Distributed transmission cable fault location diagnosis system and location diagnosis method

Abstract

The present invention relates to a distributed transmission cable fault location diagnosis system and location diagnosis method, which includes three parts: a monitoring terminal, a central station, and a user system. The monitoring terminal includes a data acquisition unit, a central control unit, a communication unit, and a power supply unit. The central station includes a pre-processing module, a database, a fault diagnosis module, and a WEB service module. The user system includes a mobile receiving terminal and a computer access terminal. According to the waveform characteristics of the traveling wave current, lightning faults and non-lightning faults are identified; for lightning faults, according to the waveform differences of the traveling wave current, lightning strike back faults and lightning strike bypass faults are identified. By optimizing the arrangement between the main line monitoring nodes, planetary monitoring nodes, and the central station, resource utilization is made more reasonable and the reliability of fault monitoring is improved. When multiple faults occur in the same section of the line, the topological structure of the present invention can find multiple fault points based on waveform superposition and combined with an offline positioning method.

Description

Technical Field

[0001] The present invention relates to the field of G06F: electrical digital data processing, in particular to the field of power transmission line inspection technology, and specifically to a distributed power transmission cable fault location diagnosis system and location diagnosis method. Background Art

[0002] High-voltage transmission cables are laid in four different ways: cable tunnels, conduits, trenches, and direct burial. Their insulation is susceptible to damage from external forces and environmental corrosion. Accidents such as single-phase grounding and interphase short circuits are common, seriously impacting the reliability of the cable power supply. Therefore, quickly and accurately identifying the fault point and promptly repairing the faulty line are crucial tasks in power operation and maintenance.

[0003] The fault diagnosis system device based on distributed installation is installed on the terminal cable body and the grounding wire of the terminal grounding box respectively. It monitors the insulation defect hidden dangers on the cable body and the fault grounding wire in real time, such as discharge traveling waves, fault traveling waves and power frequency current signals. It combines with optical fiber to achieve time synchronization and accurately and quickly locate faults.

[0004] Existing fault location devices are typically distributed along transmission cables in predetermined numbers and intervals. However, the frequency of faults varies significantly across seasons and regions, leading to a waste of resources. Furthermore, in extreme weather conditions or when multiple faults occur simultaneously, these devices struggle to accurately identify the specific fault location due to the combined signal signatures. Summary of the Invention

[0005] Purpose of the invention: To provide a distributed transmission cable fault location and diagnosis system, and further provide a location and diagnosis method based on the above-mentioned fault location and diagnosis system, so as to solve the above-mentioned problems existing in the prior art.

[0006] Technical solution: First, a distributed transmission cable fault location and diagnosis system is proposed, which includes three parts: monitoring terminal, central station, and user system.

[0007] The monitoring terminal includes a data acquisition unit, a central control unit, a communication unit and a power supply unit.

[0008] The central station includes a pre-processing module, a database, a fault diagnosis module, and a WEB service module.

[0009] The user system includes a mobile receiving terminal and a computer access terminal.

[0010] The monitoring terminals are set up at predetermined intervals along the transmission cable to collect current information from the transmission cable and convert the current information into digital signals including non-fault signals and suspected fault signals.

[0011] The central station is in communication with the monitoring terminal and is used to regularly read digital signals from the monitoring terminal, or passively receive suspected fault signals actively sent by the monitoring terminal, and compare the digital signals and suspected fault signals with the built-in database to determine whether a fault has occurred and the type of fault.

[0012] The user system is remotely connected to the central station for receiving the processed fault information from the central station and pushing it to the user side.

[0013] This fault location and diagnosis system uses Rogowski coils within monitoring terminals to directly acquire current information on transmission lines. A high-speed FPGA collects cable fault traveling wave signals and power frequency signals, which are uploaded to the diagnostic monitoring system platform in real time. Using artificial intelligence deep learning algorithms, the system diagnoses and outputs fault diagnosis results, which are sent to maintenance personnel via SMS alerts. The system also allows for real-time viewing of relevant information via a web client. Furthermore, the fault location is determined through collaborative analysis of multiple fault current traveling wave detection points and precise GPS / BD time synchronization to ensure all equipment is synchronized. Data processing ultimately determines the fault location.

[0014] In an embodiment of the first aspect, a density peak clustering algorithm is used to determine the location of the central station and the location and number of monitoring terminals based on the arrangement of the transmission cables, so that a multi-level topological fault location network is formed between the monitoring terminals and the central station;

[0015] The central station is a first-level node, and the monitoring terminals include mainline monitoring nodes distributed along the transmission cable, as well as several planetary monitoring nodes dispersed around the transmission cable in a planetary pattern. Each transmission cable has at least two mainline monitoring nodes, with at least one central station on either side of the mainline monitoring node. Each central station has at least two planetary monitoring nodes within a predetermined range on either side. Adjacent planetary monitoring nodes are interconnected to form a network.

