Integrated diagnosis system for ground fault line selection and traveling wave distance measurement based on distribution network

Abstract

The application discloses a grounding fault line selection and travelling wave distance measurement integrated diagnosis system based on a power distribution network, which comprises a data acquisition module, a data processing module, an integrated fault diagnosis module, a communication transmission module and a display and early warning module, and relates to the technical field of power distribution network fault diagnosis. The integrated diagnosis system effectively integrates the functions of grounding fault line selection and travelling wave distance measurement, collects the electrical parameters of the power distribution network in real time through the data acquisition module, analyzes and processes the electrical parameters through the data processing module, accurately determines the position of the fault line and the fault point through the integrated fault diagnosis module, adopts a multi-criterion fusion line selection algorithm and an accurate travelling wave distance measurement algorithm, improves the accuracy and reliability of the grounding fault diagnosis, extracts the core features such as the amplitude, direction and mutation time of the travelling wave head through a mobile recognition window, effectively eliminates noise interference, and improves the recognition accuracy of the wave head.

Description

Technical Field

[0001] This invention relates to the field of power distribution network fault diagnosis technology, specifically to an integrated diagnostic system for ground fault location and traveling wave ranging in power distribution networks. Background Technology

[0002] Grounding faults are a common type of fault in power distribution networks. Timely and accurate detection and location of grounding faults are crucial for ensuring the safe and stable operation of the power distribution network. Currently, the handling of grounding faults in power distribution networks mainly involves two aspects: fault location and fault distance measurement.

[0003] The reference patent, titled "An Integrated Verification Method and System for Intelligent Fault Recording and Traveling Wave Ranging" (Patent Publication No.: CN120275762A, Patent Publication Date: 2025-07-08), includes: installing equipment on both sides of the faulty line to monitor the initial traveling wave reaching the two monitoring points and perform double-end traveling wave ranging; recording the arrival time of the initial traveling wave generated by the fault point between the two monitoring points along the transmission line to both ends, and the distance from the fault point to the monitoring points at both ends; calculating the single-end traveling wave ranging equation based on traveling wave ranging theory; and using a comprehensive ranging algorithm combined with the single-end traveling wave ranging equation to complete the integrated control of intelligent fault recording and traveling wave ranging, integrating the two functions into one device, effectively reducing equipment investment and maintenance costs, solving the technical problem of synchronous acquisition of power frequency recording data and high frequency traveling wave data, ensuring the real-time performance and accuracy of the data, and improving the reliability of fault diagnosis.

[0004] Based on the description in the above documents, the following problems still exist in the current complex distribution network diagnostic operations:

[0005] 1. Existing distribution network grounding fault selection technology relies on a single criterion or simple parameter comparison, without fully considering the complex topology of the distribution network, the superposition of multiple fault types, and the impact of interference signals;

[0006] 2. Existing traveling wave ranging technology suffers from difficulties in signal processing, low time synchronization accuracy, and large distance calculation errors;

[0007] 3. Existing waveform recording functions simply record electrical parameter waveforms before and after a fault, without optimizing data quality. Therefore, this invention provides an integrated diagnostic system for ground fault location and traveling wave ranging based on power distribution networks. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention provides an integrated diagnostic system for ground fault location and traveling wave ranging based on power distribution networks. This system solves the problems of low accuracy in fault location, frequent misjudgments and omissions, poor reliability of traveling wave ranging, insufficient positioning accuracy, and insufficient precision of waveform recording signals in complex power distribution network diagnostic operations.

[0009] To achieve the above objectives, the present invention provides the following technical solution: an integrated diagnostic system for ground fault location and traveling wave ranging based on a power distribution network, comprising:

[0010] The data acquisition module includes multiple sensors installed at different locations in the power distribution network, used to collect electrical parameter data of the power distribution network in real time;

[0011] The data processing module preprocesses the electrical parameters transmitted by the data acquisition module to remove noise from the signal and convert the data.

[0012] The integrated fault diagnosis module determines whether electrical parameters exceed the preset fault initiation threshold and triggers the fault diagnosis process and waveform recording function after a fault occurs. It uses a moving recognition window to extract the amplitude, direction and abrupt change of the traveling wave front. Based on the extracted traveling wave characteristic parameters, it compares the differences in amplitude and direction of the traveling waves of each outgoing line. Based on the line selection results, it locates the time difference between the initial traveling wave front of the faulty line and the reflected wave front of the fault point. It also calculates the distance to the fault point by combining the line wave velocity parameters.

