High line loss and electricity larceny rapid diagnosis device and method

Abstract

The invention relates to the technical field of high line loss and electricity larceny diagnosis, in particular to a high line loss and electricity larceny rapid diagnosis device and method.The high line loss and electricity larceny rapid diagnosis device comprises line loss checking instruments which are deployed at an outgoing line switch and a meter box incoming line of a transformer area power grid and used for collecting three-phase real-time electric parameter data; and the analysis platform is in communication connection with the line loss checking instrument and is used for receiving data uploaded by the line loss checking instrument and executing line loss calculation and positioning analysis. The method has the advantages that topology automatic identification is achieved through feature signal injection and analysis, the line loss prediction model is constructed based on the phase delay amount and the intensity variation amount, the positioning precision is improved to the meter box level, furthermore, electric energy balance verification is completed through a localization communication architecture, dependence on a public network is eliminated, and the method is more effective. And in combination with GPS information, not only can physical position association and visual marking be realized, but also minute-level abnormity early warning is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of high line loss and electricity stealing diagnosis, and particularly relates to a high line loss and electricity stealing rapid diagnosis device and method. BACKGROUND

[0002] With the deepening of the construction of smart grid, the popularization of power utilization information collection system, high-speed power line carrier (HPLC) and other technologies provides a preliminary data basis for the line loss monitoring of transformer area. In the prior art, there are several anti-electricity stealing methods based on big data analysis or dynamic grid outlier detection, which aims to identify abnormal power utilization behavior by analyzing the sampling information and innovation variance of three-phase voltage.

[0003] However, the prior art scheme has the following limitations: firstly, the physical topology of the line and the signal transmission characteristics are not deeply utilized, resulting in insufficient line loss positioning accuracy and difficulty in accurately positioning to the meter box level; secondly, the traditional method relies on a centralized data platform and a public network such as 4G for communication and calculation, and cannot be effectively implemented in remote signal-free areas; thirdly, the establishment and maintenance of the transformer area equipment archives seriously depend on manual input, which is low in efficiency and prone to errors, and cannot meet the needs of rapid diagnosis and operation and maintenance. SUMMARY

[0004] To solve the above technical problems, the present application provides the following technical solutions: In a first aspect, the present application provides a high line loss and electricity stealing rapid diagnosis device, comprising a line loss investigation instrument, which is deployed at an outgoing switch and a meter box incoming line of a transformer area power grid, and is used to collect three-phase real-time electric parameter data; An analysis platform is in communication connection with the line loss investigation instrument, and is used to receive the data uploaded by the line loss investigation instrument, and perform line loss calculation and positioning analysis.

[0005] As a preferred scheme of the high line loss and electricity stealing rapid diagnosis device of the present application, the line loss investigation instrument comprises: A topology identification module is used to inject a characteristic signal into the power grid loop, and receive and analyze the characteristic signal from the lower node to obtain the timestamp, phase delay and intensity information of the signal; A data freezing module is used to freeze electric parameter data according to instructions; A data forwarding module is used to forward instructions and data through a target protocol.

[0006] As a preferred scheme of the high line loss and electricity stealing rapid diagnosis device of the present application, the analysis platform comprises, A topology model construction module is used to construct a topology model of the transformer area branch based on the identification result of the topology identification module and the device address information, in combination with GPS positioning data, wherein the topology model represents the line connection relationship and supports line loss prediction; The line loss calculation and early warning module is used for obtaining a predicted line loss rate of a monitored transformer area based on a topology model, associating device physical positions, and triggering early warning and visual marking when the predicted line loss rate is abnormal.

[0007] As a preferred scheme of the high line loss and electricity stealing rapid diagnosis device, the line loss investigation instrument further comprises a local maintenance module. The local maintenance module is used for storing and managing electric energy meter archives in the meter box, supporting automatic acquisition of electric meter barcode information, and associating the line loss investigation instrument with the meter relationship.

[0008] In a second aspect, the application provides a high line loss and electricity stealing rapid diagnosis method, which comprises the following steps: The subordinate line loss investigation instrument deployed at the meter box incoming line receives and analyzes the characteristic signal to obtain target data, and reports the target data and device address information to the analysis platform, wherein the target data comprises a timestamp, a phase delay amount and a strength change amount of the signal. The analysis platform completes topology identification and generates a topology model of the transformer area branch based on the target data and device address information reported by each subordinate line loss investigation instrument. Based on the topology model, the predicted line loss rate of the monitored transformer area is calculated, and the prediction result is associated to the device physical position in combination with GPS positioning information. The meter freezing data and the freezing data of the line loss investigation instrument are obtained, and the current and energy error values at the transformer area outgoing switch and the meter box incoming line are compared step by step based on the electric energy balance principle. An abnormal early warning is triggered by analyzing the change trend of the predicted line loss rate, and a high loss or negative loss interval is visually marked in the topology graph generated by the topology model.