[0016] In an embodiment of the first aspect, a method for arranging the mainline monitoring nodes, the planetary monitoring nodes, and the central station includes:

[0017] S1. Establish a mathematical model, which is defined as follows:

[0018] At least two main line monitoring nodes are installed on each transmission line L. i , and two adjacent main line monitoring nodes on the same transmission line i Separate by a predetermined distance;

[0019] Each two transmission cables L form a link through the central station C, and each central station C is directly connected to at least two planetary monitoring nodes. j .

[0020] S2. Determine the priorities of the mainline monitoring node and the planetary monitoring node.

[0021] S3. Use the density peak clustering algorithm to select multiple nodes whose local density is greater than a predetermined value and whose mutual distance reaches a predetermined value as cluster centers, and define the cluster center as the location of the central station.

[0022] In an embodiment of the first aspect, the monitoring terminal includes a data acquisition unit, a power supply unit, and a central control unit communication unit.

[0023] The data acquisition units are arranged at predetermined intervals along the transmission cables, are actively awakened at a predetermined frequency, and are passively awakened around the clock; the power supply unit independently supplies power to each data acquisition unit; the central control unit is electrically connected to the data acquisition unit, performs preliminary judgment on the current information collected by the data acquisition unit, and filters out non-fault signals and suspected fault signals; the communication unit is telecommunication-connected to the central control unit, and receives non-fault signals and suspected fault signals sent from the central control unit in real time.

[0024] In an embodiment of the first aspect, the central station includes a database, a pre-processing module, a fault diagnosis module, and a Web service module.

[0025] The database establishes communication with the main station through the local area network and regularly updates fault data. During the period of frequent thunderstorms in summer, the database wakes up 12 times a day, and wakes up for 10 minutes every 2 hours.

[0026] In addition, the central station can obtain local weather data and increase the wake-up frequency when the weather data indicates the possibility of thunderstorms. During the rest of the season, the database is in sleep mode and wakes up three times a day, for 5 minutes every 8 hours.

[0027] The pre-processing module is connected to the communication unit by telecommunications, reads non-fault signals and suspected fault signals and performs pre-judgment; the fault diagnosis module is used to read the pre-judgment results processed by the pre-processing module, and compares the pre-judgment results with the database to determine the fault diagnosis information; the Web service module establishes communication with the fault diagnosis module and sends the fault diagnosis information to the user system through the network.

[0028] In an embodiment of the first aspect, a user system includes a computer access terminal and a mobile receiving terminal.

[0029] The computer access terminal establishes communication with the central station to access the fault diagnosis information from the central station; the mobile receiving terminal is interconnected with the computer access terminal to receive the demodulated fault diagnosis information sent by the computer access terminal.

[0030] In an embodiment of the first aspect, the data acquisition unit includes a Rogowski coil, a temperature and humidity monitoring module, an acquisition board, a digital-to-analog conversion chip, an FPGA chip, a single-chip microcomputer, and an optical fiber transmission module.

[0031] When the Rogowski coil senses the current signal on the cable, it converts it into a voltage signal and sends it to the acquisition board. The operational amplifier enhances the drive capability, and the interference-prone single-ended signal is then converted into a differential signal. This differential signal has strong anti-interference capabilities, effectively suppresses EMI, and ensures precise timing positioning. The differential signal then enters a high-precision 14-bit ADC chip, which converts the analog value into a digital value and transmits it to the FPGA chip. The FPGA chip uses a high-stability, high-frequency active crystal oscillator for signal acquisition. After high-speed acquisition by the FPGA, the signal is transmitted to the microcontroller. The microcontroller combines the collected voltage signal with GPS information and temperature and humidity data, and packages it for transmission to the host computer. Fiber optic synchronization is implemented in the FPGA. After the fiber optic transmitter module converts the optical signal into an electrical signal, it is simultaneously transmitted to a pin on the FPGA chip of both devices. Whenever this pin on both devices receives a rising edge, the counters in their FPGA chips are simultaneously reset, achieving time synchronization between the two devices.

[0032] Secondly, a distributed transmission cable fault location and diagnosis method is proposed. The method is implemented based on the above-mentioned fault location and diagnosis system, and the steps are as follows:

[0033] Step 1: Build a fault location and diagnosis system, collect traveling wave waveforms of different fault types, and continuously update and improve the database through adaptive learning;

[0034] Collect traveling wave waveforms of different fault types and establish a fault traveling wave waveform database; based on the traveling wave current waveform, the system identifies lightning faults and non-lightning faults. When the traveling wave tail time is less than a predetermined threshold, it is identified as a lightning fault;

[0035] Identify whether the lightning fault is a lightning strike back fault or a lightning shielding fault. If there is a reverse polarity pulse in front of the main wave, it is identified as a lightning strike back fault; if there is no reverse polarity pulse in front of the main wave, it is identified as a lightning shielding fault.