[0013] The communication transmission module is responsible for transmitting and exchanging data between the data acquisition module, the data processing module, and the integrated fault diagnosis module;

[0014] The display and early warning module shows the operating status and fault information of the power distribution network in charts and text, and notifies the operators to handle the fault with audible and visual alarm signals and messages after a fault is detected.

[0015] Preferably, the sensors in the data acquisition module include:

[0016] Voltage sensors are used to collect voltage values ​​from each line;

[0017] Current sensors are used to collect the current values ​​of each line. A combined current acquisition device is installed at each outgoing line end to collect the A, B, and C three-phase current signals and the zero-sequence current signal of each outgoing line.

[0018] Distributed traveling wave ranging terminals are used to capture traveling wave signal data generated at periodic time nodes before and after a fault occurs.

[0019] Preferably, the noise removal and data conversion operations in the data processing module are as follows:

[0020] Based on the characteristics of different electrical parameters and the noise frequency range, one or more filtering methods such as low-pass filtering, high-pass filtering, and band-pass filtering are selected for processing.

[0021] Integrated high-speed A / D conversion module and GPS / BeiDou dual-mode synchronization module;

[0022] The high-speed A / D conversion module converts standard voltage signals into digital signals. The GPS / BeiDou dual-mode synchronization module receives satellite second pulses and B-code time signals, and compensates for time deviations through a Kalman filter algorithm. This allows the data from the bus and each outgoing line to be collected simultaneously, and the location of the collection point is determined and each digital signal is marked with an absolute time stamp.

[0023] Preferably, the operation in the integrated fault diagnosis module that determines whether electrical parameters exceed a preset fault initiation threshold is as follows:

[0024] Extract the zero-sequence current and zero-sequence voltage values ​​obtained at the corresponding time when multiple faults occurred in the recent historical data. Remove the maximum and minimum values ​​of the two values ​​respectively, and calculate the average of the remaining values ​​to obtain the fault initiation threshold for the zero-sequence current and zero-sequence voltage values. When the real-time collected values ​​meet either of the corresponding values, the fault diagnosis process is triggered.

[0025] The electrical parameters are collected and processed in real time and then compared with the fault initiation threshold in real time. When the electrical parameters exceed the fault initiation threshold, it is determined that a fault may occur, triggering the fault diagnosis process. At the same time, the waveform recording function is started, which records the waveform of the electrical parameters for a period of time before and after the fault occurs.

[0026] Preferably, the operation of extracting the amplitude, direction, and abrupt change time of the traveling wave front using a moving recognition window in the integrated fault diagnosis module is as follows:

[0027] Extract the electrical parameter data obtained after starting the waveform recording function, establish a traveling wave change coordinate axis with time node as the horizontal axis category and signal value as the vertical axis category, and input the extracted electrical parameter data into the traveling wave change coordinate axis in time order to generate change curves;

[0028] The motion recognition window is set as F(x, y), where x is the horizontal length of the motion recognition window and y is the vertical width of the motion recognition window. The motion recognition window moves from the starting point of the changing curve, and the changing curve is segmented based on the curve segment extracted by the motion recognition window. The change situation is determined by magnifying and analyzing the curve segment to obtain the moment of the wavefront change.

[0029] The amplitude and direction of the traveling wavefront are derived from the analysis of the abrupt change in the wavefront.

[0030] Preferably, the operation of determining the moment of wavefront abrupt change through magnified analysis of the curve segment is as follows:

[0031] Extract curve segments sequentially, introduce the changes of multiple undisplayed node data into the curve segments between the current time nodes to determine whether there are extreme points in the curve segments that simultaneously have rising and falling changes;

[0032] If an extreme point exists, a noise threshold is set, and the amplitude corresponding to the extreme point is compared with three times the noise threshold.

[0033] If the amplitude corresponding to the extreme point is greater than three times the noise threshold, and it appears for the first time in the current order of curve segment extraction, then the current extreme point is the moment of wavefront abrupt change.

[0034] Preferably, the operation of deriving the amplitude and direction of the traveling wavefront based on the analysis of the abrupt change in the wavefront is as follows:

[0035] Centered on the identified moment of wavefront change, the signal amplitudes of multiple sampling points before and after the point are taken, and the peak amplitude within the current sampling interval is calculated, which is the amplitude of the traveling wavefront.

[0036] The direction of the wavefront is then determined by the change in signal amplitude after the abrupt change in the wavefront. If the signal amplitude increases, the direction of the wavefront is positive; otherwise, if the signal amplitude decreases, the direction of the wavefront is negative. A positive wavefront corresponds to a fault point that is far from the monitoring point, while a negative wavefront corresponds to a fault point that is close to the monitoring point.

