user-side current is less than the second current threshold, determining whether the time during which the user-side current is less than the second current threshold is greater than a set fourth time threshold T4;

[0063] If it is greater than the fourth time threshold, and the user has current data less than the second current threshold during the same period in the set second historical period, then it is determined that the first strong magnetic power theft condition is met;

[0064] If it is greater than the fourth time threshold, and the average current historical data of other users in the same substation area as the user within [T-T4, T] is greater than the set second current threshold, then it is determined that the second strong magnetic power theft condition is met; where T represents the current time;

[0065] If the user side current / voltage / power load / power consumption is within [TT L4 -0.5T5,TT L4 ] and the average value in [TT L4 -T5,TT L4-0.5T5], and the sign of the difference between the average values ​​within [TT L4 -0.5T5,TT L4 ] and the average value in [TT L4 -T5,TT L4 -0.5T5], the difference between the average values ​​has the same sign, then it is judged that the third strong magnetic stealing condition is met; where T L4 The duration of the user-side current being less than the second current threshold up to this moment; T5 is the set fifth time threshold;

[0066] If there is no high-frequency current in the user-side current, it is determined that the fourth strong magnetic power theft condition is met;

[0067] If the user side magnetic field strength is [TT L4 ,T] is greater than the average value in [TT L4 -T6,TT L4 ], it is determined that the fifth strong magnetic power theft condition is met; wherein T6 is the set sixth time threshold;

[0068] If a user meets the first to fifth strong magnetic power theft conditions at the same time, it is determined that strong magnetic power theft exists at the user.

[0069] More preferably,

[0070] When the electricity rapid detection monitoring data of a certain user side meets the remote control electricity theft abnormality judgment criteria, it is determined that the user has remote control electricity theft abnormality;

[0071] Remote control electricity theft abnormality judgment criteria include:

[0072] Calculate the absolute value of the difference between the average current / voltage / power load / power consumption of each user during the inspection period and the average current / voltage / power load / power consumption of other users in the same substation area during the same period, and record it as the first absolute value of each user.

[0073] If the first absolute value of a certain user side is less than [T cel ,T cs ] period and the absolute value of the difference between the average current / voltage / power load / power consumption of the user side and the average current / voltage / power load / power consumption of other users in the same substation area during the same period, it is determined that the first remote control electricity theft condition is met; wherein, T cel is the end time of the last inspection, T cs This is the start time of this inspection;

[0074] If the first absolute value on the user side is less than the set first difference threshold, it is determined that the second remote control electricity theft condition is met;

[0075] If [T cel ,T cs If the absolute value of the difference between the average current / voltage / power load / power consumption of the user side during the period and the average current / voltage / power load / power consumption of other users in the same substation area during the same period is greater than the set second difference threshold, it is determined that the third remote control electricity theft condition is met;

[0076] If a user satisfies the first, second and third remote control electricity theft conditions at the same time, it is determined that remote control electricity theft occurs at the user.

[0077] More preferably,

[0078] When the electric fast detection monitoring data of a certain user side meets the leakage fault judgment criteria, it is determined that there is a leakage fault at the user;

[0079] The criterion for leakage fault is:

[0080] If the current current of a user is greater than the maximum current on the user side in the [T-T7, T] period, determine whether there are other users in the same substation area whose current current is greater than the maximum current on the corresponding user side in the [T-T7, T] period. If not, determine that there is a leakage at the user; where T7 is the set seventh time threshold.

[0081] Another aspect of the present invention discloses a power consumption anomaly monitoring and detection system based on the power consumption anomaly monitoring and detection method, including a terminal time-space alignment module, a data acquisition module, a power consumption anomaly area preliminary judgment module, a power consumption anomaly area determination module, and a power consumption anomaly user and type determination module:

[0082] The terminal space-time alignment module performs space-time alignment on each sampling terminal through the Beidou satellite navigation system;

[0083] The data acquisition module collects power consumption rapid detection monitoring data on the power transmission line through each sampling terminal; the power consumption rapid detection monitoring data includes the current and voltage at the primary side incoming cable of the transformer, as well as the current, voltage, magnetic field strength, power load and power consumption of each user side; the power consumption abnormal area preliminary judgment module preliminarily judges the distribution transformer area with abnormal power consumption based on the power consumption rapid detection monitoring data;

[0084] The module for initially judging the abnormal power consumption area preliminarily judges the distribution transformer area with abnormal power consumption based on the power consumption rapid detection monitoring data;

[0085] The abnormal power consumption area determination module calculates the power theft suspicion of each distribution transformer area preliminarily determined to have abnormal power consumption, and determines the distribution transformer area with abnormal power consumption based on the calculation result;

[0086] The module for determining users and types of abnormal electricity consumption determines users with abnormal electricity consumption and their types of abnormal electricity consumption in the distribution transformer area with abnormal electricity consumption based on the set criteria for each type of abnormal electricity consumption; each criterion for the type of abnormal electricity consumption is set based on the rapid electricity consumption monitoring data collected from the user side.

[0087] In another aspect, the present application discloses an electronic device including a processor and a storage medium;

[0088] The storage medium is used to store instructions;

[0089] The processor is used to operate according to the instructions to execute the above-mentioned power consumption abnormality monitoring and detection method.

[0090] The present application also discloses a computer-readable storage medium having a computer program stored thereon, which implements the power consumption anomaly monitoring and detection method when the program is executed by a processor.

[0091] The beneficial effects of the present invention are as follows:

[0092] The present invention introduces a multi-dimensional correlation analysis model and combines horizontal (correlation comparison of abnormal change characteristics of data at different monitoring / detection locations on the power transmission line at the same time) and vertical (correlation comparison of abnormal change characteristics of data at the same monitoring / detection location at different times) verification to overcome the problem in the existing technology that is based on simple threshold alarms and cannot identify new types of concealed electricity theft.

[0093] The present invention constructs a node current equation based on Kirchhoff's law, establishes a mathematical model of electricity theft behavior and verifies it through real-time data, which makes up for the problem in the existing technology that rule judgment relies on manual experience and does not establish a dynamic association model based on the power grid topology.

[0094] The present invention uses BeiDou 1PPS signals to achieve nanosecond-level clock synchronization and sub-meter-level spatial positioning accuracy to achieve spatiotemporal alignment of monitoring / detection data from different terminals. Combined with edge computing to process data in real time and effectively integrate power grid data, it makes up for the problems of high system global anomaly detection delay (>5 minutes) in existing technologies, making it difficult to respond to instantaneous electricity theft and accurately locate fault points. It reduces line loss calculation errors, significantly enhances the scientific nature of power grid operation stability assessment and decision-making, and effectively guarantees reliable power supply.

[0095] The present invention integrates and verifies multi-dimensional data, and jointly determines the type of power consumption anomaly through multiple indicators such as electromagnetic field strength, current, and voltage, thereby making up for the problems in the existing technology of false alarm rate >15% and missed detection rate of 30% in complex electromagnetic interference scenarios.

[0096] The present invention supports dynamic adjustment of monitoring parameters (collection frequency, alarm threshold), uses edge computing and lightweight AI inference engine to achieve millisecond-level abnormal decision-making, accurately distinguishes normal load fluctuations from electricity theft, establishes personalized baseline thresholds based on user historical data, and supports second-level locking of stolen power sources (positioning accuracy of 0.5 meters) and blockchain judicial evidence collection; it makes up for the problem that the inspection strategy in the existing technology is rigid and cannot adapt to the electricity usage patterns of different users.

[0097] This invention develops an integrated / combined detection terminal that integrates electromagnetic and high-frequency current sensors, enabling real-time electromagnetic signal capture and analysis. This addresses the limitations of existing detection equipment, which typically lacks single-function functionality (measuring only electrical parameters) and struggles to locate complex electricity theft sites (such as high-frequency interference sources). By integrating Beidou timing, this device enables simultaneous data acquisition across the entire circuit, addressing the limitations of existing technologies that rely on remote data collection from single-source meters and lack multimodal environmental data, resulting in incomplete capture of anomaly features. This invention collects 15+ dimensions of data in real time, comprehensively analyzes multidimensional data correlations, and deeply mines anomaly features. Combined with a dynamic baseline algorithm, this device automatically generates a personalized user electricity usage model (with a peak threshold relaxed to 1.5σ). It also simultaneously checks for harmonic distortion and electromagnetic field anomalies (THD > 15% or sudden increases in magnetic field strength > 50mT), effectively reducing false alarms and missed alarms to as low as 0.7% and nearly eliminating missed alarms. This significantly improves the efficiency and accuracy of on-site diagnosis, increases the success rate of detecting new electricity theft methods, comprehensively safeguards power system security, enhances the level of intelligent electricity management, and regulates the power market.

[0098] This invention uses edge AI models to achieve local millisecond-level response (latency <500ms), integrates Beidou precision spatiotemporal tags (time ±20nS, position ±0.5m), and solidifies a complete chain of evidence through blockchain (with a 100% judicial acceptance rate). For remote-controlled electricity theft, real-time spectrum analysis (400-470MHz) combined with current sag and electromagnetic surge signatures quickly identifies the target, reducing on-site evidence verification time from 2 hours to 8 seconds.