[0036] The identified fault information is input into the deep learning neural network for self-learning, and the learning results are fed back to the database.

[0037] Step 2: The data acquisition unit divides the line into several sections as a dividing point. After a fault occurs, the section where the fault point is located is determined based on the direction and magnitude of the AC line power frequency fault current, or the direction and polarity of the AC / DC line traveling wave current, thereby narrowing the fault point search range.

[0038] Step 3: The central control unit performs a preliminary judgment on the current information collected by the data acquisition unit, and screens out non-fault signals and suspected fault signals.

[0039] Step 4: The non-fault signals and suspected fault signals filtered out by the central control unit are transmitted to the central station. The pre-processing module in the central station reads the above signals and performs pre-judgment:

[0040] All suspected fault signals are regarded as fault signals and transferred to the fault diagnosis module;

[0041] The non-fault signal is compared with the fault threshold again. If the signal strength is greater than or equal to the fault threshold, it is transferred to the fault diagnosis module for further judgment; if the signal strength is less than the fault threshold, it is directly discarded.

[0042] Step 5: The fault diagnosis module reads the pre-judgment result processed by the pre-processing module, and compares the pre-judgment result with the database to determine the fault diagnosis information including the fault location and fault type.

[0043] Step 6: The fault diagnosis information is sent to the computer access terminal, which demodulates the fault diagnosis information and provides it to the user via mobile phone text messages or WEB to implement fault alarm.

[0044] Beneficial effects: The present invention relates to a distributed transmission cable fault location diagnosis system and location diagnosis method, which can identify lightning faults and non-lightning faults based on the waveform characteristics of the traveling wave current; for lightning faults, it can identify lightning strike back faults and lightning bypass faults based on the waveform differences of the traveling wave current. By optimizing the layout between the main line monitoring nodes, planetary monitoring nodes, and central stations, resource utilization is made more reasonable and the reliability of fault monitoring is improved. When multiple faults occur in the same section of the line, the topological structure of the present invention can find multiple fault points based on waveform superposition and offline positioning methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is the architecture diagram of the distributed transmission cable fault location and diagnosis system.

[0046] Figure 2 A schematic diagram of a multi-level topological fault location network is formed between the monitoring terminal and the central station.

[0047] Figure 3 This is a diagram of the positioning principle involved in the present invention.

[0048] Figure 4 This is a fault location diagnosis flowchart. DETAILED DESCRIPTION

[0049] In the following description, numerous specific details are provided to provide a more thorough understanding of the present invention. However, it will be apparent to those skilled in the art that the present invention may be practiced without one or more of these details. In other instances, certain technical features well known in the art are not described to avoid confusion with the present invention.

[0050] Example 1:

[0051] See Figure 1 This embodiment provides a distributed transmission cable fault location and diagnosis system, which includes three components: monitoring terminals, a central station, and a user system. The monitoring terminals are installed at predetermined intervals along the transmission cable and are used to collect current information from the transmission cable and convert it into digital signals. The digital signals include normal fault signals and suspected fault signals.

[0052] The monitoring terminal includes a data acquisition unit, a power supply unit, a central control unit, and a communication unit. The data acquisition units are installed at predetermined intervals along the transmission cable and are activated at a predetermined frequency and activated passively around the clock. The power supply unit independently supplies power to each data acquisition unit. The central control unit is electrically connected to the data acquisition units and performs preliminary evaluation of the current information collected by the data acquisition units, screening out non-fault signals and suspected fault signals. The communication unit is connected to the central control unit by telecommunications and receives non-fault signals and suspected fault signals from the central control unit in real time.

[0053] The data acquisition unit includes a Rogowski coil, a temperature and humidity monitoring module, an acquisition board, a digital-to-analog conversion chip, an FPGA chip, a single-chip microcomputer, and an optical fiber transmission module.

[0054] The Rogowski coil directly acquires current information on the transmission line; a temperature and humidity monitoring module, located on one side of the Rogowski coil, monitors the temperature and humidity of the surrounding environment of the AC and DC line conductors. An acquisition board collects current information from the Rogowski coil; a digital-to-analog conversion chip converts this current information into a digital quantity; an FPGA chip receives the converted digital quantity from the digital-to-analog conversion chip; a single-chip microcomputer packages the collected voltage signal, combined with GPS information and temperature and humidity information, and transmits it to a host computer. A fiber optic transmission module communicates between two adjacent data acquisition units, converting optical signals into electrical signals and simultaneously transmitting them to a predetermined pin on the FPGA chip to trigger a rising edge signal. Whenever this rising edge signal is received on these pins of two data acquisition units, the counters in their FPGA chips are simultaneously reset to zero, achieving time synchronization between the two devices.