[0037] Preferably, the operation in the integrated fault diagnosis module that compares the differences in the magnitude and direction of the traveling wave amplitude of each outgoing line is as follows:

[0038] A convolutional neural network classification model is generated by training a large amount of sample data of different fault types, and the extracted feature parameters are input into the trained convolutional neural network classification model to output a fault probability value for each outgoing line.

[0039] Compare the waveform amplitudes of all outgoing lines together, and then check the waveform direction of each outgoing line. The outgoing line with the highest fault probability value has the largest waveform amplitude and the waveform direction is different from other outgoing lines. That is, the outgoing line with the highest fault probability value is the faulty line.

[0040] Preferably, the operation of calculating the distance to the fault point in the integrated fault diagnosis module by combining the line wave velocity parameters is as follows:

[0041] Extract the waveform curve on the faulty line and determine the first wavefront that propagates directly from the fault point to the monitoring point after the fault occurs. This is the initial traveling wavefront. The second characteristic wavefront that is reflected back to the monitoring point after the initial traveling wave reaches the fault point is the fault point reflected wavefront.

[0042] The corresponding time node t1 is obtained from the amplitude change point of the initial traveling wave wavefront, and the corresponding time node t2 is obtained from the amplitude change point of the reflected wave wavefront at the fault point. The time difference ΔT = t2 - t1 is obtained.

[0043] Based on historical data, the reaction time of the monitoring point when receiving the returned traveling wave signal and obtaining the signal parameters is t3. The reaction time t3 is compensated into the time difference, and the distance from the fault point to the monitoring point is determined as L by combining the time and wave velocity parameters.

[0044] Preferably, the formula for calculating the distance L from the fault point to the monitoring point is:

[0045] L = [v × (△T - t3)] / 2;

[0046] Where v is the traveling wave velocity, which is calculated by back-calculating the arrival time difference of the traveling wave at two monitoring points with a known distance.

[0047] This invention provides an integrated diagnostic system for ground fault location and traveling wave ranging in power distribution networks. Compared with existing technologies, it has the following advantages:

[0048] 1. This integrated diagnostic system for ground fault location and traveling wave ranging based on power distribution networks effectively integrates the functions of ground fault location and traveling wave ranging. The data acquisition module collects electrical parameters of the power distribution network in real time. After analysis and processing by the data processing module, the integrated fault diagnosis module accurately determines the location of the faulty line and fault point. The system adopts a multi-criteria fusion algorithm for line selection and a precise traveling wave ranging algorithm, which improves the accuracy and reliability of ground fault diagnosis. The system extracts core features such as the amplitude, direction, and abrupt change time of the traveling wave front by moving the identification window, effectively eliminating noise interference and improving the accuracy of wave front identification.

[0049] 2. This integrated diagnostic system for ground fault location and traveling wave ranging based on distribution networks extracts zero-sequence current and zero-sequence voltage values ​​from historical fault data, removes extreme values, and takes the average value as the threshold. This takes into account the individual differences in power grid operation and avoids interference from extreme data, ensuring the accuracy of fault initiation judgment and reducing false triggering and missed triggering. The faulty line is confirmed by the dual criteria of maximum amplitude and direction different from other outgoing lines. Multi-dimensional verification ensures the reliability of the location results and is suitable for complex distribution network scenarios with multiple outgoing lines.

[0050] 3. This integrated diagnostic system for ground fault location and traveling wave ranging based on power distribution network integrates GPS and Beidou dual-mode synchronization modules, compensates for time deviations through Kalman filtering algorithm, assigns absolute time stamps to all collected data, ensures time alignment of data from multiple collection points, and introduces monitoring point response time t3 compensation in distance calculation to correct time difference ΔT, significantly reducing ranging error and making fault location more accurate.

[0051] 4. This integrated diagnostic system for ground fault location and traveling wave ranging based on the power distribution network uses a waveform recording function that is triggered only after the fault is initiated to record waveform data during key time periods before and after the fault, reducing redundant data. At the same time, combined with a wavefront recognition mechanism, it automatically marks key feature points such as wavefront abrupt changes, providing high-quality and high-value data support for subsequent diagnosis. Moreover, after identifying the faulty line, the integrated fault diagnosis module immediately starts the ranging process based on the traveling wave data of that line, without the need for cross-module data interaction and waiting, which greatly shortens the fault diagnosis cycle and improves the fault handling response speed. Attached Figure Description

[0052] Figure 1 This is a schematic diagram of the integrated diagnostic system of the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] Please see Figure 1 The present invention provides a technical solution:

[0055] An integrated diagnostic system for ground fault location and traveling wave ranging based on distribution networks includes:

[0056] The data acquisition module includes multiple sensors installed at different locations in the power distribution network, used to collect electrical parameter data of the power distribution network in real time;

[0057] The data processing module preprocesses the electrical parameters transmitted by the data acquisition module to remove noise from the signal and convert the data.