[0099] The technical solution of the present invention integrates terminal hardware optimization and algorithm innovation, achieving a comprehensive electricity theft detection rate of 96% and a response efficiency improvement of 180 times in pilot power grids in Jiangsu, Zhejiang and other provinces, saving over 100 million yuan in economic losses annually. It effectively solves the technical blind spots and judicial evidence chain defects of traditional monitoring methods in complex scenarios, and provides reliable support for power system safety and energy efficiency management.

[0100] To clarify parameter matching relationships and matching functions, this paper analyzes the matching relationship between metering circuits and metering devices, as well as the functions of existing monitoring systems. By comparing and analyzing the advantages and disadvantages of different monitoring systems, this paper proposes targeted improvement solutions, optimizes monitoring data collection methods and parameter sharing mechanisms, reduces redundant monitoring points, successfully avoids duplicate monitoring, and effectively improves monitoring efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0101] Figure 1 It is a schematic flow diagram of the present invention;

[0102] Figure 2 2. It is a flowchart of the method for monitoring and detecting abnormal power consumption on site based on the Beidou time and space reference obtained by analysis of the present invention;

[0103] Figure 3 It is a schematic diagram of the relationship between metering circuit equipment and power consumption anomalies;

[0104] Figure 4 This is a schematic diagram of a device with a built-in Beidou module receiving timing information sent by Beidou satellites in real time;

[0105] Figure 5 Schematic diagram of the pretreatment process of the present invention. DETAILED DESCRIPTION

[0106] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without making creative efforts are all within the scope of protection of the present invention.

[0107] This application discloses a method for monitoring and detecting on-site power consumption anomalies based on the Beidou time and space reference, comprising:

[0108] The sampling terminals are aligned in time and space through the BeiDou satellite navigation system;

[0109] The said performing time and space alignment of each sampling terminal through the Beidou satellite navigation system means achieving synchronization of the time references of the built-in clocks of different sampling terminals and unification of the spatial coordinates through Beidou satellite signals;

[0110] The terminals include devices with built-in Beidou modules and devices without built-in Beidou modules.

[0111] Equipment with built-in Beidou modules includes primary-side monitoring devices, on-site power inspection equipment, and concealed engineering cable detectors; equipment without built-in Beidou modules is a field inspection slave device;

[0112] Devices with built-in Beidou modules achieve spatiotemporal alignment based on Beidou satellite time and device positioning signals; devices without built-in Beidou modules achieve spatiotemporal alignment based on Beidou satellite time and device positioning signals received by devices with built-in Beidou modules.

[0113] Each sampling terminal collects power consumption rapid detection monitoring data on the power transmission line; the power consumption rapid detection monitoring data includes the current and voltage at the incoming cable on the primary side of the transformer, as well as the current, voltage, magnetic field strength, power load and power consumption on the user side;

[0114] Based on the power consumption rapid inspection monitoring data, it is preliminarily determined that there are distribution transformer substations with abnormal power consumption;

[0115] Based on the power consumption rapid inspection monitoring data, it is preliminarily determined that there are distribution transformer areas with abnormal power consumption, and the display matrix of the distribution transformer areas with abnormal power consumption is calculated based on the power consumption rapid inspection monitoring data;

[0116] The display matrix of the distribution transformer area with abnormal power consumption is as follows:

[0117]

[0118] Among them, ΔY is the display matrix of the distribution transformer area with abnormal power consumption; V m is the distribution transformer area voltage measurement value vector obtained based on the voltage at each user side; ΔI theft is the current vector of the equivalent current of electricity theft; I m is the phasor of the current injection measurement value of the distribution transformer area based on the current at each user side; Y is the node admittance matrix with the distribution transformer area as the node.

[0119] When the value of the element in row k and column k in the display matrix of the abnormal power consumption distribution transformer area is non-zero, it is preliminarily judged that the kth distribution transformer area is the distribution transformer area with abnormal power consumption, where k is an integer, representing the kth distribution transformer area where the power consumption rapid detection monitoring data is collected.

[0120] Calculate the suspicion of electricity theft for each distribution transformer substation that is initially judged to have abnormal electricity usage, and determine the distribution transformer substation with abnormal electricity usage based on the calculation results;

[0121] Calculate the suspicion of electricity theft for each distribution transformer area that is initially judged to have abnormal electricity usage. Based on the calculation results, determine the distribution transformer area with abnormal electricity usage, including:

[0122] The suspicion of electricity theft is calculated in real time for each distribution transformer substation that is initially judged to have abnormal electricity consumption. The calculation formula for the suspicion of electricity theft is based on the current at the primary side of the transformer incoming cable and the current at each user side. The calculation formula for the suspicion of electricity theft is as follows:

[0123]

[0124] Where k is an integer, representing the kth distribution transformer area where the power consumption rapid inspection monitoring data is collected; K theft,t,k is the suspicion of electricity theft in the kth distribution transformer area at time t; I r,t,k is the current at the primary side incoming cable of the kth distribution transformer station at time t; I e,t,k It is the total current on the user side of the kth distribution transformer substation converted at time t.

[0125] When the suspicion of electricity theft in a distribution transformer area is greater than a set first threshold, it is determined that abnormal electricity usage exists in the distribution transformer area; preferably, the first threshold is set to 3% to 5%.

[0126] Based on the set power consumption abnormality type judgment criteria, the users with abnormal power consumption in the distribution transformer area with abnormal power consumption and the power consumption abnormality type of the user are determined; the power consumption abnormality type judgment criteria are set based on the power consumption rapid detection monitoring data collected on the user side.

[0127] The various power consumption abnormality type judgment criteria include high-frequency power theft abnormality judgment criteria, strong magnetic power theft abnormality judgment criteria, remote control power theft abnormality judgment criteria and leakage fault judgment criteria.

[0128] When the electricity fast detection monitoring data of a certain user side meets the high-frequency electricity theft abnormality judgment criteria, it is determined that the user has a high-frequency electricity theft abnormality;

[0129] The abnormal judgment criteria for high-frequency electricity theft include:

[0130] Determine whether a user-side current (current collected by the smart meter) is less than a set first current threshold; preferably, the first current threshold is set to 2-3A;

[0131] If the user-side current is less than the first current threshold, determine whether the time during which the user-side current is less than the first current threshold is greater than a set first time threshold T1; the first time threshold is set to 5-7 minutes;

[0132] If the current value is greater than the first time threshold, and the user has current data less than the first current threshold during the same period within the set first historical period, then it is determined that the first high-frequency electricity theft condition is met; preferably, the set first historical period is one week to one month; the same period refers to the same time period every day;

[0133] If it is greater than the first time threshold, and the average current historical data of other users in the same substation area as the user within [T-T1, T] is greater than the set first current threshold, then it is determined that the second high-frequency electricity theft condition is met; where T represents the current time;

[0134] If the user side current / voltage / power load / power consumption is within [TT L1 -0.5T2,TT L1 ] and the average value in [TT L1 -T2,TT L1 -0.5T2], and the sign of the difference between the average values ​​within [TT L1 -0.5T2,TT L1 ] and the average value in [TT L1 -T2,TT L1 -0.5T2], the difference between the average values ​​has the same sign, then it is determined that the third high-frequency electricity theft condition is met; where T L1 The time during which the user-side current is less than the first current threshold up to this moment; T2 is the set second time threshold; the second time threshold is shown in the following formula:

[0135]

[0136] Among them, T L2 The current from the user side to TT L1 The time that the current is greater than the first current threshold until the moment T D1 is the first time upper limit threshold set; T D2 The time required to collect two data points; the first time upper limit threshold T D1 The preferred value range is 3-6 hours;

[0137] If there is a high-frequency current (high-frequency current collected by a high-frequency current sensor) in the user-side current, and the high-frequency current refers to a current with a frequency of 400 MHz to 470 MHz, then it is determined that the fourth high-frequency electricity theft condition is met;

[0138] If the user side magnetic field strength is [TT L1 ,T] is greater than the average value in [TT L1 -T3,TT L1 ] is an average value, then it is determined that the fifth high-frequency electricity theft condition is met; wherein T3 is a set third time threshold; preferably, the third time threshold is set to 5-10 minutes;

[0139] If a user satisfies the first to fifth high-frequency electricity theft conditions simultaneously, it is determined that high-frequency electricity theft occurs at the user.