[0009] As a preferred scheme of the high line loss and electricity stealing rapid diagnosis method, the topology model of the transformer area branch comprises, The analysis platform sends a topology identification instruction to the line loss investigation instrument. The line loss investigation instrument injects a characteristic signal with specific characteristic code bits and waveforms generated by the on-off of a constant resistance load or a constant current load in response to the instruction. After receiving the characteristic signal, the subordinate line loss investigation instrument analyzes and records the arrival timestamp, phase delay amount and signal strength attenuation amount of the signal, and reports the identification result containing the device address information. The analysis platform analyzes the hierarchical connection relationship between devices based on the information reported by each node, and constructs a transformer area branch topology model for describing line connection and supporting line loss prediction.

[0010] As a preferred scheme of the high line loss and electricity stealing rapid diagnosis method of the present application, wherein: the construction of the topological model depends on the physical characteristics of the characteristic signal in the transmission process; The quantitative relationship between the predicted line loss rate and the signal parameters is established, including, The actual line loss rates of a plurality of sample transformer areas are obtained based on field investigation, and the phase delay and the intensity change of the characteristic signal are recorded synchronously; Taking the phase delay and the intensity change as independent variables and the actual line loss rate as dependent variable, data fitting is performed to construct a prediction model. The prediction model is used to calculate the predicted line loss rate according to the real-time collected phase delay and intensity change.

[0011] As a preferred scheme of the high line loss and electricity stealing rapid diagnosis method of the present application, wherein: the step-by-step comparison of the current and power error values of the transformer outlet switch and the meter box inlet is performed by using a localized architecture.

[0012] As a preferred scheme of the high line loss and electricity stealing rapid diagnosis method of the present application, wherein: the method further comprises, The meter bar code is scanned by the field operation terminal to record the identification information of the electric energy meter in the meter box; The meter identification recorded by the scanning is automatically associated and bound with the line loss investigation instrument currently connected by the field operation terminal, a temporary equipment file is formed and uploaded to the analysis platform; When the analysis platform detects or the operator finds that the file association is abnormal through the field operation terminal, the field operation terminal starts the offline comparison mechanism to guide the operator to retrieve the secondary file data of the field meter and manually check and correct on the terminal.

[0013] As a preferred scheme of the high line loss and electricity stealing rapid diagnosis method of the present application, wherein: the method further comprises, after the diagnosis process is completed, automatically deleting the line loss investigation instrument temporary file created in the concentrator for this diagnosis.

[0014] Compared with the prior art, the present application has the following advantages: the line loss positioning is changed from the original carpet type investigation to accurate guidance, that is, by sending a characteristic signal to the target phase loop, and then analyzing the phase delay and attenuation degree when the signal is transmitted back, the position of electricity stealing or electricity leakage can be directly locked to a specific meter box, and the error is less than 5%, and minute-level abnormality early warning can be realized. The most critical is that the whole process does not depend on mobile phone signals, and even in remote mountainous areas, data acquisition can be realized by Bluetooth and power line carrier, the communication cost can be saved by more than 30%, and the data is safer. It is more convenient to operate, the meter and the equipment can be automatically bound through the scanning mode, and the garbage data can be automatically cleaned up after use, and the archiving time is compressed from several hours to several minutes. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0016] Figure 1 is a general architecture diagram; Figure 2 is a relationship diagram of the predicted line loss rate and the strength change amount; Figure 3 is a relationship diagram of the predicted line loss rate and the phase delay amount; Figure 4 is a relationship diagram of the predicted line loss rate and the strength change amount and the phase delay amount; Figure 5 is a code bit schematic diagram of the constant resistance load characteristic signal in the present application; Figure 6 is a waveform diagram of the constant resistance load characteristic signal in the present application; Figure 7 is a code bit schematic diagram of the constant current load characteristic signal in the present application; Figure 8 is a waveform diagram of the constant current load characteristic signal in the present application.

[0017] Figure 9 is a flowchart of the high line loss and electricity stealing rapid diagnosis method. DETAILED DESCRIPTION

[0018] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without any creative effort should be within the scope of protection of the present application.

[0019] Embodiment 1, refer to Figure 1 The first embodiment of the present application provides a high line loss and electricity stealing rapid diagnosis device, which comprises: A line loss investigation instrument is arranged at the outgoing switch and the meter box incoming line of the transformer area power grid, and is used to collect three-phase real-time electrical parameter data. It can be understood that the three-phase real-time electrical parameter data includes but is not limited to three-phase voltage, current, power, frequency, phase, electrical quantity and switch quantity state data.

[0020] It should be noted that the line loss troubleshooting instrument is connected to the monitored loop in parallel or series. Further, the field operation terminal performs initial configuration on the line loss troubleshooting instrument through Bluetooth / infrared, without changing the existing line structure.

[0021] Preferably, by distributed deployment at key nodes of the power grid (total outlet of the transformer area and inlet of each meter box), not only can a monitoring network covering all branches be constructed, providing homogeneous and synchronous electrical parameter basic data for subsequent topology identification and line loss calculation, but also the problem of unclear line loss distribution caused by insufficient monitoring points in the prior art is solved.