[0058] The integrated fault diagnosis module determines whether electrical parameters exceed the preset fault initiation threshold and triggers the fault diagnosis process and waveform recording function after a fault occurs. It uses a moving recognition window to extract the amplitude, direction and abrupt change of the traveling wave front. Based on the extracted traveling wave characteristic parameters, it compares the differences in amplitude and direction of the traveling waves of each outgoing line. Based on the line selection results, it locates the time difference between the initial traveling wave front of the faulty line and the reflected wave front of the fault point. It also calculates the distance to the fault point by combining the line wave velocity parameters.

[0059] The communication transmission module is responsible for transmitting and exchanging data between the data acquisition module, the data processing module, and the integrated fault diagnosis module;

[0060] The display and early warning module shows the operating status and fault information of the power distribution network in charts and text, and notifies the operators to handle the fault with audible and visual alarm signals and messages after a fault is detected.

[0061] The communication transmission module adopts a dual-backup communication method, utilizes the existing optical fiber communication network of the distribution network, and uses the Ethernet protocol (TCP / IP) to transmit data. It is suitable for data transmission between modules on the bus side and within the substation. The distributed traveling wave ranging terminal and the data processing module use 5G / 4G wireless communication, which supports data upload from edge computing nodes. At the same time, it is equipped with a LoRa backup communication link to deal with scenarios with poor 5G / 4G signal coverage.

[0062] The display and early warning module supports functions such as real-time data curve display, fault information list display, historical data query, fault location map annotation, and can export fault reports.

[0063] By effectively integrating ground fault location and traveling wave ranging functions, the electrical parameters of the distribution network are collected in real time through the data acquisition module. After analysis and processing by the data processing module, the integrated fault diagnosis module accurately determines the location of the faulty line and fault point. The use of a multi-criteria fusion line selection algorithm and a precise traveling wave ranging algorithm improves the accuracy and reliability of ground fault diagnosis. By extracting core features such as traveling wavefront amplitude, direction, and abrupt change time through a moving recognition window, noise interference is effectively eliminated, and the wavefront recognition accuracy is improved.

[0064] In this embodiment of the invention, the sensor in the data acquisition module includes:

[0065] Voltage sensors are used to collect voltage values ​​of each line. They are installed on the distribution network bus and each outgoing switch side. Each outgoing line corresponds to one set of voltage sensors to collect line voltage and phase voltage data. The sampling frequency is set to 20kHz to ensure coverage of the high-frequency components of the traveling wave signal.

[0066] Current sensors are used to collect the current values ​​of each line. A combined current acquisition device is installed at each outgoing line end to collect the A, B, and C three-phase current signals and the zero-sequence current signal of each outgoing line.

[0067] Distributed traveling wave ranging terminals are used to capture traveling wave signal data generated at periodic time nodes before and after a fault occurs. They are deployed in a distributed manner along the power distribution network lines, with the terminals installed at line towers or cable joints. They are waterproof, dustproof, and resistant to electromagnetic interference.

[0068] Among them, the voltage sensor is a capacitive voltage divider type voltage sensor, the current sensor is a Rogowski coil current sensor, and the distributed traveling wave ranging terminal is equipped with a high-performance FPGA chip, a built-in high-speed data cache unit, supports local temporary storage of fault waveform data, and has the function of resuming interrupted transmission.

[0069] In this embodiment of the invention, the operations of noise removal and data conversion in the data processing module are as follows:

[0070] Based on the characteristics of different electrical parameters and the noise frequency range, one or more filtering methods such as low-pass filtering, high-pass filtering, and band-pass filtering are selected for processing.

[0071] First, the collected voltage and current signals are subjected to spectrum analysis to determine the noise frequency range. Low-pass filtering is used for power frequency interference and low-frequency noise; high-pass filtering is used for high-frequency noise; and band-pass filtering is used for interference in a specific frequency range (such as communication interference). After filtering, the signal is smoothed and a moving average algorithm is used to further reduce the impact of residual noise.