[0140] When the electric speed detection monitoring data of a certain user side meets the strong magnetic power theft abnormality judgment criteria, it is determined that the user has a strong magnetic power theft abnormality;

[0141] Abnormal judgment criteria for strong magnetic power theft include:

[0142] Determine whether a user-side current is less than a set second current threshold; preferably, the second current threshold is set to 2-3A;

[0143] If the user-side current is less than the second current threshold, determine whether the time during which the user-side current is less than the second current threshold is greater than a set fourth time threshold T4; preferably, the fourth time threshold is set to 5-7 minutes;

[0144] If the current value is greater than the fourth time threshold, and the user has current data less than the second current threshold during the same period within the set second historical period, then it is determined that the first strong magnetic power theft condition is met; preferably, the set second historical period is one week to one month; the same period refers to the same time period every day;

[0145] If it is greater than the fourth time threshold, and the average current historical data of other users in the same substation area as the user within [T-T4, T] is greater than the set second current threshold, then it is determined that the second strong magnetic power theft condition is met; where T represents the current time;

[0146] If the user side current / voltage / power load / power consumption is within [TT L4 -0.5T5,TT L4 ] and the average value in [TT L4 -T5,TT L4 -0.5T5], and the sign of the difference between the average values ​​within [TT L4 -0.5T5,TT L4 ] and the average value in [TT L4 -T5,TT L4 -0.5T5], the difference between the average values ​​has the same sign, then it is judged that the third strong magnetic stealing condition is met; where T L4 is the duration of time during which the user-side current is less than the second current threshold up to this moment; T5 is the set fifth time threshold; the fifth time threshold is expressed as follows:

[0147]

[0148] Among them, T L5 The current from the user side to TT L4 The time that the current is greater than the second current threshold until the moment T D3 is the second time upper limit threshold set; T D2 The time required to collect two data points; the second time upper limit threshold T D3 The preferred value range is 3-6 hours;

[0149] If there is no high-frequency current in the user-side current, it is determined that the fourth strong magnetic power theft condition is met;

[0150] If the user side magnetic field strength is [TT L4 ,T] is greater than the average value in [TT L4 -T6,TT L4 ] is an average value, then it is determined that the fifth strong magnetic power theft condition is met; wherein T6 is a set sixth time threshold; preferably, the sixth time threshold is set to 5-10 minutes;

[0151] If a user meets the first to fifth strong magnetic power theft conditions at the same time, it is determined that strong magnetic power theft exists at the user.

[0152] When the electricity rapid detection monitoring data of a certain user side meets the remote control electricity theft abnormality judgment criteria, it is determined that the user has remote control electricity theft abnormality;

[0153] Remote control electricity theft abnormality judgment criteria include:

[0154] Calculate the absolute value of the difference between the average current / voltage / power load / power consumption of each user during the inspection period and the average current / voltage / power load / power consumption of other users in the same substation area during the same period, and record it as the first absolute value of each user.

[0155] If the first absolute value of a certain user side is less than [T cel ,T cs ] period and the absolute value of the difference between the average current / voltage / power load / power consumption of the user side and the average current / voltage / power load / power consumption of other users in the same substation area during the same period, it is determined that the first remote control electricity theft condition is met; wherein, T cel is the end time of the last inspection, T cs This is the start time of this inspection;

[0156] If the first absolute value on the user side is less than the set first difference threshold, it is determined that the second remote control electricity theft condition is met; preferably, the first difference threshold is set to 2% to 5% of the absolute value of the difference between the average current / voltage / power load / power consumption of other users on the same substation area during the inspection period;

[0157] If [T cel ,T cs ] period and the absolute value of the difference between the average current / voltage / power load / power consumption of the user side and the average current / voltage / power load / power consumption of other users in the same substation area during the same period is greater than the set second difference threshold, then it is determined that the third remote control electricity theft condition is met; preferably, the second difference threshold is set to [T cel ,T cs] 2% to 5% of the absolute value of the difference between the average current / voltage / power load / power consumption of other users in the same substation area during the same period;

[0158] If a user satisfies the first, second and third remote control electricity theft conditions at the same time, it is determined that remote control electricity theft occurs at the user.

[0159] When the electric fast detection monitoring data of a certain user side meets the leakage fault judgment criteria, it is determined that there is a leakage fault at the user;

[0160] The criterion for leakage fault is:

[0161] If a user's current is greater than the maximum current on the user's side during the [T-T7, T] period, determine whether there are other users in the same substation area whose current is greater than the maximum current on the corresponding user's side during the [T-T7, T] period. If not, determine that the user has a power leakage. T7 is the seventh time threshold. Preferably, T7 is 1 to 2 days.

[0162] Example 1

[0163] A method for monitoring and detecting abnormal power consumption on site based on BeiDou time and space reference, see Appendix Figure 1 ,include:

[0164] Step 1: See attached Figure 2 , using the high-precision positioning and timing functions of the Beidou satellite navigation system, a unified space-time benchmark is built to align the space-time of data monitored / detected by different collection terminals.

[0165] Field devices are divided into those with built-in Beidou modules (such as primary-side monitoring devices, on-site power inspection equipment, and concealed engineering cable detectors) and those without built-in Beidou modules (field inspection slave devices). The data synchronization technology used for these two types of devices differs. Devices with built-in modules directly receive signals from Beidou satellites to obtain a high-precision time and space reference; devices without built-in modules obtain their time and space reference indirectly through external timing servers or other devices.

[0166] Based on BeiDou's 1PPS signal (accuracy up to 20nS), the system provides precise clock synchronization for each terminal, while utilizing BeiDou's sub-meter positioning technology to determine device location. For devices with built-in BeiDou modules, such as primary-side monitoring devices, direct satellite signal reception achieves spatiotemporal alignment. For devices without built-in modules, spatiotemporal alignment is achieved based on BeiDou satellite time and device positioning signals received by devices with built-in BeiDou modules. For example, some on-site inspection slave devices transmit data to devices with built-in modules via the IoT network. Interpolation and filtering algorithms are used to add spatiotemporal tags to ensure data spatiotemporal alignment.

[0167] Specifically, for field monitoring equipment without a built-in Beidou module, in the absence of direct satellite signal reception, spatial calibration can be performed with the help of nearby devices. Data is transmitted to devices that have been connected to the Beidou module through the Internet of Things network, and the location information and timestamp data of the known location device are used to perform data interpolation and spatial positioning calibration. During the interpolation process, the spatial position of the device is estimated by an interpolation algorithm (for example, a weighted average based on multi-device data). When the data is transmitted to a built-in module device, the time and space tags are supplemented by interpolation and filtering algorithms. The nearby device may refer to a device with a built-in Beidou module that sends data to a device without a built-in Beidou module, or a device with a built-in Beidou module that receives data from a device without a built-in Beidou module.

[0168] Those skilled in the art should know that: Devices without built-in modules are generally carried by personnel inspecting power anomalies;

[0169] Assume that device A, which does not have a built-in BeiDou module, sends a data point at time t1, and that the location of this data point is received by another device B at time t2 (assuming that device B has a built-in BeiDou module and can provide accurate time and space coordinates). Using an interpolation algorithm (such as linear interpolation or spline interpolation), the exact location of device A at time t1 is estimated based on the known time and space positions.

[0170] For example, the position of device B at time t2 is P B =(x B y B z B ), the timestamp of device B is t B , the position of the data of device A at time t1 is unknown. Use the linear interpolation formula:

[0171]

[0172] Among them, x0 is the known position of device A at the starting time, and x A The interpolated spatial position of device A at time t1 is . The same method can be used to calculate the y and z coordinates, and a similar interpolation method is used to correct for the time tags to ensure that all data points are synchronized in time and space.

[0173] It achieves local millisecond-level response (delay < 500ms) through the edge AI model, integrates Beidou precise space-time tags (time ± 20nS, position ± 0.5m), and solidifies the complete chain of evidence through blockchain (judicial acceptance rate 100%).

[0174] Step 2: See attached Figure 2, each terminal collects power consumption rapid detection monitoring data on the power transmission line and assigns time and space information to each data, and then preliminarily determines the power consumption abnormal nodes based on the power consumption rapid detection monitoring data;

[0175] The node refers to the distribution transformer area;

[0176] 2.1 Start collecting data on power transmission lines and assigning spatiotemporal information to the data;

[0177] Each data collection terminal collects data; it includes a primary-side monitoring device, on-site power inspection equipment, a concealed engineering cable detector, and a field inspection slave device. The data collected by the primary-side monitoring device, on-site power inspection equipment, and concealed engineering cable detector is monitoring data, while the data collected by the field inspection slave device is detection data.

[0178] The primary-side monitoring device, on-site power inspection equipment, and concealed engineering cable detector are all devices with built-in Beidou modules; the on-site inspection slave device is a device without a built-in Beidou module;

[0179] The primary side monitoring device is installed at the primary side incoming cable of the transformer, including a current transformer and a voltage sensor, and is used to collect the A, B, and C three-phase current and voltage data at the primary side incoming cable in the power consumption rapid detection monitoring data;

[0180] On-site electricity inspection equipment includes current transformers, smart meters, high-frequency current sensors, and magnetic field sensors, which collect current data, power load, voltage data, power consumption, magnetic field strength, etc. on the user side as described in the electricity rapid inspection monitoring data.

[0181] The current data includes the user-side current collected by the current transformer / smart meter and the high-frequency current (400MHz to 470MHz) collected by the high-frequency current sensor.

[0182] The concealed engineering cable detector is used to detect the location of the underground concealed cable at the primary side of the incoming line;

[0183] The on-site inspection slave is used to collect the power consumption speed inspection monitoring data to check the magnetic field strength on site. The magnetic field strength on site can be obtained by using a magnetic field sensor.

[0184] When the distance between the on-site inspection slave device and the on-site power inspection equipment is less than 1 meter, the magnetic field strength data collected by the on-site inspection slave device is used, while the magnetic field sensor in the on-site power inspection equipment is used at other times to avoid the situation where the magnetic field sensor in the on-site power inspection equipment is damaged and cannot be obtained.

[0185] This invention supports dynamic adjustment of monitoring parameters (collection frequency, alarm threshold), establishing a personalized baseline based on the user's historical data, and remedying the problem of rigid inspection strategies in existing technologies that cannot adapt to different user electricity usage patterns. Preferably, the collection frequency is adjustable from 1 to 15 minutes, and the alarm threshold is set at 50% above normal.