[0022] The analysis platform is in communication connection with the line loss troubleshooting instrument, and is used for receiving data uploaded by the line loss troubleshooting instrument, and performing line loss calculation and positioning analysis.

[0023] Further, the line loss troubleshooting instrument comprises: The topology identification module is used for injecting a characteristic signal into the power grid loop, and receiving and analyzing the characteristic signal from the lower node, so as to obtain the time stamp, phase delay and intensity information of the signal; It can be understood that the time stamp is used for calculating the delay, the phase delay is used for reflecting the line impedance, and the intensity information is used for reflecting the line loss.

[0024] It should be noted that this module is not limited to passive collection, but changes to actively inject an identifiable probe signal into the power grid. By analyzing the loss and delay of the signal from point A to point B, the connection relationship and electrical characteristics of the line can be accurately inferred. This not only replaces the traditional inefficient and error-prone manual survey, but also realizes the basis of meter box level positioning.

[0025] The data freezing module is used for freezing electrical parameter data according to instructions; It can be understood that, in order to accurately calculate the line loss, it is necessary to ensure that the power consumption data of the total outlet line and each branch are obtained at the same time. Specifically, according to the freezing instruction received from the analysis platform, all troubleshooting instruments and meter boxes are locked and the current voltage, current, power, and power data are recorded at the same time.

[0026] The data forwarding module is used for forwarding instructions and data through a target protocol; It should be noted that the data forwarding module supports multiple communication protocols (HPLC, RS485, Bluetooth, etc.), so as to ensure that the device can be connected in various transformer area environments.

[0027] The local maintenance module is used for storing and managing the meter box file of the electric energy meter, and supports automatic acquisition of the electric meter barcode information, so as to associate the line loss troubleshooting instrument with the meter box.

[0028] Further, the analysis platform comprises, a topology model construction module, configured to construct a topology model of the branch of the transformer area based on the identification result of the topology identification module and the device address information, in combination with GPS positioning data, wherein the topology model represents the connection relationship of the line and supports line loss prediction; It can be understood that the topology model construction module is used to fuse the electrical relationship and the physical position, and construct a digital twin model that can be used for analysis.

[0029] Specifically, the topology model construction module calculates the hierarchical relationship between devices by receiving the signal parameters and device addresses reported by all topology identification modules, and then binds these electrical nodes with the physical coordinates obtained through GPS to generate a visual transformer area power grid topology diagram containing geographic location information.

[0030] Preferably, this not only enables the operation and maintenance personnel to understand the power grid structure from the electrical logic, but also from the geographical space, laying the foundation for accurate positioning.

[0031] a line loss calculation and early warning module, configured to obtain the predicted line loss rate of the monitored transformer area based on the topology model, associate the device physical position, and trigger early warning and visual marking when the predicted line loss rate is abnormal.

[0032] It should be noted that when an anomaly (high line loss or negative line loss) is found, the abnormal interval (such as a specific meter box) is highlighted on the previously constructed visual topology diagram, and a minute-level warning is issued.

[0033] In summary, the present application has the beneficial effects of a high line loss and electricity stealing rapid diagnosis device, which constructs an efficient and accurate transformer area line loss and electricity stealing diagnosis system by integrating the cooperative work of the line loss investigation instrument and the analysis platform. The line loss investigation instrument realizes accurate injection and analysis of characteristic signals, real-time acquisition and solidification of electrical parameters, and automatic association of household meter relationship by virtue of topology identification, data freezing, multi-protocol forwarding and local archive management functions, providing a rich and reliable on-site data basis for diagnosis. Further, the analysis platform constructs a topology model that can represent the connection relationship of the line by using these data in combination with GPS information, and realizes the calculation of the predicted line loss rate, the association of the device position, and the rapid early warning and visual marking when abnormal, thereby improving the line loss positioning accuracy to the meter box level and greatly reducing the cost and error of manual inspection.

[0034] Embodiment 2, refer to Figure 9 As the first embodiment of the present application, the embodiment provides a high line loss and electricity stealing rapid diagnosis method, comprising: S100: injecting a characteristic signal into a target phase circuit by a line loss investigation instrument deployed at a transformer area outgoing switch; S200: receiving and analyzing the characteristic signal by the lower line loss troubleshooting instrument deployed at the meter box incoming line to obtain target data, and reporting the target data and its device address information to the analysis platform, wherein the target data includes the timestamp, phase delay amount and intensity change amount of the signal; S300: completing topology identification and generating a topology model of the transformer branch based on the target data and its device address information reported by each lower line loss troubleshooting instrument; S400: calculating the predicted line loss rate of the monitored transformer area based on the topology model, and associating the prediction result to the physical location of the device in combination with the GPS positioning information; S500: obtaining the frozen data of the meter and the frozen data of the line loss troubleshooting instrument, and comparing the current and energy error values of the transformer outgoing switch and the meter box incoming line based on the principle of energy balance; S600: triggering abnormal early warning by analyzing the change trend of the predicted line loss rate, and visually marking the high loss or negative loss interval in the topology graph generated by the topology model.