[0055] When the Rogowski coil senses the current signal on the cable, it converts it into a voltage signal and sends it to the acquisition board. The operational amplifier enhances the drive capability, and the interference-prone single-ended signal is then converted into a differential signal. This differential signal has strong anti-interference capabilities, effectively suppresses EMI, and ensures precise timing positioning. The differential signal then enters a high-precision 14-bit ADC chip, which converts the analog value into a digital value and transmits it to the FPGA chip. The FPGA chip uses a high-stability, high-frequency active crystal oscillator for signal acquisition. After high-speed acquisition by the FPGA, the signal is transmitted to the microcontroller. The microcontroller combines the collected voltage signal with GPS information and temperature and humidity data, and packages it for transmission to the host computer. Fiber optic synchronization is implemented in the FPGA. After the fiber optic transmitter module converts the optical signal into an electrical signal, it is simultaneously transmitted to a pin on the FPGA chip of both devices. Whenever this pin on both devices receives a rising edge, the counters in their FPGA chips are simultaneously reset, achieving time synchronization between the two devices.

[0056] The central station is in communication with the monitoring terminal, and is used to periodically read digital signals from the monitoring terminal, or passively receive suspected fault signals actively sent by the monitoring terminal, and compare the digital signals and the suspected fault signals with the built-in database to determine whether a fault has occurred and the type of fault;

[0057] The user system is connected to the central station through remote communication to receive fault information processed by the central station and push it to the user side.

[0058] The central station includes a database, a pre-processing module, a fault diagnosis module, and a Web service module. The database establishes communication with the main station via the local area network and regularly updates fault data. During the summer, when thunderstorms are common, the database wakes up 12 times a day, or 10 minutes every two hours. In addition, the main station can obtain local meteorological data and increase the wake-up frequency when the meteorological data indicates the possibility of a thunderstorm. During the rest of the season, the database is in a dormant state, waking up three times a day, or 5 minutes every eight hours. The pre-processing module is connected to the communication unit by telecommunications, reading non-fault signals and suspected fault signals and performing pre-judgments. The fault diagnosis module is used to read the pre-judgment results processed by the pre-processing module and compare the pre-judgment results with the database to determine the fault diagnosis information. The Web service module establishes communication with the fault diagnosis module and sends the fault diagnosis information to the user system via the network.

[0059] The user system includes a computer access terminal and a mobile receiving terminal. The computer access terminal establishes communication with the central station to access the fault diagnosis information from the central station; the mobile receiving terminal is connected to the computer access terminal to receive the demodulated fault diagnosis information sent by it.

[0060] The above system mainly has the following functions:

[0061] Fault recording waveform

[0062] (1) Record the power frequency waveform and traveling wave waveform of the line when the fault occurs.

[0063] (2) Collect traveling wave waveforms of different fault types, such as lightning strikes, tree barriers, wind deflection, forest fires, bird damage, and other fault current waveforms, and establish a fault traveling wave waveform database. (Based on the traveling wave current waveform, the system identifies lightning faults and non-lightning faults. The tail time of the traveling wave of the lightning fault current will be less than 40 microseconds. Among them, lightning faults are further divided into lightning strike back and lightning shielding fault. The main wave of the lightning strike back fault has a reverse polarity pulse, while the lightning shielding fault has no reverse polarity pulse.)

[0064] Fault Location

[0065] The monitoring terminal serves as a demarcation point, dividing the line into several sections. After a fault occurs, the fault point should be located in the section according to the direction and magnitude of the AC line's power-frequency fault current, or the direction and polarity of the traveling-wave current on AC and DC lines, thereby narrowing the search area. (Generally, a set of distributed fault diagnosis devices (one for each of the A, B, and C phases) is installed every 20 to 30 kilometers.)

[0066] Fault Identification

[0067] The system can manually or automatically identify lightning faults and non-lightning faults based on the traveling wave current waveform. (Based on the traveling wave current waveform, the system identifies lightning faults and non-lightning faults. The tail time of the traveling wave of the lightning fault current will be less than 40 microseconds.)

[0068] Fault alarm

[0069] After a line fault occurs, the diagnostic results can be provided to users via mobile phone text messages, web publishing, etc. to provide fault alarms.

[0070] Operation monitoring

[0071] When the line is running, it can monitor the operating current of the AC line in real time, and monitor the temperature and humidity of the surrounding environment of the AC and DC line conductors when necessary.