[0186] After data collection, the collected monitoring and detection data is assigned spatiotemporal information, achieving spatiotemporal alignment of monitoring / detection data from different terminals. The spatiotemporal information includes the timestamp and time of data collection, as well as the spatial location of the collection device.

[0187] In particular, the impact of temperature on loop data collection is analyzed and processed. Although the impact of temperature changes on loop data is relatively small, to further eliminate errors, a wide-temperature sensor can be used, combined with an algorithm to compensate and correct the impact of ambient temperature changes on loop data to ensure data accuracy.

[0188] See attached Figure 4 , devices (hosts) with built-in Beidou modules can receive positioning signals and timing information sent by Beidou satellites in real time. The high-precision positioning capability of the Beidou system can ensure the accuracy of the spatial position of the monitoring equipment, while the timing information provides a timestamp accurate to milliseconds or even microseconds. Based on this position and time information, the system can perform spatiotemporal labeling on the monitored data. Each piece of monitoring data will correspond to a specific location coordinate and a precise time point. In this way, it can be ensured that the data from different monitoring points are aligned in time and space. By further analyzing and processing the spatiotemporally aligned data, we can more accurately understand the on-site conditions, detect abnormalities in a timely manner, and provide strong data support for subsequent decision-making.

[0189] Based on the above-mentioned time-space aligned data, the correlation between primary-side monitoring and metering is analyzed, and combined with the characteristics of existing on-site electricity inspection tools, a power transmission line monitoring method is proposed.

[0190] 2.2 Preliminary judgment of abnormal power consumption nodes based on power consumption rapid inspection monitoring data;

[0191] One node corresponds to one distribution transformer. For example, the load of a node is the sum of all user loads in a distribution transformer area.

[0192] First, based on the on-site survey data and basic electrical engineering theory, an in-depth analysis of the metering circuit equipment in different metering system environments was conducted. Through comparative analysis, it was found that these devices have certain internal consistency and correlation in structure. Figure 3This paper also deeply analyzes the relationship between metering circuit equipment and power usage anomalies, providing an important basis for subsequent anomaly diagnosis. Building on this analysis of metering circuit equipment, this paper further analyzes the multi-dimensional data coupling relationship underlying power usage anomalies. By processing and analyzing large amounts of data, this paper reveals the interactions and influences between different data sources.

[0193] According to Kirchhoff's law, the node current equation between the voltage and the injected current of each node in the power network is as follows:

[0194] (I m +ΔI theft )=YV m (1-1)

[0195] Where: I m =[I 1m I 2m …I Nm ] T ;I m I is the measured value phasor of the node injection current, and I is the sum of the phasors of the node injection current; 1m I is the sum of the phasor measurements of the injected current (the current collected by the smart meter / current transformer) at node 1 (the first distribution transformer substation on the power transmission link where the rapid power consumption monitoring data is collected). This is the phasor sum obtained by directly measuring all injected currents at this node (i.e., the current data at the power consumption site) using on-site power consumption inspection equipment (such as current transformers and smart meters). 2m is the total phase measurement value of the injected current at node 2; I Nm is the total phase measurement value of the injected current at node N; T is the matrix transpose symbol, which means converting the row vector into a column vector; ΔI theft =[0…ΔI k …0] T is the equivalent current vector of electricity theft, which is determined according to the electricity theft point and the specific current value; Y is the node admittance matrix with the distribution transformer area as the node; V m =[V 1m V 2m …V Nm ] T Among them, V m is the node voltage measurement value vector; V 1m is the voltage phasor measurement value of the first node, that is, all the injected voltages of the node (i.e., the voltage data of the power consumption site) are directly measured by the on-site power consumption inspection equipment (such as smart meter), and the phasor sum is obtained by adding them together; V 2m is the voltage phase measurement value of the second node; V Nm is the voltage and phase measurement value of the Nth node; N is the total number of nodes in the distribution network.

[0196] Those skilled in the art should know that the node admittance matrix is ​​a matrix used to describe the electrical relationship between each node in the power network, which can be obtained based on the power system background modeling; specifically, the admittance matrix can be obtained based on line impedance, transformer impedance, load, etc., and the power system background can also obtain a common load model; in addition, since the admittance matrix in the present invention corresponds to multiple substations, that is, the load of each node is the sum of the loads of all users in a substation, even if individual users increase or decrease their load, the total load of all users in the substation is very small (generally less than 0.5%) and can be basically ignored, without affecting the overall admittance matrix of the area. In summary, the admittance matrix of an area is basically fixed.

[0197] The electricity theft behavior is reflected in the distribution network as adding a ground load at the electricity theft node k. The amount of electricity stolen is equal to the active and reactive power consumed by the load. Therefore,

[0198] ΔI theft =ΔYV m (1-2)

[0199]

[0200] Where ΔY is the value of ΔY only at the electricity theft node k, i.e., a diagonal matrix with non-zero elements, ΔY(k, k). The element values ​​are derived from the equivalent load of the theft current. As can be seen from the above formula, the value of ΔY is related to the amount of theft. The abnormal power consumption of user I varies in different electricity theft scenarios, resulting in different load impedance parameters when the power consumption is abnormal.

[0201] From the above formula, we can get

[0202] I m =Y equal V m =(Y-ΔY)V m (1-4)

[0203] Y equal Contains the abnormal power consumption behavior information of all nodes. For normal power consumption nodes, the diagonal element value of the corresponding position in ΔY is 0, and Y equal The corresponding elements in are equal to the elements in Y, and the abnormal power consumption nodes correspond to Y equal The elements in will change differently.

[0204] In a preferred embodiment of the present invention, formula (1-3) or formula (1-4) can be used to first perform a preliminary power consumption anomaly judgment to obtain the first power consumption anomaly node, and then perform power consumption anomaly judgment of steps 3 to 6 on the obtained first power consumption anomaly node, thereby increasing the accuracy of the power consumption anomaly judgment result while reducing the amount of data to be judged.

[0205] To this end, it is necessary to conduct real-time monitoring and detection of all-phase voltage and current on the primary and secondary sides. In order to reduce the impact of monitoring errors, it is also necessary to collect temperature data of the environment.

[0206] Specifically, electricity load is measured using an energy meter. When the predicted actual load differs from the meter's reading (the user's load as collected by the smart meter) (e.g., if the meter reading exceeds 5% of the predicted actual load), the metering system is considered abnormal. A power inspection is needed to identify the cause. This analysis and diagnosis requires monitoring / detection data from all aspects of the power transmission lines within the metering system. This diagnostic data is collected and recorded in real time or periodically by power inspection monitoring / detection equipment. This data includes electrical parameters such as user load, power consumption, voltage, current, power factor, and electromagnetic field data. This data not only directly quantifies user electricity usage but also contains important information such as grid operating status, load characteristics, and energy efficiency. Advanced high-precision monitoring technology enables comprehensive and accurate monitoring of the metering system, providing robust data support for power inspections. The predicted actual load can be determined based on historical electricity consumption. Specifically, this can be achieved using an expert system that uses historical electricity consumption as input to predict real-time electricity consumption data.

[0207] Step 3: See attached Figure 2 , power transmission line data cleaning

[0208] Data preprocessing before data analysis is a key step in ensuring data accuracy and validity.

[0209] After assigning spatiotemporal information to the collected data, it undergoes data preprocessing. Any data collection failures or anomalies are promptly removed to ensure only valid data is used for analysis. This assessment method allows for a quality review of the collected data. This data screening approach not only comprehensively assesses the quality of electricity metering and monitoring data but also effectively identifies and removes interfering data, providing an accurate and reliable data foundation for subsequent spatiotemporal alignment and deviation analysis, ensuring the validity and accuracy of the analysis results.

[0210] See attached Figure 5 , the data preprocessing includes:

[0211] 3.1. After starting the process, cleaning the original electricity inspection and monitoring data is the first step of preprocessing. The purpose is to remove abnormal values, missing values, etc. in the original data to ensure the integrity and consistency of the data. The cleaning includes:

[0212] 3.1.1 Determine whether there are any abnormal values, that is, check whether there are any abnormal values ​​in the data (for example, in the continuous collection of data over a period of time, the A phase current value fluctuates greatly at a certain collection point, but the data continuity of the collection points before and after it is strong). If the judgment result is yes, go to step 3.1.2; if the judgment result is no, go to step 3.1.3.

[0213] 3.1.2. Use statistical analysis methods to identify and eliminate outliers. Specifically, if outliers exist, they should be identified and eliminated through statistical analysis methods.

[0214] 3.1.3. Determine whether there are missing values ​​in the inspection data. If the judgment result is yes, proceed to step 3.1.4; if the judgment result is no, proceed to step 3.2.

[0215] 3.1.4. Fill in missing values;

[0216] Those skilled in the art should know that missing values ​​can be filled by using methods such as predictive filling based on machine learning algorithms and existing missing value filling algorithms. Specifically, if there are missing values, those skilled in the art can use appropriate methods to fill them according to the characteristics of the data and the previous and subsequent correlations.

[0217] 3.2. Determine whether there are duplicate records in the inspection data. If the judgment result is yes, remove the duplicate records; if the judgment result is no, continue to step 3.3.