[0035] It should be noted that, in order to solve the problems of low positioning accuracy, difficult topology identification and dependence on public network communication in the prior art, the steps S100-S600 are used to realize automatic topology identification by characteristic signal injection and analysis, and a line loss prediction model is constructed based on the phase delay amount and intensity change amount to improve the positioning accuracy to the meter box level. Further, the energy balance verification is completed by using the localized communication architecture, which eliminates the dependence on public network. Furthermore, in combination with the GPS information, not only the physical location association and visual marking can be realized, but also the minute-level abnormal early warning can be realized.

[0036] Embodiment 3, refer to Figures 2-9 For an embodiment of the present application, based on the above embodiment, a high line loss and electricity stealing rapid diagnosis method is provided.

[0037] In the present application, step S100 is performed by the line loss troubleshooting instrument deployed at the transformer outgoing switch, which injects a characteristic signal into the target phase circuit, wherein the target phase circuit is the A phase circuit, including the following steps A1-A2: A1: the line loss troubleshooting instrument controls the on-off operation of the constant resistance load or constant current load inside it to generate a characteristic signal with specific characteristic code bits and waveform, and injects the signal into the target phase circuit.

[0038] Referring to the accompanying Figures 5-8 As shown in the figure, the line loss troubleshooting instrument can generate a non-random, device preset and unique current or power pulse sequence by controlling the on-off of the constant resistance load or constant current load, that is, a whistle of a specific frequency and rhythm (characteristic signal) is blown in the background noise of the power grid, so that the receiver can clearly distinguish it from the noise.

[0039] It should be noted that the power grid is a three-phase system, and if the signal is injected into three phases at the same time, the signal coupling and analysis will be very complex. If the signal is injected into the target phase, all downstream devices can establish a connection relationship by detecting the characteristic signal in the target phase loop, which can make the topology identification clear and feasible.

[0040] In use, the upper line loss troubleshooting instrument located at the outlet switch of the transformer area receives the topology identification instruction generated by the analysis platform / system, and controls a constant resistance load (consumes constant power) or a constant current load (generates constant current) through its internal electronic switch to turn on and off according to the preset code bit rule, so that a corresponding and unique characteristic voltage or current fluctuation is generated on the A-phase loop of the power grid and propagates downstream along the A-phase line, passing through each level of line, switch and meter box.

[0041] Preferably, the generated characteristic signal can be effectively distinguished from inherent noise, harmonics and normal power load fluctuations in the power grid, and the lower level troubleshooting instrument can reliably detect and analyze.

[0042] A2: Before step S100, scan the meter barcode through the field operation terminal to record the identification information of the electric energy meter in the meter box, wherein the field operation terminal can be a mobile phone or a special PDA; The field operation terminal automatically associates and binds the meter identification scanned and recorded with the line loss troubleshooting instrument currently connected in communication, forms a temporary device file and uploads it to the analysis platform; When the analysis platform detects or the operator discovers that the file association is abnormal through the field operation terminal, the field operation terminal starts the offline comparison mechanism, guides the operator to retrieve the secondary file data of the on-site meter, and manually checks and corrects on the terminal.

[0043] In use, the operator arrives at the scene, scans the bar code of each electric energy meter in the meter box using the field operation terminal, and the terminal APP automatically binds the scanned meter ID with the ID of the line loss troubleshooting instrument currently in communication, forms a temporary device file, and uploads it to the analysis platform.

[0044] The analysis platform uses this temporary device file to correspond the topology identification result (electrical connection relationship) with the specific meter (physical device), and realizes positioning to the meter box level.

[0045] When the system comparison finds that the electric quantity data is abnormal, or the manual perception finds that the positioning result is unreasonable, the offline comparison mechanism can be started, and the operator can directly retrieve the secondary file data (such as the meter address read directly through the RS485 port) from the meter local without networking, and compare and correct the file recorded by scanning.

[0046] Preferably, by automatic association through code scanning, the cumbersome process of traditional manual transcription and background input can be avoided, the time-consuming of file establishment is shortened by 90%, and artificial input errors are eliminated from the source to ensure the accuracy of line loss positioning to the level of the meter box.

[0047] In an alternative embodiment, the injection of characteristic signals into the target phase circuit in step S100 can also use the way of high-frequency pulse coding injection based on power electronic converter, that is, a small H-bridge power electronic converter is integrated in the troubleshooting instrument at the outlet switch of the transformer area, and a high-frequency pulse voltage signal with a frequency of 20-100 kHz and an amplitude of 1%-3% of the rated voltage is superimposed and injected into the target phase circuit through IGBT or MOSFET devices. Digital coding modulation technology (such as BPSK or FSK) is used to embed device address code, time stamp and other information into the high-frequency carrier to form a characteristic signal with specific code bits and waveform.