[0072] See Figure 3 This device uses the principle of traveling wave two-terminal ranging: when a line fault occurs, a high-frequency traveling wave signal is generated. The traveling wave signal propagates from the fault point along the conductor to both ends. Monitoring terminals installed at both ends of the line collect the initial traveling wave of the fault and measure and record the wave head time. The time difference between the two traveling wave heads can be calculated. If the line length between the two monitoring points and the traveling wave speed are known, the fault point location can be calculated. The calculation formula is as follows:

[0073]

[0074]

[0075] Where, Indicates the distance between the monitoring terminal M and the fault point C, represents the distance between monitoring terminal N and fault point C, and L represents the distance between monitoring terminal M and monitoring terminal N; It represents the time it takes for the traveling wave at the fault point C to be transmitted to the monitoring terminal M. It indicates the time it takes for the traveling wave from the fault point C to be transmitted to the monitoring terminal N.

[0076] The troubleshooting process is as follows:

[0077] Low-frequency and high-frequency coefficients

[0078] Wavelet analysis methods have excellent time-frequency localization capabilities. The wavelet transform process decomposes a signal into low-frequency coefficients and high-frequency coefficients. Wavelet transforming the original signal can achieve data dimensionality reduction and feature extraction. Wavelet decomposition involves performing inner product operations on wavelet bases of varying scales and the measured signal. The Mallat algorithm is a commonly used wavelet decomposition algorithm. Mallat decomposition of measured power system fault signals yields scaling coefficients and wavelet coefficients. The scaling coefficients generate the low-frequency portion of the signal, while the wavelet coefficients generate the high-frequency portion.

[0079] Signal singularity detection

[0080] Selecting an appropriate wavelet basis to perform wavelet decomposition on the measured power system fault signal yields the high-frequency portion of the fault signal. Analysis of this high-frequency portion allows detection of signal singularities and the time of the power system fault, but does not determine the location of the fault. To locate the fault point, a modulus maximum algorithm is required to further examine each sampling point in the power system. When a power system fault occurs, if the transient fault signal is singular, the singular points in the signal can be used to diagnose the moment of the fault. Since the singular points of a signal are modulus maxima of the wavelet transform, but these are not necessarily singular points of the signal, wavelet methods often employ threshold values ​​for power system fault diagnosis. If the modulus maximum is greater than the threshold, a fault is considered to have occurred at that location and time; if the modulus maximum is less than the threshold, a fault is considered to have not occurred at that location and time. The algorithm for detecting singular points in power system transient signals using wavelet theory proceeds as follows:

[0081] S1. Use Mallat algorithm to perform wavelet decomposition on the sudden change signal of power system, decompose the highest layer low-frequency coefficient and high-frequency coefficient, and calculate the maximum value of wavelet transform coefficient ;

[0082] S2, threshold and duration Set up;

[0083] S3. Combination time length Filter the maximum modulus of wavelet transform coefficients and determine the screening threshold and , if the following equations are satisfied at the same time:

[0084]

[0085]

[0086] Then the modulus maximum is retained, otherwise the modulus maximum is not retained.

[0087] S4. Observe the distribution of the retained modulus maxima, thereby realizing the diagnosis of the power system fault time and fault location.

[0088] Example 2:

[0089] See Figure 4 , this embodiment proposes a distributed transmission cable fault location and diagnosis method as follows:

[0090] Collect traveling wave waveforms of different fault types and continuously update and improve the database through adaptive learning; collect traveling wave waveforms of different fault types and establish a fault traveling wave waveform database; based on the traveling wave current waveform, the system identifies lightning faults and non-lightning faults. When the traveling wave tail time is less than the predetermined threshold, it is identified as a lightning fault; identify whether the lightning fault is a lightning strike or a lightning shielding fault. If there is a reverse polarity pulse in front of the main wave, it is identified as a lightning strike fault; if there is no reverse polarity pulse in front of the main wave, it is identified as a lightning shielding fault; input the identified fault information into the deep learning neural network for self-learning, and feed the learning results back to the database.

[0091] The database is updated actively at a nonlinear frequency, or passively when activated by the central station. It is connected to meteorological data and regularly receives weather information for the designated area. The database communicates with the central station via a local area network and regularly updates fault data. During the summer, when thunderstorms are common, the database wakes up 12 times daily, for 10 minutes every two hours. The central station also accesses local meteorological data and increases its wake-up frequency when it indicates a possible thunderstorm. During the rest of the year, the database remains dormant, waking up three times daily for 5 minutes every eight hours.

[0092] The data acquisition unit acts as a dividing point to divide the line into several sections. After a fault occurs, the section where the fault point is located is determined based on the direction and magnitude of the AC line's power frequency fault current, or the direction and polarity of the AC / DC line's traveling wave current, thereby narrowing the fault point search range.