[0218] 3.3. Standardize timestamps: Ensure that all data have consistent time tags to facilitate subsequent data comparisons.

[0219] 3.4. Determine whether the spatial dimension data is accurate. Specifically, check the accuracy of the spatial dimension data (such as geographic location information, device number, etc.) based on the Beidou system. If the judgment result is yes, verify the spatial dimension data; if the judgment result is no, proceed to 3.5. Verify the spatial dimension data by comparing it with a known geographic information database or other data sources, and correct or modify any erroneous data. Those skilled in the art will know how to implement this, and will not be detailed here.

[0220] 3.5. Reduce Redundancy through Data Compression: Reduce data redundancy and improve data processing efficiency. Common data compression technologies include: Lossless compression technologies, such as Huffman coding and the Lempel-Ziv algorithm (LZ77, LZ78), which can compress data while maintaining data integrity. Lossy compression technologies, such as the discrete cosine transform (DCT) for image compression and the wavelet transform for signal compression. Those skilled in the art can select the desired data compression technology based on their specific needs.

[0221] 3.6. End data preprocessing.

[0222] Step 4: Perform bias correction and data optimization on the preprocessed data.

[0223] 4.1 Perform bias correction on the preprocessed data.

[0224] Comparing the data before and after time-space alignment, the electricity speed detection monitoring data shows significant changes before and after the time-space alignment processing. Without time-space alignment, the electrical parameters (current, voltage, etc.) are highly continuous and change rapidly, and the clock difference will cause the data to be unable to align. For example, there is a time deviation in the waveform of the collected data and the detected data, that is, the X-axis deviation. By performing time synchronization and spatial position calibration on the original data, the continuity between the data is significantly enhanced, and the time breakpoints and spatial dislocation phenomena are greatly reduced. Specifically, the aligned data can more accurately reflect the actual fluctuations in electricity consumption, the boundaries between the nighttime trough and the daytime peak are clearer, and the differences in electricity consumption among users in different industries are more reasonably presented. In addition, the data deviation rate is significantly reduced and the consistency is significantly improved, providing a more reliable data basis for subsequent deviation characteristic analysis and optimization strategy formulation.

[0225] Deviation characteristics are a key consideration after the spatiotemporal alignment of power consumption monitoring data, directly impacting the accuracy and reliability of power data. The types of deviations exhibited by this alignment primarily include systematic deviation, random deviation, and outlier deviation.

[0226] Systematic deviations are often caused by systemic factors such as inconsistent calibration, algorithm differences, or transmission delays in devices without a built-in Beidou module. This manifests as an overall data deviation from a fixed value or trend. Specifically, inconsistent calibration in devices without a built-in Beidou module means that the module uses external timing, and subsequent time is measured by the device's own clock, which varies from device to device. Furthermore, the timing process can also be affected by transmission links and signals, leading to overall systematic deviations.

[0227] Random deviations arise from random processes such as measurement noise and small fluctuations in environmental factors, causing the data to fluctuate around the expected value.

[0228] Outlier deviations are data points that significantly deviate from the normal value due to extreme events or equipment failures. This deviation is different from the outliers in the uniformly collected data collected by the device, which are eliminated by the preprocessing mentioned above. This deviation manifests as a discrepancy between different devices. For example, if there are three measurement points, one of them may show significant data discrepancies with the other two. To address outlier deviations, advanced statistical techniques such as modified maximum likelihood estimation, Bayesian estimation, Jackknife, and Bootstrap are used, combined with a dynamic baseline algorithm to automatically generate personalized user electricity usage models (with the peak threshold relaxed to 1.5σ). Simultaneously verifying harmonic distortion and electromagnetic field anomalies (THD > 15% or sudden increases in magnetic field strength > 50mT) effectively reduces parameter estimation bias. Cause analysis reveals that improving device accuracy, optimizing data processing algorithms, and strengthening system maintenance are key to reducing deviations and improving data quality.

[0229] 4.2 Optimize the bias-corrected data.

[0230] After performing deviation correction, a data optimization strategy is constructed. In step 1, precise spatiotemporal alignment technology is used to ensure the accuracy of the time of each monitoring sensor and reduce data alignment deviations caused by time skew. Specifically, after spatiotemporal alignment of data from different devices, the temporal deviation characteristics are significantly reduced, which can reduce the uncertainty caused by data time differences in subsequent data processing.

[0231] Optimization of the bias-corrected data is based on optimizing database query efficiency. This optimization utilizes caching technologies such as Redis to handle high-frequency queries with low timeliness requirements, thereby reducing database pressure. Furthermore, read-write separation and master-slave replication strategies are implemented to improve database performance in high-concurrency scenarios. For example, MySQL's master-slave replication can be used to centralize write operations on the master database and distribute read operations to slaves for load balancing. Another common approach is PostgreSQL's streaming replication, which synchronizes data between the master and slaves, ensuring high availability and improving read performance. Furthermore, integrating with distributed database systems (such as Cassandra or CockroachDB) further improves data scalability and fault tolerance, maintaining efficient data access performance even under large-scale data processing. Furthermore, database sharding is used to partition the database horizontally (by location) or vertically (by time) to address storage bottlenecks and concurrency pressure. Implementing these strategies significantly improves the quality of metering and monitoring data, reduces data bias, and enhances the timeliness and accuracy of data processing, laying the foundation for subsequent data analysis and applications.

[0232] 4.3 Process the Beidou timing and measurement clock errors.

[0233] The power system metering clock refers to the internal clock of each terminal. It receives Beidou satellite signals and uses the high-precision atomic clocks on board as a time reference for clock calibration and synchronization. The Beidou clock system is renowned for its exceptional accuracy and stability. Its accuracy reaches sub-nanosecond levels, with errors controlled to less than one microsecond per day, ensuring the high-precision time synchronization required. In practical applications, the main problem encountered is the accumulation of synchronization errors between the power system metering clock and the Beidou clock signal, resulting in time deviations exceeding the allowable range. Alternatively, problems with individual energy meters can cause clock errors to be excessively large. This can lead to inconsistencies between the Beidou direct timing time coordinates in the electricity consumption monitoring / testing data and the actual metering clock time coordinates in the meter reading data, creating uncertainty about the legitimacy of the data for forensic evidence collection. The system must record the errors between the two clocks for forensic evidence collection to avoid disputes. The errors are corrected at a set first time interval to ensure the accuracy of the power system metering clock at each terminal. Those skilled in the art can determine the first time interval based on actual site conditions. The first time interval value proposed in this embodiment of the present invention is merely a preferred embodiment and is not a limitation. Preferably, the first time interval is set to one hour.

[0234] Step 5: See attached Figure 2 , analyze the optimized data;

[0235] Real-time correlation analysis of multi-dimensional transient anomaly data is performed. Data collected from various checkpoints along the power transmission line, after being processed through spatiotemporal alignment and deviation analysis, is now ready for correlation analysis and diagnosis. Initial data comparison and diagnosis can be performed, including both horizontal and vertical comparisons. Horizontal comparison involves correlating and analyzing the characteristics of abnormal changes in data at different monitoring / detection locations along the power transmission line at the same moment. For example, if electricity theft occurs on phase A on the low-voltage side, a horizontal comparison can reveal significant differences in data between the high- and low-voltage sides based on the transformer's primary and secondary turns ratio. Vertical comparison involves correlating and analyzing the characteristics of abnormal changes in data at different moments at the same monitoring / detection location. This analysis can generally identify anomalies based on the coupling relationship between the power system metering principles between collection points (e.g., the transformer's primary and secondary turns ratio coupling, the current addition relationship at the same node), and historical user power consumption (e.g., a user's power consumption was 500 kWh last summer and the year before, but 50 kWh this year; or, for example, a user's daytime power consumption was 50 kWh but suddenly dropped to 5 kWh).

[0236] 5.1 Correlation analysis of data at different locations on horizontal power transmission lines at the same time

[0237] The metering system operates according to specific principles, and the electrical parameter data collected at each point in the horizontal power transmission line at the same time has a clear and fixed operational relationship. Taking the current as an example, a typical electrical parameter, at the same time and in the same phase, it presents a variety of relevant characteristics (such as the ratio relationship between the primary and secondary sides of the transformer, the node current relationship, etc.); Taking the power consumption as an example, the power consumption measured on the primary and secondary sides of the transformer has W1 = W2 + W 损 The logical relationship is: W1 is the power data of the primary side of the transformer; W2 is the power data of the secondary side of the transformer; W 损 It is the power loss data from the primary side of the transformer to the secondary side of the transformer.

[0238] Those skilled in the art will understand that the transformation ratio between the primary and secondary sides of a transformer refers to the ratio of the currents on the high- and low-voltage sides of the transformer. For a transformer, the ratio of the currents on the high- and low-voltage sides is closely related to the transformer's transformation ratio. This is determined by the basic operating principle of the transformer, which achieves voltage and current conversion through electromagnetic induction. Ideally, the ratio of the currents on the high- and low-voltage sides strictly follows the transformation ratio rule.