[0048] In another alternative embodiment, the injection of characteristic signals into the target phase circuit in step S100 can also use the way of current distortion injection based on silicon-controlled phase trigger, that is, a TRIAC module is built-in in the troubleshooting instrument at the outlet switch of the transformer area and is connected in series to the target phase circuit. By triggering the conduction accurately at a specific phase angle (such as-5° to +5°) before and after the voltage zero-crossing point, a transient current pulse lasting about 1-2 ms is generated, resulting in a recognizable distortion gap in the current waveform in that period. The time position and depth of the distortion gap constitute the characteristic code bits.

[0049] In the embodiments of the present application, the lower-level line loss troubleshooting instrument deployed at the meter box inlet receives and analyzes the characteristic signal to obtain the target data, and reports the target data and its device address information to the analysis platform in step S200, including the following steps B1-B2: B1: The lower-level line loss troubleshooting instrument deployed at the meter box inlet receives the characteristic signal and analyzes the characteristic signal to obtain its arrival time stamp, phase delay amount and intensity change amount.

[0050] It can be understood that the line loss troubleshooting instrument located downstream can accurately lock and confirm that the received signal is the characteristic signal injected by the upper level by continuously monitoring the target phase circuit and using its known specific signal as a template from the complex power grid background noise.

[0051] It should be noted that the arrival time stamp is used to reflect the absolute time when the signal is accurately identified, and is also the basis for calculating the phase delay amount; the phase delay amount a is used to reflect the phase delay experienced by the characteristic signal from the sending point to the receiving point, which is the key to judging the electrical performance and connection relationship of the line, and the intensity change amount m is used to reflect the amplitude attenuation of the characteristic signal from the sending point to the receiving point.

[0052] In an alternative embodiment, the resolving of the characteristic signal in step B1 to obtain its arrival time stamp, phase delay and intensity variation can also use a resolving method based on digital matching filtering and cross-correlation peak detection, i.e. the troubleshooting instrument continuously collects the A-phase voltage and current signals at a sampling rate of 100 kHz, and when the correlation peak value exceeds a set threshold (such as 0.7 times the template energy), the detection is triggered. The system time corresponding to the maximum value of the correlation peak is recorded as the arrival time stamp; the difference between the correlation peak phase and the initial phase of the template is calculated to obtain the phase delay a; the ratio of the correlation peak amplitude to the theoretical amplitude of the template is calculated to obtain the intensity variation m.

[0053] In another alternative embodiment, the resolving of the characteristic signal in step B1 to obtain its arrival time stamp, phase delay and intensity variation can also be a feature extraction method based on wavelet transform modulus maxima, i.e. using the troubleshooting instrument to perform 5-layer wavelet decomposition on the A-phase loop signal, detecting the modulus maxima point in the 3-4 layer detail coefficients, and determining the time corresponding to the point as the signal arrival time stamp; the instantaneous phase at this scale is obtained by Hilbert transform, and compared with the initial phase of the transmitting end to obtain the phase delay a; the attenuation gradient of the modulus maxima amplitude with the scale is calculated to back-propagate the intensity variation m in the signal transmission process.

[0054] B2: Report the time stamp, phase delay, intensity variation and local device address information to the analysis platform.

[0055] It should be noted that the lower-level line loss troubleshooter takes its device address as the source address of the data packet, and takes the time stamp, phase delay and intensity variation as the data payload, which are packaged together and sent to the analysis platform through the data forwarding module.

[0056] In the embodiments of the present application, the analysis platform in step S300 completes topology identification and generates a topology model of the substation branch based on the target data reported by each lower-level line loss troubleshooter and its device address information, including the following steps C1-C4: C1: The analysis platform sends a topology identification instruction to the line loss troubleshooter.

[0057] In an optional implementation, the analysis platform sending topology identification commands to the line loss detector in step C1 can also be achieved through an edge autonomous triggering mechanism. That is, instead of directly issuing topology identification commands in real time, the analysis platform pre-programs the topology identification strategy (such as triggering conditions, execution cycles, and signal encoding rules) into the local maintenance platform of each line loss detector using localized configuration tools (such as Bluetooth or infrared). The detector has a built-in edge computing module that monitors the load fluctuation rate, line loss rate change rate, or abnormal events (such as current surges or phase imbalances) in the distribution area in real time. When the monitored value exceeds a preset threshold, the topology identification process is automatically triggered. The detector at the outgoing switch actively injects a characteristic signal, while the detector at the meter box passively receives and parses it. After identification, the results are cached locally and reported in batches when communicating with the analysis platform.

[0058] In another optional implementation, the analysis platform can send the topology identification command to the line loss detector in step C1 via carrier broadcast + address filtering command propagation. Specifically, the analysis platform broadcasts a topology identification command frame to the power line bus via the HPLC high-speed carrier module of the transformer concentrator. The command frame is encoded using an address mask (e.g., the target address is a specific meter box address range). All detectors receive the carrier broadcast frame, but only the downstream detector with a matching address responds and performs the parsing operation. The outgoing switch detector, acting as a command relay node, first decodes and confirms itself as the signal source node after receiving the broadcast command, and then injects a characteristic signal into phase A.