[0072] It integrates a high-speed A / D conversion module and a GPS / BeiDou dual-mode synchronization module to ensure high-precision conversion of analog signals to digital signals. It selects a module that supports BDS-3 and GPS L1 / L2 dual-frequency reception and receives satellite signals through an external high-gain antenna to ensure accurate time synchronization even in complex terrain.

[0073] The high-speed A / D conversion module converts standard voltage signals into digital signals. The GPS / BeiDou dual-mode synchronization module receives satellite second pulses and B-code time signals, and compensates for time deviations through a Kalman filter algorithm. This allows the data from the bus and each outgoing line to be collected simultaneously, and the location of the collection point is determined and each digital signal is marked with an absolute time stamp.

[0074] In this embodiment of the invention, the operation of determining whether electrical parameters exceed a preset fault initiation threshold in the integrated fault diagnosis module is as follows:

[0075] Extract the zero-sequence current and zero-sequence voltage values ​​obtained at the corresponding time when multiple faults occurred in the recent historical data. Remove the maximum and minimum values ​​of the two values ​​respectively, and calculate the average of the remaining values ​​to obtain the fault initiation threshold for the zero-sequence current and zero-sequence voltage values. When the real-time collected values ​​meet either of the corresponding values, the fault diagnosis process is triggered.

[0076] The electrical parameters are collected and processed in real time and then compared with the fault initiation threshold in real time. When the electrical parameters exceed the fault initiation threshold, it is determined that a fault may occur, triggering the fault diagnosis process. At the same time, the waveform recording function is started, which records the waveform of the electrical parameters for a period of time before and after the fault occurs.

[0077] By extracting zero-sequence current and zero-sequence voltage values ​​from historical fault data, removing extreme values ​​and taking the average as the threshold, the system considers both individual differences in power grid operation and avoids interference from extreme data, ensuring the accuracy of fault initiation judgment and reducing false triggering and missed triggering. The faulty line is confirmed by the dual criteria of maximum amplitude and direction different from other outgoing lines. Multi-dimensional verification ensures the reliability of the line selection results and is suitable for complex distribution network scenarios with multiple outgoing lines.

[0078] In this embodiment of the invention, the operation of extracting the amplitude, direction, and abrupt change time of the traveling wave front using a moving recognition window in the integrated fault diagnosis module is as follows:

[0079] Extract the electrical parameter data obtained after starting the waveform recording function, establish a traveling wave change coordinate axis with time node as the horizontal axis category and signal value as the vertical axis category, and input the extracted electrical parameter data into the traveling wave change coordinate axis in time order to generate change curves;

[0080] The motion recognition window is set as F(x, y), where x is the horizontal length of the motion recognition window and y is the vertical width of the motion recognition window. The motion recognition window moves from the starting point of the changing curve, and the changing curve is segmented based on the curve segment extracted by the motion recognition window. The change situation is determined by magnifying and analyzing the curve segment to obtain the moment of the wavefront change.

[0081] The amplitude and direction of the traveling wavefront are derived from the analysis of the abrupt change in the wavefront.

[0082] Extract the traveling wave signal (voltage or current traveling wave, with priority given to voltage traveling waves with higher signal-to-noise ratio) from the recorded waveform data. Establish a traveling wave variation coordinate axis with time as the horizontal axis and signal amplitude as the vertical axis. Input the recorded waveform data into the coordinate axis in chronological order to generate a continuous traveling wave variation curve.

[0083] The moving recognition window starts from the starting point of the changing curve and moves along the horizontal axis with the window length as the step and along the vertical axis with the window width as the step. After each movement, the curve segment within the window is extracted to achieve uniform segmentation of the entire traveling wave curve. For each extracted curve segment, 5 undisplayed node data (calculated using a linear interpolation algorithm) are introduced to enrich the curve details and analyze whether there are extreme points of rise and fall in the curve segment.

[0084] In this embodiment of the invention, the operation of determining the change and deriving the moment of wavefront abrupt change through magnified analysis of the curve segment is as follows:

[0085] Extract curve segments sequentially, introduce the changes of multiple undisplayed node data into the curve segments between the current time nodes to determine whether there are extreme points in the curve segments that simultaneously have rising and falling changes;

[0086] If an extreme point exists, a noise threshold is set, and the amplitude corresponding to the extreme point is compared with three times the noise threshold.

[0087] If the amplitude corresponding to the extreme point is greater than three times the noise threshold, and it appears for the first time in the current order of curve segment extraction, then the current extreme point is the moment of wavefront abrupt change.