[0093] See Figure 2 For the arrangement of transmission cables, the density peak clustering algorithm is used to determine the location of the central station, as well as the location and number of monitoring terminals, so that a multi-level topological fault location network is formed between the monitoring terminals and the central station; wherein the central station is a first-level node, and the monitoring terminals include mainline monitoring nodes distributed along the transmission cables, and several planetary monitoring nodes dispersed in a planetary shape around the transmission cables.

[0094] Among them, the arrangement methods between the main line monitoring nodes, planetary monitoring nodes, and central stations include:

[0095] A mathematical model is established and defined as follows:

[0096] At least two main line monitoring nodes are installed on each transmission line L. i , and two adjacent main line monitoring nodes on the same transmission line i Separate by a predetermined distance;

[0097] Each two transmission cables L form a link through the central station C, and each central station C is directly connected to at least two planetary monitoring nodes. j , the planetary monitoring nodes connected to two adjacent central stations C j The following conditions are met:

[0098]

[0099]

[0100] Where, 、 Represents any two planetary monitoring nodes connected to two adjacent central stations C, Indicates the distance between the two planetary monitoring nodes with the largest separation between two adjacent central stations C; 、 Represents the nodes of two adjacent central stations; Indicates the straight-line distance between two adjacent central stations; Indicates the distance between the two planetary monitoring nodes with the shortest interval connected to two adjacent central stations C;

[0101] Determine the priorities of the mainline monitoring node and the planetary monitoring node. The priority function is as follows:

[0102]

[0103] Where, 、 、 、 is the weight coefficient, , Indicates the main line monitoring node i The distance to the central station C, Indicates the main line monitoring node i The distance level to the central station C; where, definition For level I, definition For level II, definition It is level III, defined is level IV; level I is 0.5, level II is 2, level III is 5, and level IV is 8; Indicates the main line monitoring node i To the set threshold of the central station C; Indicates the main line monitoring node i To the planetary monitoring node j distance, Indicates the main line monitoring node i To Central Station j The distance level of For level I, definition For level II, definition It is level III, defined is level IV; level I is 0.5, level II is 2, level III is 5, and level IV is 8; Indicates the main line monitoring node i The actual frequency of detected faults, Mainline monitoring node i The average frequency of detected faults during a predetermined period, Represents a planetary monitoring node j The actual frequency of detected faults, Represents a planetary monitoring node j The average frequency of faults detected during a predetermined period;

[0104] Using the density peak clustering algorithm, multiple nodes with local density greater than a predetermined value and a mutual distance reaching a predetermined value are selected as cluster centers, and the cluster center is defined as the location of the central station;

[0105] The density peak clustering algorithm needs to calculate the local density of nodes and the threshold of node spacing. The calculation formula for the local density of nodes is as follows:

[0106]

[0107]

[0108] Where, represents the local density of the current node, Indicates the i The local density of nodes, represents the distance between the two corresponding nodes, Representation node i and nodes j The spacing, represents the cutoff distance;

[0109] The position of the planetary monitoring node is used as the independent variable, and the density peak clustering algorithm is used again to select a node with the maximum local density value and the maximum mutual distance as the cluster center, and define the cluster center as the position of the main line monitoring node;

[0110] The calculation formula for the threshold of node spacing is as follows:

[0111]

[0112] Where, Indicates the threshold of node spacing, which represents the distance between the node and the nearest node when the local density is greater than the current node. represents the local density of the j-th node, Representation node i and nodes j The spacing, represents the maximum local density of a predetermined node p.

[0113] The central control unit makes a preliminary judgment on the current information collected by the data acquisition unit, and filters out non-fault signals and suspected fault signals.

[0114] The non-fault signals and suspected fault signals filtered out by the central control unit are transmitted to the central station. The pre-processing module in the central station reads the above signals and performs pre-judgment: all suspected fault signals are regarded as fault signals and transferred to the fault diagnosis module; the non-fault signals are compared with the fault threshold again for judgment. If the signal strength is greater than or equal to the fault threshold, it is transferred to the fault diagnosis module for further judgment; if the signal strength is less than the fault threshold, it is directly discarded.

[0115] The fault diagnosis module reads the pre-judgment result processed by the pre-processing module, and compares the pre-judgment result with the database to determine the fault diagnosis information including the fault location and fault type.

[0116] The fault diagnosis information is sent to the computer access terminal, which demodulates the fault diagnosis information and provides it to the user in the form of mobile phone text messages or WEB publishing to realize fault alarm.