[0059] C2: The line loss detector responds to the command by injecting a characteristic signal with specific characteristic code positions and waveforms generated by the switching on and off of a constant resistance load or a constant current load.

[0060] C3: After receiving the characteristic signal, the lower-level line loss detector analyzes and records the signal's arrival timestamp, phase delay, and signal strength attenuation, and reports the identification results, including the device address information.

[0061] C4: Based on the information reported by each node, the analysis platform analyzes the hierarchical connection relationship between devices and constructs a transformer branch topology model to describe line connections and support line loss prediction.

[0062] It should be noted that, on the one hand, the analysis platform automatically infers the parent-child hierarchical relationship between devices by analyzing the signal propagation path and timing based on the device addresses and timestamps / phase delays reported by each node. For example: if a signal arrives at node A first and then at node B, and the signal strength at node A is much greater than that at node B, then A can be determined to be the superior of B. On the other hand, by binding the identified hierarchical relationship with the signal strength attenuation (i.e., strength change) reported by each node and GPS location information, a comprehensive topology model is formed that includes connectivity, line loss characteristics, and spatial location.

[0063] Ideally, the constructed topology model not only makes the invisible power grid connections visible, manageable, and analyzable, but also serves as a direct basis for subsequent calculations to predict line loss rates and locate anomalies at the meter box level.

[0064] In this embodiment of the application, step S400 calculates the predicted line loss rate of the monitored area based on the topology model, and associates the prediction result with the physical location of the equipment by combining GPS positioning information, including the following steps D1-D2: D1: The analysis platform calculates the predicted line loss rate of each branch and the overall transformer area based on the phase delay and intensity change of each node in the topology model through a preset prediction model.

[0065] It should be noted that by manually sampling 100 data points (including manually detected line loss rate, intensity change m and phase delay a collected by the equipment), where line loss power = (input power - output power) / input power x 100%, then: like Figure 2 As shown, when a mathematical model is established for the relationship between the predicted line loss rate w and the intensity change m, then: (Formula 1); In the formula: k represents the empirical constant for adjusting the sensitivity of the model.

[0066] Furthermore, intensity features are constructed based on the actual line loss rate obtained from field surveys and the intensity changes obtained from analytical characteristic signals, such as... Figure 3 As shown, when a mathematical model is established for the relationship between the predicted line loss rate w and the phase delay a, then: (Formula 2); In the formula: k represents the empirical constant for adjusting the sensitivity of the model.

[0067] Furthermore, a topological model of the transformer substation branches is established based on the correlation between delay characteristics and intensity characteristics, such as... Figure 4 As shown, when a mathematical model is established to represent the relationship between the predicted line loss rate w and the intensity change m and phase delay a, it can be seen that the intensity change m and phase delay a are positively correlated. Combining the characteristic relationships of Formula 1 and Formula 2 above, the formula for calculating the predicted line loss rate is: In the formula: w represents the line loss rate predicted based on the characteristic signals collected by the line loss detector; a represents the phase delay amount extracted from the characteristic signals, with a value range of 0~90º; m represents the intensity change amount extracted from the characteristic signals.

[0068] For example, when the line loss detector at the meter box inlet receives a characteristic signal and the phase delay is a = 29º, and the intensity change is m = 12% derived from the characteristic signal, then: ; Based on the above calculations, the predicted line loss rate of the acquired feature signal is w=42.6%.

[0069] Further comparisons were made with multiple sets of data, and the results are shown in Table 1 below: Table 1. Comparison of Predicted Line Loss Rate and Actual Line Loss Rate Based on Partial Implementation Data According to the data in Table 1 above, the average error between the line loss rate w predicted by the topology model and the line loss rate detected manually is less than 5%. Therefore, the topology model can bring a better user experience, especially by greatly reducing the manual inspection of line loss, and at the same time, it can quickly locate the transformer area where line loss occurs.

[0070] In an optional implementation, the predicted line loss rate of each branch and the overall distribution area calculated in step S400 can also be obtained using a distribution area line loss prediction method based on a graph neural network (GNN). This involves abstracting the distribution area power grid topology into graph data, where nodes represent line loss detectors at various levels (total distribution area node, branch nodes, and meter box nodes), and edges represent connecting lines. The input feature vector of each node not only includes the phase delay 'a' and intensity change 'm' obtained from the parsed feature signals of that node, but also integrates multi-dimensional parameters such as the node's real-time load factor, historical line loss rate, line length, conductor cross-sectional area, and ambient temperature. The analysis platform constructs a graph convolutional neural network (GCN) or graph attention network (GAT) model, and by learning the feature propagation rules between nodes, outputs the predicted line loss rate of each node (i.e., each branch) and the overall distribution area line loss rate in one go.