[0088] The waveform recording function is triggered only after a fault is initiated, recording waveform data for key time periods before and after the fault, reducing redundant data. At the same time, combined with the wavefront recognition mechanism, key feature points such as wavefront abrupt changes are automatically marked, providing high-quality and high-value data support for subsequent diagnosis. Moreover, after identifying the faulty line, the integrated fault diagnosis module immediately starts the ranging process based on the traveling wave data of that line, without the need for cross-module data interaction and waiting, which greatly shortens the fault diagnosis cycle and improves the fault handling response speed.

[0089] In this embodiment of the invention, the operation of determining the amplitude and direction of the traveling wavefront based on the moment of abrupt change in the wavefront is as follows:

[0090] Centered on the identified moment of wavefront change, the signal amplitudes of multiple sampling points before and after the point are taken, and the peak amplitude within the current sampling interval is calculated, which is the amplitude of the traveling wavefront.

[0091] The direction of the wavefront is then determined by the change in signal amplitude after the abrupt change in the wavefront. If the signal amplitude increases, the direction of the wavefront is positive; otherwise, if the signal amplitude decreases, the direction of the wavefront is negative. A positive wavefront corresponds to a fault point that is far from the monitoring point, while a negative wavefront corresponds to a fault point that is close to the monitoring point.

[0092] In this embodiment of the invention, the operation of comparing the differences in the magnitude and direction of the traveling wave amplitude of each outgoing line in the integrated fault diagnosis module is as follows:

[0093] A convolutional neural network classification model is generated by training a large amount of sample data of different fault types, and the extracted feature parameters are input into the trained convolutional neural network classification model to output a fault probability value for each outgoing line.

[0094] Compare the waveform amplitudes of all outgoing lines together, and then check the waveform direction of each outgoing line. The outgoing line with the highest fault probability value has the largest waveform amplitude and the waveform direction is different from other outgoing lines. That is, the outgoing line with the highest fault probability value is the faulty line.

[0095] The process involves collecting sample data (at least 1000 sets) of different fault types (such as single-phase grounding, two-phase grounding, and three-phase grounding), different fault locations, and different fault resistances. Each set of samples includes characteristic parameters such as the amplitude, direction, and abrupt change time of the traveling wave front. The sample data is divided into a training set and a test set in a 7:3 ratio. The training set is used to train the convolutional neural network model (the network structure includes an input layer, two convolutional layers, one pooling layer, a fully connected layer, and an output layer). The test set is used to verify the model performance. Training is stopped when the model accuracy is ≥95%, and the model parameters are saved.

[0096] In this embodiment of the invention, the operation of calculating the distance to the fault point in the integrated fault diagnosis module by combining the line wave velocity parameters is as follows:

[0097] Extract the waveform curve on the faulty line and determine the first wavefront that propagates directly from the fault point to the monitoring point after the fault occurs. This is the initial traveling wavefront. The second characteristic wavefront that is reflected back to the monitoring point after the initial traveling wave reaches the fault point is the fault point reflected wavefront.

[0098] The corresponding time node t1 is obtained from the amplitude change point of the initial traveling wave wavefront, and the corresponding time node t2 is obtained from the amplitude change point of the reflected wave wavefront at the fault point. The time difference ΔT = t2 - t1 is obtained.

[0099] Based on historical data, the reaction time of the monitoring point when receiving the returned traveling wave signal and obtaining the signal parameters is t3. The reaction time t3 is compensated into the time difference, and the distance from the fault point to the monitoring point is determined as L by combining the time and wave velocity parameters.

[0100] Based on historical data, the average response time of the monitoring points after receiving the traveling wave signal is statistically analyzed to complete signal acquisition, conversion, and feature extraction.

[0101] By integrating GPS and BeiDou dual-mode synchronization modules and using Kalman filtering algorithm to compensate for time deviations, all collected data are given an absolute time stamp, ensuring time alignment of data from multiple collection points. At the same time, the monitoring point reaction time t3 is compensated in the distance calculation to correct the time difference ΔT, which greatly reduces the ranging error and makes the fault location more accurate.

[0102] In this embodiment of the invention, the formula for calculating the distance L from the fault point to the monitoring point is:

[0103] L = [v × (△T - t3)] / 2;

[0104] Where v is the traveling wave velocity, which is calculated by back-calculating the arrival time difference of the traveling wave at two monitoring points with a known distance.

