A method and device for online identification and diagnosis of breakdown faults in cross-connected grounded cable lines

Abstract

The present invention discloses an online identification and diagnosis method and device for breakdown faults in a cross-connected grounded cable line. The device includes a data acquisition unit and an analysis and diagnosis unit, and is applicable to a single-core cable cross-connected grounding method. A data acquisition unit needs to be installed at each grounding box position within the cross-connected section of the cable line. The data acquisition unit synchronously collects signals on the interconnected row or single-core grounding wire through an acquisition sensor, and monitors the on-site situation in real time. When a breakdown fault occurs in the cable line, the fault signal is collected, and the signal polarity is calculated using a fault curve model and a feature extraction algorithm. Based on the basic principle that the waveform propagation directions at both ends of the fault point are opposite, the fault cross-connected cable loop is first determined, and then the cable fault interval section is further determined, eliminating the inconvenience of fault finding on the cross-connected grounded cable line, and facilitating cable operation and maintenance personnel to quickly find faults.

Description

Technical Field

[0001] The present invention relates to the field of online monitoring of power system cables, and in particular to an online identification and diagnosis method and device for a breakdown fault of a cross-connected grounded cable line. Background Art

[0002] During operation, power cables gradually experience insulation aging and even insulation breakdown, leading to cable line failures. Damaged cable insulation can easily lead to major incidents such as phase-to-phase short circuits and single-phase grounding, severely impacting power supply reliability. After a cable fault occurs, precise fault location must be determined. Currently, operations and maintenance units primarily use fault-finding instruments to manually locate the fault after a cable breakdown. However, due to the uncertain nature of cable faults, a wide variety of fault-finding instruments are required. Interpreting fault data parameters relies heavily on technician experience, making the task challenging and often taking hours or even days. With technological advancements, online fault warning and location devices are increasingly being used in transmission lines. However, these devices still have limitations. While they are effective for overhead lines, cables with direct grounding at one end, and low-voltage three-core cables, they are less effective for long-distance cables with multiple cross-connected sections. With the expansion of cities, the number of long-distance cable lines is increasing, creating an urgent need for diagnostic systems that can monitor cable breakdown faults online and locate the fault zone.

[0003] The GB50217-2018 standard for designing power engineering cables stipulates that single-core power cables for long-distance onshore AC systems utilize a cross-connection grounding method, dividing the line into several units. Within each unit, the cable is divided into three equally spaced sections. Two sets of insulating joints are installed within each section. The metal sheath at the insulating joints is then transposed using coaxial cable to a cross-connection grounding box, where it is then grounded via a cable sheath protector. A set of insulating joints is installed between sections, with both ends of the insulating joints directly grounded. This ensures that the induced voltages on the metal sheaths of the equally spaced cables within each section cancel each other out due to their 120° phase shift.

[0004] The existing diagnostic method for online cable fault warning and positioning devices uses the principle of double-end traveling wave ranging. This method is effective for overhead lines, three-core medium and low voltage cable lines, and high-voltage single-core cable lines with simple grounding methods, but it is not very effective for long-distance, multi-segment cable lines with cross-interconnected grounding methods.

[0005] Due to the above problems, there is an urgent need for a diagnostic method and device that can monitor cable fault signals on site and accurately determine the cable fault interval in order to comprehensively monitor cable fault signals, quickly determine the fault interval, and then find the fault point. Summary of the Invention

[0006] In order to solve the above technical problems, the present invention proposes an online identification and diagnosis method and device for breakdown faults in cross-connected grounded cable lines, which performs real-time monitoring of each grounding point of the cable line. When an abnormally large signal appears, the fault signal is collected, and the polarity is automatically analyzed based on the theoretical model. The signal propagation direction is determined by the signal polarity direction, thereby quickly determining the fault phase and fault interval, providing assistance for the rapid detection and repair of cable faults.

[0007] The technical solution adopted in the present invention is:

[0008] An online identification and diagnosis device for breakdown faults in cross-connected grounded cable lines, a data acquisition unit and an analysis and diagnosis unit;

[0009] The data acquisition unit is used to synchronously collect high-frequency fault waveforms on the grounding box interconnection row or single-core grounding wire when the cable breaks down;

[0010] The analysis and diagnosis unit is used to summarize the collected data of each data collection unit in the complete cross-interconnection section, perform polarity analysis and calculation on the collected high-frequency fault waveform, identify the positive and negative polarity of the fault waveform, and summarize the polarity of the fault waveform collected by each data collection unit in the cross-interconnection section. Based on the principle that the polarity of the traveling waves at both ends of the fault point is opposite, the fault circuit is first determined from the polarity of the fault waveform at the direct grounding points at both ends, and then the fault interval segment is determined from the fault circuit, so as to achieve rapid interval determination of the breakdown fault waveform in the cross-interconnection section.