[0071] In another optional implementation, the predicted line loss rate for each branch and the overall transformer area calculated in step S400 can also be obtained through a hierarchical recursive Bayesian probability model. Specifically, based on the generated topological hierarchy model (transformer area-branch-meter box), a probability distribution function for the line loss rate is established at each level. The prior distribution of the root node (overall transformer area) is set based on historical line loss statistics. Each branch node takes the distribution of its parent node as input, combines the phase delay 'a' and intensity change 'm' derived from the characteristic signals of its own node as observational evidence, and recursively updates the posterior probability distribution using Bayes' theorem, outputting the expected value and confidence interval of the branch's line loss rate. Finally, the predicted line loss rate for the overall transformer area is obtained by weighted summation of the expected values ​​of each branch, while the probability distribution of the overall line loss rate is calculated using Monte Carlo sampling.

[0072] D2: Match the device address information in the topology model with the pre-stored GPS positioning information, thereby mapping the predicted line loss rate and its corresponding abnormal warning information to the specific physical location of the meter box on the visualization interface.

[0073] It should be noted that, based on GPS coordinates, the location of each meter box can be displayed on the map. According to its predicted line loss rate and preset threshold (e.g., >10% is high loss, <-2% is negative loss), different colors (e.g., red for highlighting, yellow for warning) or icons are used to dynamically mark the meter boxes.

[0074] In this embodiment of the application, step S500 involves acquiring frozen data from household meters and frozen data from line loss detectors. Based on the principle of power balance, the current and power error values ​​at the outlet switch of the transformer area and the inlet of the meter box are compared step by step, including the following steps E1-E3: It should be noted that the step-by-step comparison of the current and energy error values ​​between the outgoing switches of the transformer area and the incoming lines of the meter box is performed using a localized architecture.

[0075] E1: The field operation terminal establishes a connection with the distribution area concentrator via Bluetooth communication and reads the frozen power and current data of the household meters on its behalf.

[0076] Specifically, when maintenance personnel arrive at the distribution area, they turn on the Bluetooth and APP of the field operation terminal (such as a dedicated PDA). The APP automatically scans and connects to the Bluetooth interface of the distribution area concentrator. The terminal sends a frozen data reading command to the concentrator through the APP. The concentrator executes the command and collects the frozen power, current and other data of all user meters at the specified time through its subordinate HPLC network, and sends the data packets to the field operation terminal via Bluetooth.

[0077] Ideally, by obtaining first-hand data directly from the concentrator, the risk of data being tampered with or forged during remote transmission is effectively eliminated. Furthermore, it is possible to complete data collection from hundreds of meters across the entire distribution area within minutes.

[0078] E2: The field operation terminal interacts with each line loss detector through the power line carrier channel to obtain the frozen electrical parameter data.

[0079] Specifically, the field operation terminal connects to the power line in the transformer area through an HPLC communication module. The terminal broadcasts data freeze and upload instructions to each line loss detector through a carrier channel. Each detector reports the three-phase voltage, current, power and other electrical parameter data frozen by its data freeze module at the time of the instruction to the field operation terminal through the carrier channel.

[0080] E3: The entire comparison process is completed within the local network, without relying on 4G mobile networks or external Internet servers.

[0081] Specifically, the APP of the on-site operation terminal completes all data parsing and comparison calculations locally and immediately generates comparison results (such as "XX branch power error exceeds the standard"). This result can be mutually verified with the prediction result of step S400 to ensure the accuracy of diagnosis. Finally, only non-sensitive results such as diagnostic conclusions (such as "XX meter box is suspected of electricity theft") can be selectively synchronized to the superior information system, while the original detailed power consumption data is kept locally.

[0082] In this embodiment of the application, step S600 triggers an anomaly warning by analyzing and predicting the changing trend of the line loss rate, and visually marks high loss or negative loss intervals in the topology map generated by the topology model, including the following steps F1-F3: F1: The analysis platform compares the predicted line loss rate with the preset threshold in real time. When the predicted line loss rate continuously exceeds the high loss threshold or falls below the negative loss threshold, it automatically triggers a minute-level anomaly warning.

[0083] For example, preset thresholds such as >10% are considered high loss and <-2% are considered negative loss.

[0084] F2: In the visualized topology map generated based on the topology model, different colors or icons are used to dynamically render and mark the corresponding branches and meter boxes according to the predicted line loss rate and severity.

[0085] F3: Automatically deletes the temporary line loss detector file created for this diagnosis in the concentrator after the diagnostic process is completed.

[0086] In summary, the beneficial effects of this invention's rapid diagnosis method for high line loss and electricity theft are that it transforms line loss location from a haphazard, blanket-style search to precise guidance. By sending characteristic signals to the target phase circuit and then analyzing the phase delay and attenuation of the returned signal, the location of electricity theft or leakage can be directly pinpointed to the specific meter box with an error of less than 5%, enabling minute-level anomaly warnings. Most importantly, the entire process does not rely on mobile phone signals; even in remote mountainous areas, data collection can be achieved using Bluetooth and power line carrier, saving over 30% on communication costs and enhancing data security. Operation is also convenient; the meter and device are automatically bound via QR code scanning, and junk data is automatically cleaned up after use, reducing record-keeping time from hours to minutes.