[0105] Furthermore, after the system is powered on, each module automatically performs a self-test (including sensors, communication links, storage units, etc.). The self-test results are displayed through the warning module. If a fault is found (such as sensor communication interruption or abnormal synchronization module signal), an audible and visual alarm is immediately triggered, indicating the fault location. After the self-test passes, the data processing module automatically loads historical fault data and calculates the fault initiation threshold; the communication transmission module establishes a connection with each acquisition unit and synchronizes the clock signal; the warning module initializes its interface and begins to receive and display power distribution network operation data in real time.

[0106] Once the system detects a fault and completes fault line selection and distance measurement, the display and early warning module immediately pops up a fault prompt window, showing information such as the fault line number, distance to the fault point, fault occurrence time, wavefront amplitude and direction. At the same time, the audible and visual alarm device is activated. Operators can view the traveling wave waveform before and after the fault, the fault location map marking, export the fault report, and arrange for maintenance personnel to go to the fault point for handling. After the fault is handled, the operator enters the handling results into the system, and the system automatically updates the historical fault data and recalculates the fault initiation threshold.

[0107] Furthermore, any content not described in detail in this specification is existing technology known to those skilled in the art.

[0108] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An integrated diagnostic system for ground fault location and traveling wave ranging in power distribution networks, characterized in that: include: The data acquisition module includes multiple sensors installed at different locations in the power distribution network, used to collect electrical parameter data of the power distribution network in real time; The data processing module preprocesses the electrical parameters transmitted by the data acquisition module to remove noise from the signal and convert the data. The integrated fault diagnosis module determines whether electrical parameters exceed the preset fault initiation threshold and triggers the fault diagnosis process and waveform recording function after a fault occurs. It uses a moving recognition window to extract the amplitude, direction and abrupt change of the traveling wave front. Based on the extracted traveling wave characteristic parameters, it compares the differences in amplitude and direction of the traveling waves of each outgoing line. Based on the line selection results, it locates the time difference between the initial traveling wave front of the faulty line and the reflected wave front of the fault point. It also calculates the distance to the fault point by combining the line wave velocity parameters. The communication transmission module is responsible for transmitting and exchanging data between the data acquisition module, the data processing module, and the integrated fault diagnosis module; The display and early warning module shows the operating status and fault information of the power distribution network in charts and text, and notifies the operators to handle the fault with audible and visual alarm signals and messages after a fault is detected.

2. The integrated diagnostic system for ground fault location and traveling wave ranging based on distribution network as described in claim 1, characterized in that: The sensors in the data acquisition module include: Voltage sensors are used to collect voltage values ​​from each line; Current sensors are used to collect the current values ​​of each line. A combined current acquisition device is installed at each outgoing line end to collect the A, B, and C three-phase current signals and the zero-sequence current signal of each outgoing line. Distributed traveling wave ranging terminals are used to capture traveling wave signal data generated at periodic time nodes before and after a fault occurs.

3. The integrated diagnostic system for ground fault location and traveling wave ranging based on distribution network as described in claim 1, characterized in that: The data processing module performs noise removal and data conversion operations in the signal as follows: Based on the characteristics of different electrical parameters and the noise frequency range, one or more filtering methods such as low-pass filtering, high-pass filtering, and band-pass filtering are selected for processing. Integrated high-speed A / D conversion module and GPS / BeiDou dual-mode synchronization module; The high-speed A / D conversion module converts standard voltage signals into digital signals. The GPS / BeiDou dual-mode synchronization module receives satellite second pulses and B-code time signals, and compensates for time deviations through a Kalman filter algorithm. This allows the data from the bus and each outgoing line to be collected simultaneously, and the location of the collection point is determined and each digital signal is marked with an absolute time stamp.

4. The integrated diagnostic system for ground fault location and traveling wave ranging based on distribution network as described in claim 1, characterized in that: The operation in the integrated fault diagnosis module that determines whether electrical parameters exceed a preset fault initiation threshold is as follows: Extract the zero-sequence current and zero-sequence voltage values ​​obtained at the corresponding time when multiple faults occurred in the recent historical data. Remove the maximum and minimum values ​​of the two values ​​respectively, and calculate the average of the remaining values ​​to obtain the fault initiation threshold for the zero-sequence current and zero-sequence voltage values. When the real-time collected values ​​meet either of the corresponding values, the fault diagnosis process is triggered. The electrical parameters are collected and processed in real time and then compared with the fault initiation threshold in real time. When the electrical parameters exceed the fault initiation threshold, it is determined that a fault may occur, triggering the fault diagnosis process. At the same time, the waveform recording function is started, which records the waveform of the electrical parameters for a period of time before and after the fault occurs.