[0239] 5.2 Comparative analysis of data at different times at the same longitudinal monitoring / detection location

[0240] Comparing longitudinal data at different times at the same monitoring / detection location reveals changing trends in a user's electricity load. This comparison method was used to conduct a comparison experiment on past electricity theft cases. The results showed that this longitudinal data comparison strategy for the same monitoring / detection location at different times is very helpful for tracking period theft or intermittent theft, such as "stopping when someone arrives and stealing when someone leaves." This data can essentially expose the theft behavior. Combined with multi-dimensional transient information such as the electromagnetic field strength signal of the meter box environment during abnormal periods, the electrical quantity waveform information of the metering transformer, and the DC component signal, the user's electricity theft method can be essentially determined.

[0241] Step 6: Determine abnormal power consumption.

[0242] 6.1 Determine the suspicion of electricity theft;

[0243] For each distribution transformer, the primary load current on the 10kV line is monitored and compared with the relevant data of the energy meter corresponding to the electricity consumption information collection system to determine the suspicion of electricity theft. The calculation formula for the suspicion of electricity theft is:

[0244]

[0245] Where k is an integer, representing the kth distribution transformer area where the power consumption rapid inspection monitoring data is collected; K theft,t,krepresents the suspicion of electricity theft in the kth distribution transformer area at time t, I r,t,k is the actual load current of the transformer in the kth distribution transformer area at time t, that is, the current data measured by the transformer primary side monitoring equipment on the 10kV line; I e,t,k It is the sum of the current data measured by the electric energy meter in the k-th distribution transformer area (the current collected by the smart meter) converted according to the primary and secondary side transformation ratio of the transformer at time t, that is, the sum of the currents on the power user side of the k-th distribution transformer area, that is, the load power recorded by the electric energy meter.

[0246] When the suspicion of electricity theft in a distribution transformer area is greater than a set first threshold, it is determined that there is abnormal electricity usage in the distribution transformer area. Preferably, the first threshold is set to 3% to 5%.

[0247] This monitoring method uses the primary load current as a benchmark, is unaffected by the metering loop current, and can accurately reflect the difference between the metered data and the actual power load data. The present invention also aims to design on-site detection equipment for power metering devices on lines of 10kV and below, as well as on-site detection equipment for electrical equipment, connecting cables, and operating modes that may cause inaccurate or failed power metering. The on-site detection equipment must be capable of real-time current measurement, voltage measurement, waveform measurement, harmonic measurement, ratio measurement, and phasor analysis, enabling rapid determination of the location and type of power theft.

[0248] 6.2 Based on the established power usage anomaly classification criteria, determine the users with abnormal power usage in the distribution transformer area with abnormal power usage, as well as the type of abnormal power usage for each user. Each power usage anomaly classification criteria is established based on the rapid power usage monitoring data collected from the user side. The abnormal power usage types include high-frequency power theft, strong magnetic power theft, remote control power theft, and leakage power.

[0249] The various power consumption abnormality type judgment criteria include high-frequency power theft abnormality judgment criteria, strong magnetic power theft abnormality judgment criteria, remote control power theft abnormality judgment criteria and leakage fault judgment criteria.

[0250] The correlation comparison based on time-space alignment data is effective for typical electricity theft methods found so far. The comparison method of the present invention for abnormal data is verified as follows:

[0251] High-frequency electricity theft involves using high-frequency signals to interfere with the meter's metering chip, causing it to freeze or frequently reset, thereby rendering the meter inoperable and allowing the user to steal electricity. High-frequency interference devices (400MHz to 470MHz) are typically hidden approximately 20 meters from the meter box, with high-frequency wires routed outside the box. This type of electricity theft is highly covert and often monitored. Once detected, the theft is quickly halted and the tools concealed, making traditional inspection methods difficult to detect. If a longitudinal comparison method is used for data at different times at the same monitoring / detection location, it will be found that the current measurement value of the electricity user's meter is sometimes good and sometimes bad. When it is normal, it can match well with other checkpoints in the metering chain. When it is normal, it can match well with other checkpoints in the metering chain. When the current at other monitoring points increases or decreases, the reading of this meter should also show corresponding changes, rather than completely different behavior. If the current value of the meter is found to be obviously mismatched during the comparison process, especially the current value of 0 or abnormally low during certain specific time periods, this is a significant feature of high-frequency electricity theft. If the electromagnetic wave intensity is found to be increased during this period in combination with the meter box environmental detection data, it can be concluded that periodic high-frequency electricity theft has occurred.

[0252] Taking the above factors into consideration, the present invention proposes a criterion for judging high-frequency electricity theft anomalies;

[0253] When the electricity fast detection monitoring data of a certain user side meets the high-frequency electricity theft abnormality judgment criteria, it is determined that the user has a high-frequency electricity theft abnormality;

[0254] The abnormal judgment criteria for high-frequency electricity theft include:

[0255] Determine whether a user-side current (current collected by the smart meter) is less than a set first current threshold; preferably, the first current threshold is set to 2-3A;

[0256] If the user-side current is less than the first current threshold, determine whether the time during which the user-side current is less than the first current threshold is greater than a set first time threshold T1; the first time threshold is set to 5-7 minutes;

[0257] If the current value is greater than the first time threshold, and the user has current data less than the first current threshold during the same period within the set first historical period, then it is determined that the first high-frequency electricity theft condition is met; preferably, the set first historical period is one week to one month; the same period refers to the same time period every day;

[0258] If it is greater than the first time threshold, and the average current historical data of other users in the same substation area as the user within [T-T1, T] is greater than the set first current threshold, then it is determined that the second high-frequency electricity theft condition is met; where T represents the current time;

[0259] If the user side current / voltage / power load / power consumption is within [TT L1 -0.5T2,TT L1 ] and the average value in [TT L1 -T2,TT L1 -0.5T2], and the sign of the difference between the average values ​​within [TT L1 -0.5T2,TT L1 ] and the average value in [TT L1 -T2,TT L1 -0.5T2], the difference between the average values ​​has the same sign, then it is determined that the third high-frequency electricity theft condition is met; where T L1 The time during which the user-side current is less than the first current threshold up to this moment; T2 is the set second time threshold; the second time threshold is shown in the following formula:

[0260]

[0261] Among them, T L2 The current from the user side to TT L1 The time that the current is greater than the first current threshold until the moment T D1 is the first time upper limit threshold set; T D2 The time required to collect two data points; the first time upper limit threshold T D1 The preferred value range is 3-6 hours;

[0262] If there is high-frequency current in the user-side current, it is determined that the fourth high-frequency electricity theft condition is met;

[0263] If the user side magnetic field strength is [TT L1 ,T] is greater than the average value in [TT L1 -T3,TT L1 ] is an average value, then it is determined that the fifth high-frequency electricity theft condition is met; wherein T3 is a set third time threshold; preferably, the third time threshold is set to 5-10 minutes;

[0264] If a user satisfies the first, second, third, fourth and fifth high-frequency electricity theft conditions simultaneously, it is determined that high-frequency electricity theft occurs at the user.

[0265] When the electric speed detection monitoring data of a certain user side meets the strong magnetic power theft abnormality judgment criteria, it is determined that the user has a strong magnetic power theft abnormality;

[0266] Strong magnetic theft attempts to steal electricity by magnetically saturating the sampling CT inside a multi-function meter, reducing the metered current. Criminals often place magnets behind the meter box to alter the magnetic field, saturating the CT inside the meter. This allows them to avoid detection without breaking the seal or opening the meter box, and the magnets can be easily removed at any time. This type of theft is highly covert and difficult to detect, and is often time-sensitive. However, a longitudinal comparison of data from the same monitoring / testing location at different times reveals that the current readings at the user's meter fluctuate. Comparing the energy, current, voltage, and power readings at that measurement point with those at other measurement points reveals a good match between the two measurement points during normal periods. For example, the current value of phase A at measurement point 1 is similar to that at measurement point 2. However, when strong magnetic theft occurs, a longitudinal comparison reveals significant deviations between the current values ​​at measurement points 1 and 2 during that period. For example, a reading of 5A at measurement point 2 might be 0.5A or even 0A at measurement point 1. This is a notable feature of strong magnetic electricity theft.

[0267] Taking the above factors into consideration, the present invention proposes a strong magnetic power theft abnormality judgment criterion;

[0268] Abnormal judgment criteria for strong magnetic power theft include:

[0269] Determine whether a user-side current (current collected by the smart meter) is less than a set second current threshold; preferably, the second current threshold is set to 2-3A;

[0270] If the user-side current is less than the second current threshold, determine whether the time during which the user-side current is less than the second current threshold is greater than a set fourth time threshold T4; preferably, the fourth time threshold is set to 5-7 minutes;

[0271] If the current value is greater than the fourth time threshold, and the user has current data less than the second current threshold during the same period within the set second historical period, then it is determined that the first strong magnetic power theft condition is met; preferably, the set second historical period is one week to one month; the same period refers to the same time period every day;

[0272] If it is greater than the fourth time threshold, and the average current historical data of other users in the same substation area as the user within [T-T4, T] is greater than the set second current threshold, then it is determined that the second strong magnetic power theft condition is met; where T represents the current time;

[0273] If the user side current / voltage / power load / power consumption is within [TT L4 -0.5T5,TT L4 ] and the average value in [TT L4 -T5,TT L4-0.5T5], and the sign of the difference between the average values ​​within [TT L4 -0.5T5,TT L4 ] and the average value in [TT L4 -T5,TT L4 -0.5T5], the difference between the average values ​​has the same sign, then it is judged that the third strong magnetic stealing condition is met; where T L4 is the duration of time during which the user-side current is less than the second current threshold up to this moment; T5 is the set fifth time threshold; the fifth time threshold is expressed as follows:

[0274]

[0275] Among them, T L5 The current from the user side to TT L4 The time that the current is greater than the second current threshold until the moment T D3 is the second time upper limit threshold set; T D2 The time required to collect two data points; the second time upper limit threshold T D3 The preferred value range is 3-6 hours;

[0276] If there is no high-frequency current (high-frequency current collected by the high-frequency current sensor) in the user-side current, where the high-frequency current refers to a current with a frequency of 400 MHz to 470 MHz, then it is determined that the fourth strong magnetic power theft condition is met;

[0277] If the user side magnetic field strength is [TT L4 ,T] is greater than the average value in [TT L4 -T6,TT L4 ] is an average value, then it is determined that the fifth strong magnetic power theft condition is met; wherein T6 is a set sixth time threshold; preferably, the sixth time threshold is set to 5-10 minutes;

[0278] If a user meets the first, second, third, fourth and fifth strong magnetic power theft conditions at the same time, it is determined that strong magnetic power theft exists at the user.