[0087] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A rapid diagnostic device for high line loss and electricity theft, characterized in that: include, Line loss detectors are deployed at the outgoing switches and meter box inlets of the power grid in the distribution area to collect real-time three-phase electrical parameter data. The analysis platform is communicatively connected to the line loss detector, and is used to receive data uploaded by the line loss detector and perform line loss calculation and location analysis.

2. The rapid diagnostic device for high line loss and electricity theft as described in claim 1, characterized in that: The line loss detector includes: The topology identification module is used to inject characteristic signals into the power grid loop and to receive and parse characteristic signals from lower-level nodes to obtain the timestamp, phase delay and strength information of the signals. The data freeze module is used to freeze electrical parameter data according to instructions; The data forwarding module is used to forward instructions and data through the target protocol.

3. The rapid diagnostic device for high line loss and electricity theft as described in claim 2, characterized in that: The analysis platform includes, The topology model construction module is used to construct a topology model of the transformer substation branch based on the identification results of the topology identification module and the device address information, combined with GPS positioning data. The topology model represents the line connection relationship and supports line loss prediction. The line loss calculation and early warning module is used to obtain the predicted line loss rate of the monitored area based on the topology model, associate the physical location of the equipment, and trigger an early warning and visual marker when the predicted line loss rate is abnormal.

4. The rapid diagnostic device for high line loss and electricity theft as described in claim 3, characterized in that: The line loss detector also includes a local maintenance module; The local maintenance module is used to store and manage the electricity meter files in the metering box, and supports automatic acquisition of meter barcode information to associate the line loss detector with the household meter.

5. A rapid diagnostic method for high line loss and electricity theft, characterized in that: include, The line loss detector deployed at the outgoing switch of the transformer area injects a characteristic signal into the target phase circuit; The lower-level line loss detector deployed at the meter box inlet receives and parses the characteristic signal to obtain target data, and reports the target data and its device address information to the analysis platform. The target data includes the signal timestamp, phase delay, and intensity change. The analysis platform, based on the target data and equipment address information reported by each lower-level line loss detector, completes topology identification and generates a topology model of the transformer substation branch. Based on the aforementioned topology model, the predicted line loss rate of the monitored area is calculated, and the prediction results are correlated with the physical location of the equipment by combining GPS positioning information. Obtain frozen data from household meters and frozen data from line loss detectors, and based on the principle of power balance, compare the current and power error values ​​at the outlet switch of the transformer area and the inlet of the meter box step by step; Anomaly warnings are triggered by analyzing the changing trend of the predicted line loss rate, and high-loss or negative-loss intervals are visually marked in the topology map generated by the topology model.

6. The method for rapid diagnosis of high line loss and electricity theft as described in claim 5, characterized in that: The topology model for generating branch lines in the transformer substation includes, The analysis platform sends a topology identification command to the line loss detector; The line loss detector responds to the instruction by injecting a characteristic signal with specific characteristic code points and waveforms generated by the switching on and off of a constant resistance load or a constant current load. After receiving the characteristic signal, the lower-level line loss detector analyzes and records the signal's arrival timestamp, phase delay, and signal strength attenuation, and reports the identification results, which include device address information. Based on the information reported by each node, the analysis platform analyzes the hierarchical connection relationship between devices and constructs a transformer substation branch topology model to describe line connections and support line loss prediction.

7. The method for rapid diagnosis of high line loss and electricity theft as described in claim 5, characterized in that: The construction of the topology model depends on the physical characteristics of the characteristic signals during transmission; Establish a quantization relationship between the predicted line loss rate and signal parameters. include, The actual line loss rate of multiple sample areas was obtained based on field surveys, and the phase delay and intensity change were recorded simultaneously through the analysis of characteristic signals. Using the phase delay and intensity change as independent variables and the actual line loss rate as the dependent variable, a prediction model is constructed by fitting the data. The prediction model is used to calculate the predicted line loss rate based on the phase delay and intensity change data collected in real time.

8. The method for rapid diagnosis of high line loss and electricity theft as described in claim 5, characterized in that: The step-by-step comparison of the current and energy error values ​​between the outlet switch of the transformer area and the inlet of the meter box is performed using a localized architecture.

9. A rapid diagnostic method for high line loss and electricity theft as described in any one of claims 5-8, characterized in that: The method also includes, Scan the meter barcode using the on-site operation terminal and enter the identification information of the electricity meter in the meter box; The on-site operation terminal automatically associates and binds the meter identifier scanned and entered with the line loss detector currently connected to the communication, forming a temporary equipment file and uploading it to the analysis platform; When the analysis platform detects or the operator discovers an anomaly in the file association through the on-site operation terminal, the on-site operation terminal initiates an offline comparison mechanism, guiding the operator to retrieve the secondary file data of the on-site electricity meter and manually verify and correct it on the terminal.

10. A rapid diagnostic method for high line loss and electricity theft as described in any one of claims 5-8, characterized in that: The method also includes automatically deleting the temporary line loss detector file created for this diagnosis in the concentrator after the diagnosis process is completed.

Images