5. The integrated diagnostic system for ground fault location and traveling wave ranging based on distribution network as described in claim 1, characterized in that: The operation of extracting the amplitude, direction, and abrupt change time of the traveling wavefront using a moving recognition window in the integrated fault diagnosis module is as follows: Extract the electrical parameter data obtained after starting the waveform recording function, establish a traveling wave change coordinate axis with time node as the horizontal axis category and signal value as the vertical axis category, and input the extracted electrical parameter data into the traveling wave change coordinate axis in time order to generate change curves; The motion recognition window is set as F(x, y), where x is the horizontal length of the motion recognition window and y is the vertical width of the motion recognition window. The motion recognition window moves from the starting point of the changing curve, and the changing curve is segmented based on the curve segment extracted by the motion recognition window. The change situation is determined by magnifying and analyzing the curve segment to obtain the moment of the wavefront change. The amplitude and direction of the traveling wavefront are derived from the analysis of the abrupt change in the wavefront.

6. The integrated diagnostic system for ground fault location and traveling wave ranging based on distribution network as described in claim 5, characterized in that: The operation to determine the moment of abrupt change in the wavefront through magnified analysis of the curve segment is as follows: Extract curve segments sequentially, introduce the changes of multiple undisplayed node data into the curve segments between the current time nodes to determine whether there are extreme points in the curve segments that simultaneously have rising and falling changes; If an extreme point exists, a noise threshold is set, and the amplitude corresponding to the extreme point is compared with three times the noise threshold. If the amplitude corresponding to the extreme point is greater than three times the noise threshold, and it appears for the first time in the current order of curve segment extraction, then the current extreme point is the moment of wavefront abrupt change.

7. The integrated diagnostic system for ground fault location and traveling wave ranging based on distribution network as described in claim 5, characterized in that: The operation of deriving the amplitude and direction of the traveling wavefront based on the analysis of the abrupt change in the wavefront is as follows: Centered on the identified moment of wavefront change, the signal amplitudes of multiple sampling points before and after the point are taken, and the peak amplitude within the current sampling interval is calculated, which is the amplitude of the traveling wavefront. The direction of the wavefront is then determined by the change in signal amplitude after the abrupt change in the wavefront. If the signal amplitude increases, the direction of the wavefront is positive; otherwise, if the signal amplitude decreases, the direction of the wavefront is negative. A positive wavefront corresponds to a fault point that is far from the monitoring point, while a negative wavefront corresponds to a fault point that is close to the monitoring point.

8. The integrated diagnostic system for ground fault location and traveling wave ranging based on distribution network as described in claim 1, characterized in that: The operation in the integrated fault diagnosis module that compares the differences in the magnitude and direction of the traveling wave amplitude of each outgoing line is as follows: A convolutional neural network classification model is generated by training a large amount of sample data of different fault types, and the extracted feature parameters are input into the trained convolutional neural network classification model to output a fault probability value for each outgoing line. Compare the waveform amplitudes of all outgoing lines together, and then check the waveform direction of each outgoing line. The outgoing line with the highest fault probability value has the largest waveform amplitude and the waveform direction is different from other outgoing lines. That is, the outgoing line with the highest fault probability value is the faulty line.

9. The integrated diagnostic system for ground fault location and traveling wave ranging based on distribution network as described in claim 1, characterized in that: The operation of calculating the distance to the fault point in the integrated fault diagnosis module, based on the line wave velocity parameters, is as follows: Extract the waveform curve on the faulty line and determine the first wavefront that propagates directly from the fault point to the monitoring point after the fault occurs. This is the initial traveling wavefront. The second characteristic wavefront that is reflected back to the monitoring point after the initial traveling wave reaches the fault point is the fault point reflected wavefront. Based on the amplitude change point of the initial traveling wave wavefront, the corresponding time node t1 is obtained, and based on the amplitude change point of the reflected wave wavefront at the fault point, the corresponding time node t2 is obtained, and the time difference ΔT = t2 - t1 is obtained. Based on historical data, the reaction time of the monitoring point when receiving the returned traveling wave signal and obtaining the signal parameters is t3. The reaction time t3 is compensated into the time difference, and the distance from the fault point to the monitoring point is determined as L by combining the time and wave velocity parameters.

10. The integrated diagnostic system for ground fault location and traveling wave ranging based on distribution network according to claim 9, characterized in that: The formula for calculating the distance L from the fault point to the monitoring point is: L = [v × (△T - t3)] / 2; Where v is the traveling wave velocity, which is calculated by back-calculating the time difference of arrival of the traveling wave at two monitoring points with a known distance.

Images