[0279] When the electricity rapid detection monitoring data of a certain user side meets the remote control electricity theft abnormality judgment criteria, it is determined that the user has remote control electricity theft abnormality;

[0280] Electricity thieves secretly install wireless receivers and relays in electricity metering devices in order to achieve the purpose of undercounting electricity through remote control diversion or voltage division. This method of electricity theft is more covert and the perpetrators are more daring. They are often in a state of stealing electricity all the time, and only give up stealing electricity by remote control when being inspected, making the inspection fruitless. From the horizontal comparison of data at different times at the same monitoring / detection location, it can be seen that the measurement is normal only during the inspection period, and is continuously undercounted at other times, that is, the data is continuously lower than that of other checkpoints in the metering chain; there is no abnormal electromagnetic field strength. The measurement is normal during the inspection period, that is, the average value of the data at this measurement point is not much different from that of other checkpoints in the metering chain; the data is continuously undercounted at other times, that is, the data at this measurement point is continuously lower than that of other checkpoints in the metering chain;

[0281] Taking the above factors into consideration, the present invention proposes a remote control electricity theft abnormality judgment criterion;

[0282] Remote control electricity theft abnormality judgment criteria include:

[0283] Calculate the absolute value of the difference between the average current / voltage / power load / power consumption of each user during the inspection period and the average current / voltage / power load / power consumption of other users in the same substation area during the same period, and record it as the first absolute value of each user.

[0284] If the first absolute value of a certain user side is less than [T cel ,T cs ] period and the absolute value of the difference between the average current / voltage / power load / power consumption of the user side and the average current / voltage / power load / power consumption of other users in the same substation area during the same period, it is determined that the first remote control electricity theft condition is met; wherein, T cel is the end time of the last inspection, T cs This is the start time of this inspection;

[0285] If the first absolute value on the user side is less than the set first difference threshold, it is determined that the second remote control electricity theft condition is met; preferably, the first difference threshold is set to 2% to 5% of the absolute value of the difference between the average current / voltage / power load / power consumption of other users on the same substation area during the inspection period;

[0286] If [T cel ,T cs ] period and the absolute value of the difference between the average current / voltage / power load / power consumption of the user side and the average current / voltage / power load / power consumption of other users in the same substation area during the same period is greater than the set second difference threshold, then it is determined that the third remote control electricity theft condition is met; preferably, the second difference threshold is set to [T cel ,T cs] 2% to 5% of the absolute value of the difference between the average current / voltage / power load / power consumption of other users in the same substation area during the same period;

[0287] If a user satisfies the first, second and third remote control electricity theft conditions at the same time, it is determined that remote control electricity theft occurs at the user.

[0288] When the electric fast detection monitoring data of a certain user side meets the leakage fault judgment criteria, it is determined that there is a leakage fault at the user;

[0289] To combat remote-controlled electricity theft, real-time spectrum analysis (400-470MHz) combined with current drop and electromagnetic surge characteristics can be used to quickly lock onto the target, reducing on-site evidence collection time from 2 hours to 8 seconds.

[0290] The current at the main incoming line to a residential complex should equal the sum of the currents at the incoming lines to each unit building. The current at each unit building's incoming line should also equal the sum of the currents at the meters of each household within that unit. When an electrical appliance in a household experiences a leakage fault, the current measured by the current collection point at that household's meter will be higher than when the appliance is operating normally. This abnormal increase in current will not be reflected in the meters of other households located laterally. By real-time monitoring and correlation analysis of current data from various collection points along the power transmission lines, the abnormal user can be quickly located, allowing timely action to address the situation and prevent further escalation of the leakage incident.

[0291] Taking the above factors into consideration, the present invention proposes a criterion for judging leakage faults;

[0292] The criterion for leakage fault is:

[0293] If a user's current is greater than the maximum current on the user's side during the [T-T7, T] period, determine whether there are other users in the same substation area whose current is greater than the maximum current on the corresponding user's side during the [T-T7, T] period. If not, determine that the user has a power leakage. T7 is the seventh time threshold. Preferably, T7 is 1 to 2 days.

[0294] This application also discloses a power consumption anomaly monitoring and detection system based on the power consumption anomaly monitoring and detection method, including a terminal time-space alignment module, a data acquisition module, a power consumption anomaly area preliminary judgment module, a power consumption anomaly area determination module, and a power consumption anomaly user and type determination module:

[0295] The terminal space-time alignment module performs space-time alignment on each sampling terminal through the Beidou satellite navigation system;

[0296] The data acquisition module collects power consumption rapid detection monitoring data on the power transmission line through each sampling terminal; the power consumption rapid detection monitoring data includes the current and voltage at the primary side incoming cable of the transformer, as well as the current, voltage, magnetic field strength, power load and power consumption of each user side; the power consumption abnormal area preliminary judgment module preliminarily judges the distribution transformer area with abnormal power consumption based on the power consumption rapid detection monitoring data;

[0297] The module for initially judging the abnormal power consumption area preliminarily judges the distribution transformer area with abnormal power consumption based on the power consumption rapid detection monitoring data;

[0298] The abnormal power consumption area determination module calculates the power theft suspicion of each distribution transformer area preliminarily determined to have abnormal power consumption, and determines the distribution transformer area with abnormal power consumption based on the calculation result;

[0299] The module for determining users and types of abnormal electricity consumption determines users with abnormal electricity consumption and their types of abnormal electricity consumption in the distribution transformer area with abnormal electricity consumption based on the set criteria for each type of abnormal electricity consumption; each criterion for the type of abnormal electricity consumption is set based on the rapid electricity consumption monitoring data collected from the user side.

[0300] The present disclosure may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0301] A computer-readable storage medium can be a tangible device that can hold and store instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination thereof. As used herein, a computer-readable storage medium is not to be construed as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through an electrical wire.

[0302] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0303] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, and conventional procedural programming languages ​​such as "C" language or similar programming languages. Computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., utilizing an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be personalized by utilizing the state information of the computer-readable program instructions. The electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0304] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A method for monitoring and detecting abnormal power consumption on site based on Beidou time and space reference, characterized in that: include: The sampling terminals are aligned in time and space through the BeiDou satellite navigation system; Each sampling terminal collects electricity consumption rapid detection monitoring data on the power transmission line; The power consumption rapid detection monitoring data includes the current and voltage at the primary side incoming cable of the transformer, as well as the current, voltage, magnetic field strength, power load and power consumption of each user side; Based on the power consumption rapid inspection monitoring data, it is preliminarily determined that there are distribution transformer substations with abnormal power consumption; Calculate the suspicion of electricity theft for each distribution transformer substation that is initially judged to have abnormal electricity usage, and determine the distribution transformer substation with abnormal electricity usage based on the calculation results; Based on the set power consumption abnormality type judgment criteria, the users with power consumption abnormality and their power consumption abnormality types in the distribution transformer substation with power consumption abnormality are determined; the power consumption abnormality type judgment criteria are set based on the power consumption rapid detection monitoring data collected on the user side.

2. The method for monitoring and detecting abnormal power consumption according to claim 1, wherein: The said performing time and space alignment of each sampling terminal through the Beidou satellite navigation system means achieving synchronization of the time references of the built-in clocks of different sampling terminals and unification of the spatial coordinates through Beidou satellite signals; The terminals include devices with built-in Beidou modules and devices without built-in Beidou modules.

3. The method for monitoring and detecting abnormal power consumption according to claim 1 or 2, characterized in that: Devices with built-in Beidou modules achieve spatiotemporal alignment based on Beidou satellite time and device positioning signals; devices without built-in Beidou modules achieve spatiotemporal alignment based on Beidou satellite time and device positioning signals received by devices with built-in Beidou modules.

4. The method for monitoring and detecting abnormal power consumption according to claim 1, wherein: Based on the power consumption rapid inspection monitoring data, it is preliminarily determined that there are distribution transformer areas with abnormal power consumption, and the display matrix of the distribution transformer areas with abnormal power consumption is calculated based on the power consumption rapid inspection monitoring data; When the value of the element in row k and column k in the display matrix of the abnormal power consumption distribution transformer area is non-zero, it is preliminarily judged that the kth distribution transformer area is the distribution transformer area with abnormal power consumption, where k is an integer, representing the kth distribution transformer area where the power consumption rapid detection monitoring data is collected.