[0117] In summary, the distributed fault diagnosis device of the present invention uses Rogowski coils to directly obtain current information on the transmission line, collects cable fault traveling wave signals and fault power frequency signals through high-speed FPGA, and uploads them to the diagnostic monitoring system platform in real time. The fault diagnosis results are diagnosed and output through artificial intelligence deep learning algorithms. The diagnosis results are pushed to the operation and maintenance personnel in the form of alarm text messages. The relevant information can be viewed in real time through the web client, and multiple fault current traveling wave detection points are used for collaborative analysis. GPS / BD is used to accurately synchronize the time of all devices to ensure that the time of all devices is synchronized. After data processing, the fault point location is finally obtained. Time synchronization is performed using optical fiber to ensure that the time of each device is relatively consistent. After a fault occurs, the system will provide the diagnosis results to the operation and maintenance personnel in various ways such as text messages and WEB publishing to realize fault alarm. According to the characteristics of the traveling wave current waveform, lightning faults and non-lightning faults are identified; for lightning faults, lightning strike back faults and lightning shielding faults are identified based on the difference in the waveform of the traveling wave current. The accurate fault point location is determined based on the traveling wave positioning method. Artificial faults are simulated to establish a preliminary fault database. If multiple faults occur on the same line, it is difficult to accurately locate the fault point due to waveform superposition. In this case, offline positioning methods can be used to find multiple fault points.

[0118] As described above, although the present invention has been shown and described with reference to specific preferred embodiments, it should not be construed as limiting the present invention itself. Various changes may be made to it in form and detail without departing from the spirit and scope of the present invention as defined in the appended claims.

Claims

1. A distributed transmission cable fault location and diagnosis system, characterized in that: include: Monitoring terminals are arranged at predetermined intervals along the transmission cable to collect current information from the transmission cable and convert the current information into digital signals; Digital signals include non-fault signals and suspected fault signals; monitoring terminals include: Mainline monitoring nodes are distributed along the transmission cables. At least two mainline monitoring nodes i are installed on each transmission cable L, and there is at least one central station C on both sides of the mainline monitoring node. A plurality of planetary monitoring nodes are dispersed around the transmission cable in a planetary shape, and there are at least two planetary monitoring nodes within a predetermined range on one side of each central station, and each two adjacent planetary monitoring nodes are interconnected to form a network; Each two transmission cables L form a link through the central station C. Each central station C is directly connected to at least two planetary monitoring nodes j. The planetary monitoring nodes j connected to two adjacent central stations C meet the following conditions: In the formula, jm and jn represent any two planetary monitoring nodes connected to two adjacent central stations C. Indicates the distance between the two planetary monitoring nodes with the largest interval between two adjacent central stations C; Cm and Cn represent the nodes of two adjacent central stations; D Cm,Cn Indicates the straight-line distance between two adjacent central stations; Indicates the distance between the two planetary monitoring nodes with the shortest interval connected to two adjacent central stations C; The central station is in communication with the monitoring terminal and is used to regularly read the digital signals from the monitoring terminal and compare them with its built-in database to determine whether a fault has occurred and the type of fault; or passively receive suspected fault signals actively sent by the monitoring terminal and compare them with its built-in database to determine whether a fault has occurred and the type of fault; The database is actively updated at a nonlinear frequency, or passively updated by being woken up by the main station; the database is connected to the meteorological data and regularly receives meteorological information of the current predetermined area; the database establishes communication with the main station via the local area network and regularly updates the fault data; The user system is remotely connected to the central station for receiving fault information processed by the central station and sending it to the user side; A multi-level topological fault location network is formed between the mainline monitoring nodes, planetary monitoring nodes, and central stations. The priority functions of the mainline monitoring nodes and planetary monitoring nodes are as follows: Where w1, w2, w3, w4 are weight coefficients, D ic represents the distance from the mainline monitoring node i to the central station C, G represents the distance level from the main line monitoring node i to the central station C; i and G j They represent the actual frequencies of faults detected by the mainline monitoring node i and the planetary monitoring node j, and are the average frequencies of faults detected by the mainline monitoring node i and the planetary monitoring node j during the predetermined period, respectively.

2. The fault location diagnosis system according to claim 1, characterized in that: The location of the central station, and the location and number of monitoring terminals are determined using a density peak clustering algorithm.

3. The fault location diagnosis system according to claim 1, characterized in that: The monitoring terminal includes: Data collection units are set up along the transmission cable at predetermined intervals, are actively awakened at a predetermined frequency, and are passively awakened around the clock; A power supply unit, which independently supplies power to each of the data acquisition units; a central control unit, electrically connected to the data acquisition unit, performing preliminary judgment on the current information collected by the data acquisition unit, and screening out non-fault signals and suspected fault signals; as well as The communication unit is connected to the central control unit by telecommunication and receives the non-fault signal and the suspected fault signal sent by the central control unit in real time.