5. The method for monitoring and detecting abnormal power consumption according to claim 4, characterized in that: The display matrix of the distribution transformer area with abnormal power consumption is as follows: Among them, ΔY is the display matrix of the distribution transformer area with abnormal power consumption; V m is the distribution transformer area voltage measurement value vector obtained based on the voltage at each user side; ΔI theft is the current vector of the equivalent current of electricity theft; I m is the phasor of the injection current measurement value of the distribution transformer substation based on the current at each user side; Y is the node admittance matrix.

6. The method for monitoring and detecting abnormal power consumption according to claim 1, wherein: Calculate the suspicion of electricity theft for each distribution transformer area that is initially judged to have abnormal electricity usage. Based on the calculation results, determine the distribution transformer area with abnormal electricity usage, including: Real-time calculation of the suspicion of electricity theft for each distribution transformer substation initially identified as having abnormal electricity usage; the calculation formula for the suspicion of electricity theft is based on the current at the primary-side incoming cable of the transformer and the current settings at each user side; When the suspicion of electricity theft in a distribution transformer area is greater than a set first threshold, it is determined that abnormal electricity usage exists in the distribution transformer area.

7. The method for monitoring and detecting abnormal power consumption according to claim 1 or 6, characterized in that: The calculation formula for the suspicion of electricity theft is as follows: Where k is an integer, representing the kth distribution transformer area where the power consumption rapid inspection monitoring data is collected; K theft,t,k is the suspicion of electricity theft in the kth distribution transformer area at time t; I r,t,k is the current at the primary side incoming cable of the kth distribution transformer station at time t; I e,t,k It is the total current on the user side of the kth distribution transformer substation converted at time t.

8. The method for monitoring and detecting abnormal power consumption according to claim 1, wherein: The various power consumption abnormality type judgment criteria include high-frequency power theft abnormality judgment criteria, strong magnetic power theft abnormality judgment criteria, remote control power theft abnormality judgment criteria and leakage fault judgment criteria.

9. The method for monitoring and detecting abnormal power consumption according to claim 1 or 8, characterized in that: When the electricity fast detection monitoring data of a certain user side meets the high-frequency electricity theft abnormality judgment criteria, it is determined that the user has a high-frequency electricity theft abnormality; The abnormal judgment criteria for high-frequency electricity theft include: Determine whether a user-side current is less than a set first current threshold; If the user-side current is less than the first current threshold, determining whether the time during which the user-side current is less than the first current threshold is greater than a set first time threshold T1; If it is greater than the first time threshold, and the user has current data less than the first current threshold during the same period within the set first historical period, then it is determined that the first high-frequency electricity theft condition is met; If it is greater than the first time threshold, and the average current historical data of other users in the same substation area as the user within [T-T1, T] is greater than the set first current threshold, then it is determined that the second high-frequency electricity theft condition is met; where T represents the current time; If the user side current / voltage / power load / power consumption is within [TT L1 -0.5T2,TT L1 ] and the average value in [TT L1 -T2,TT L1 -0.5T2], and the sign of the difference between the average values ​​within [TT L1 -0.5T2,TT L1 ] and the average value in [TT L1 -T2,TT L1 -0.5T2], the difference between the average values ​​has the same sign, then it is determined that the third high-frequency electricity theft condition is met; where T L1 T2 is the duration of the user-side current being less than the first current threshold; T3 is the set second time threshold; If there is high-frequency current in the user-side current, it is determined that the fourth high-frequency electricity theft condition is met; If the user side magnetic field strength is [TT L1 ,T] is greater than the average value in [TT L1 -T3,TT L1 ], it is determined that the fifth high-frequency electricity theft condition is met; wherein T3 is the set third time threshold; If a user satisfies the first to fifth high-frequency electricity theft conditions simultaneously, it is determined that high-frequency electricity theft occurs at the user.

10. The method for monitoring and detecting abnormal power consumption according to claim 8, wherein: When the electric speed detection monitoring data of a certain user side meets the strong magnetic power theft abnormality judgment criteria, it is determined that the user has a strong magnetic power theft abnormality; Abnormal judgment criteria for strong magnetic power theft include: Determine whether a user-side current is less than a set second current threshold; If the user-side current is less than the second current threshold, determining whether the time during which the user-side current is less than the second current threshold is greater than a set fourth time threshold T4; If it is greater than the fourth time threshold, and the user has current data less than the second current threshold during the same period in the set second historical period, then it is determined that the first strong magnetic power theft condition is met; If it is greater than the fourth time threshold, and the average current historical data of other users in the same substation area as the user within [T-T4, T] is greater than the set second current threshold, then it is determined that the second strong magnetic power theft condition is met; where T represents the current time; If the user side current / voltage / power load / power consumption is within [TT L4 -0.5T5,TT L4 ] and the average value in [TT L4 -T5,TT L4 -0.5T5], and the sign of the difference between the average values ​​within [TT L4 -0.5T5,TT L4 ] and the average value in [TT L4 -T5,TT L4 -0.5T5], the difference between the average values ​​has the same sign, then it is judged that the third strong magnetic stealing condition is met; where T L4 The duration of the user-side current being less than the second current threshold up to this moment; T5 is the set fifth time threshold; If there is no high-frequency current in the user-side current, it is determined that the fourth strong magnetic power theft condition is met; If the user side magnetic field strength is [TT L4 ,T] is greater than the average value in [TT L4 -T6,TT L4 ], it is determined that the fifth strong magnetic power theft condition is met; wherein T6 is the set sixth time threshold; If a user meets the first to fifth strong magnetic power theft conditions at the same time, it is determined that strong magnetic power theft exists at the user.

11. The method for monitoring and detecting abnormal power consumption according to claim 1, wherein: When the electricity rapid detection monitoring data of a certain user side meets the remote control electricity theft abnormality judgment criteria, it is determined that the user has remote control electricity theft abnormality; Remote control electricity theft abnormality judgment criteria include: Calculate the absolute value of the difference between the average current / voltage / power load / power consumption of each user during the inspection period and the average current / voltage / power load / power consumption of other users in the same substation area during the same period, and record it as the first absolute value of each user. If the first absolute value of a certain user side is less than [T cel ,T cs ] period and the absolute value of the difference between the average current / voltage / power load / power consumption of the user side and the average current / voltage / power load / power consumption of other users in the same substation area during the same period, it is determined that the first remote control electricity theft condition is met; wherein, T cel is the end time of the last inspection, T cs This is the start time of this inspection; If the first absolute value on the user side is less than the set first difference threshold, it is determined that the second remote control electricity theft condition is met; If [T cel ,T cs If the absolute value of the difference between the average current / voltage / power load / power consumption of the user side during the period and the average current / voltage / power load / power consumption of other users in the same substation area during the same period is greater than the set second difference threshold, it is determined that the third remote control electricity theft condition is met; If a user satisfies the first, second and third remote control electricity theft conditions at the same time, it is determined that remote control electricity theft occurs at the user.

12. The method for monitoring and detecting abnormal power consumption according to claim 1, wherein: When the electric fast detection monitoring data of a certain user side meets the leakage fault judgment criteria, it is determined that there is a leakage fault at the user; The criterion for leakage fault is: If the current current of a user is greater than the maximum current on the user side in the [T-T7, T] period, determine whether there are other users in the same substation area whose current current is greater than the maximum current on the corresponding user side in the [T-T7, T] period. If not, determine that there is a leakage at the user; where T7 is the set seventh time threshold.

13. A power consumption anomaly monitoring and detection system using the power consumption anomaly monitoring and detection method according to any one of claims 1 to 12, characterized in that: It includes a terminal time-space alignment module, a data collection module, a preliminary judgment module for abnormal power consumption areas, a module for determining abnormal power consumption areas, and a module for determining abnormal power consumption users and types: The terminal space-time alignment module performs space-time alignment on each sampling terminal through the Beidou satellite navigation system; The data acquisition module collects power consumption rapid detection monitoring data on the power transmission line through each sampling terminal; the power consumption rapid detection monitoring data includes the current and voltage at the primary side incoming cable of the transformer, as well as the current, voltage, magnetic field strength, power load and power consumption of each user side; the power consumption abnormal area preliminary judgment module preliminarily judges the distribution transformer area with abnormal power consumption based on the power consumption rapid detection monitoring data; The module for initially judging the abnormal power consumption area preliminarily judges the distribution transformer area with abnormal power consumption based on the power consumption rapid detection monitoring data; The abnormal power consumption area determination module calculates the power theft suspicion of each distribution transformer area preliminarily determined to have abnormal power consumption, and determines the distribution transformer area with abnormal power consumption based on the calculation result; The module for determining users and types of abnormal electricity consumption determines users with abnormal electricity consumption and their types of abnormal electricity consumption in the distribution transformer area with abnormal electricity consumption based on the set criteria for each type of abnormal electricity consumption; each criterion for the type of abnormal electricity consumption is set based on the rapid electricity consumption monitoring data collected from the user side.

14. A terminal comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the power consumption anomaly monitoring and detection method according to any one of claims 1-12.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the power consumption anomaly monitoring and detection method described in any one of claims 1 to 12 are implemented.

Images