4. The fault location diagnosis system according to claim 3, characterized in that: The central station includes: The database establishes communication with the main station through the local area network and regularly updates the fault data; A pre-processing module, connected to the communication unit by telecommunications, reads non-fault signals and suspected fault signals and performs pre-determination; a fault diagnosis module, configured to read the pre-judgment result processed by the pre-processing module, and compare the pre-judgment result with the database to determine fault diagnosis information; The Web service module establishes communication with the fault diagnosis module and sends the fault diagnosis information to the user system through the network.

5. The fault location diagnosis system according to claim 1, characterized in that: The user system includes: A computer access terminal establishes communication with the central station and accesses fault diagnosis information from the central station; The mobile receiving terminal is interconnected with the computer access terminal and receives the demodulated fault diagnosis information sent by the computer access terminal.

6. The fault location diagnosis system according to claim 3, characterized in that: The data acquisition unit further comprises: Rogowski coil, directly obtains current information on the transmission line; A temperature and humidity monitoring module is provided on one side of the Rogowski coil to monitor the temperature and humidity of the surrounding environment of the AC and DC line conductors; an acquisition board, used for acquiring current information from the Rogowski coil; A digital-to-analog conversion chip, used to convert the current information into a digital quantity; The FPGA chip receives the digital quantity converted by the digital-to-analog conversion chip; The single chip microcomputer combines the collected voltage signal with GPS information and temperature and humidity information to transmit it to the host computer; The optical fiber transmission module communicates between two adjacent data acquisition units and is used to convert optical signals into electrical signals. At the same time, it transmits the rising edge signal to the predetermined pin of the FPGA chip to trigger the rising edge signal. Whenever the pins of the two data acquisition units receive the rising edge signal, the counters in their FPGA chips are cleared at the same time to achieve time synchronization between the two devices.

7. A method for locating and diagnosing faults in distributed power transmission cables, implemented based on the fault locating and diagnosing system according to any one of claims 1 to 6, characterized in that: The steps include: Step 1: Build a fault location and diagnosis system, collect traveling wave waveforms of different fault types, and continuously update and improve the database through adaptive learning; Step 2: The data acquisition unit divides the line into several sections as a demarcation point. After a fault occurs, the fault point is determined based on the direction and magnitude of the AC line power frequency fault current, or the direction and polarity of the AC / DC line traveling wave current, thereby narrowing the fault point search range. Step 3: The central control unit makes a preliminary judgment on the current information collected by the data acquisition unit, and screens out non-fault signals and suspected fault signals; Step 4: The non-fault signals and suspected fault signals filtered out by the central control unit are transmitted to the central station. The pre-processing module in the central station reads the above signals and performs pre-judgment: All suspected fault signals are regarded as fault signals and transferred to the fault diagnosis module; The non-fault signal is compared with the fault threshold again. If the signal strength is greater than or equal to the fault threshold, it is transferred to the fault diagnosis module for further judgment; if the signal strength is less than the fault threshold, it is directly discarded; Step 5: The fault diagnosis module reads the pre-judgment result processed by the pre-processing module and compares the pre-judgment result with the database to determine the fault diagnosis information including the fault location and fault type; Step 6: The fault diagnosis information is sent to the computer access terminal, which demodulates the fault diagnosis information and provides it to the user via mobile phone text messages or WEB to implement fault alarm.

8. The fault location and diagnosis method according to claim 7, characterized in that: Step 1 further includes: Step 1-1, collect traveling wave waveforms of different fault types and establish a fault traveling wave waveform database; Step 1-2: Based on the traveling wave current waveform, the system identifies lightning faults and non-lightning faults. When the traveling wave tail time is less than a predetermined threshold, it is identified as a lightning fault. Step 1-3: Identify whether the lightning fault is a lightning strike back fault or a lightning shielding fault. If there is a reverse polarity pulse in the main wave front, it is identified as a lightning strike back fault; if there is no reverse polarity pulse in the main wave front, it is identified as a lightning shielding fault. Steps 1-4: Input the identified fault information into the deep learning neural network for self-learning, and feed the learning results back to the database.

9. The fault location and diagnosis method according to claim 7, characterized in that: The process of building a fault location and diagnosis system further includes: S1. Establish a mathematical model that meets the following conditions: At least two main line monitoring nodes i are installed on each transmission cable L, and two adjacent main line monitoring nodes i on the same transmission cable are separated by a predetermined distance; Each two transmission cables L form a link through the central station C, and each central station C is directly connected to at least two planetary monitoring nodes j; S2. Determine the priority of the mainline monitoring node i and the planetary monitoring node j; S3. Use the density peak clustering algorithm to select multiple nodes whose local density is greater than a predetermined value and whose mutual distance reaches a predetermined value as cluster centers, and define the cluster center as the location of the central station.

Images