[0011] Furthermore, the data acquisition unit includes:

[0012] Three data acquisition sensors are installed on the three-phase grounding wires to sense the high-frequency traveling waves generated when the cable breaks down;

[0013] The isolation protection circuit is used to divide and isolate the high-frequency traveling wave sensed by the data acquisition sensor, eliminate the high voltage surge and common mode voltage in the high-frequency traveling wave, and send it to the signal conditioning circuit;

[0014] The signal conditioning circuit is used to de-jitter, de-noise and amplify the amplitude of the high-frequency traveling wave processed by the isolation protection circuit and send it to the signal synchronization circuit;

[0015] The signal synchronization circuit is used to synchronize the time of the three high-frequency traveling waves that have completed signal conditioning, and transmit the synchronized signals to the high-speed acquisition circuit;

[0016] Three high-speed acquisition circuits are used to sample the waveforms of the synchronized high-speed traveling wave signals, convert analog-to-digital signals into digital signals, and transmit the digital signals to the control circuit;

[0017] The control circuit is used to perform preliminary processing on the digital signal, package the data according to the communication protocol to obtain a high-frequency fault waveform data packet, and transmit the high-frequency fault waveform data packet to the communication circuit;

[0018] The communication circuit establishes a communication connection with the analysis and diagnosis unit according to the TCP / IP protocol, and is used to send the data packets processed by the control circuit to the analysis and diagnosis unit;

[0019] The synchronous timing circuit is remotely synchronized by the analysis and diagnosis unit to ensure that the breakdown fault waveforms collected by each data acquisition unit are at the same time.

[0020] Furthermore, the three data acquisition sensors use directional high-frequency Rogowski coil sensors with a 3dB response bandwidth of 5MHz; at the same detection point, the high-frequency Rogowski coil sensors are respectively installed on three interconnected ground bars or single-core ground wires with the same installation direction.

[0021] Furthermore, the analysis and diagnosis unit includes:

[0022] The communication module is used to establish a communication connection and synchronous timing with the data acquisition unit, receive and analyze the high-frequency fault waveform data packet transmitted by the data acquisition unit, and transmit the high-frequency fault waveform to the polarity calculation module;

[0023] The polarity calculation module is used to receive the high-frequency fault waveform analyzed from the communication module, and use the established fault waveform curve model and feature extraction algorithm to identify the waveform polarity. The fault waveform curve model simulates the high-frequency fault waveform as an underdamped oscillation motion of an object with a correction coefficient, which conforms to the law of propagation and attenuation of the breakdown fault signal in the cable. By inputting the coordinate information of the second and third peaks and troughs of the analyzed high-frequency fault waveform into the fault waveform curve model, the damped oscillation waveform onset peak value, underdamping coefficient, oscillation frequency and curve phase angle under no damping coefficient in the curve model are calculated to obtain the high-frequency fault waveform. The feature extraction algorithm calculates the time period of underdamped oscillation from the fitting curve of the high-frequency fault waveform from back to front, removes the influence of high-frequency spike pulse interference and establishes three unrelated feature quantities, and uses weighted coefficients to calculate the values ​​of the three feature quantities to form a fault waveform polarity correction formula. The above weighted coefficients are obtained by neural network training based on laboratory samples, and the fault waveform polarity correction formula is used to calculate the polarity coefficient, so as to judge the polarity direction of the fault waveform of each data acquisition unit respectively, and collect the fault waveform polarity of the data acquisition units in the same cross-connected segment and send it to the interval judgment module;

[0024] The interval judgment module is used to determine the cable line breakdown fault interval between the grounding boxes based on the synchronous triggering fault waveform polarity of the 12 data acquisition unit measurement points in the same cross-connected section collected by the polarity calculation module. Based on the fault interval judgment algorithm and the basic principle that the fault waveforms at both ends of the breakdown fault point have opposite polarities, the module further determines the cable line breakdown fault interval between the grounding boxes.

[0025] The data display module is used to receive data from the polarity calculation module and the interval judgment module, and display the fault waveform and fault interval diagnosis results.

[0026] Furthermore, the analysis and diagnosis unit is connected to each data acquisition unit via an optical fiber communication line to achieve synchronous timing.

[0027] Furthermore, the fault waveform curve model is:

[0028]

[0029] f(t) is the curve of the waveform amplitude changing with time, t is the time from 0 to infinity in microseconds, F0 is the peak value of the damped oscillation waveform, β is the underdamping coefficient, ω is the oscillation frequency of the waveform without the damping coefficient, is the corrected oscillation frequency with a damping coefficient, Ф is the phase angle of the curve; A0sin(γt1) is a randomly appearing spike signal, represented by a half-cycle ultra-narrowband sinusoidal signal, where γ is the frequency of the ultra-narrowband sinusoidal signal, which is much larger than ω, and t0 represents the time when the signal appears. is the duration of the signal, i.e., half a cycle. A0sin(γt1) is in the first half cycle of the curve model. g(t) is the random glitch interference below 10 MHz, with an amplitude much smaller than F0.

[0030] By obtaining the discrete sequence of high-frequency fault waveforms f(n) = (f(1), f(n2), ... f(n)), the actual parameters of the curve (F0, β, ω, Ф) are calculated, thereby calculating the waveform characteristics. The actual parameters of the curve (F0, β, ω, Ф) are calculated by the following formula:

[0031]

[0032]

[0033]

[0034]

[0035] Where f(p2) and f(p3) are the second and third peak-to-trough amplitudes in the discrete waveform sequence f(n), respectively. p2 , t p3are the times of the second and third peaks and troughs in the discrete waveform sequence f(n), respectively. i is the zero-crossing position sequence in f(n), and Δt is the sampling interval.

[0036] Furthermore, the fault waveform polarity correction formula is:

[0037]

[0038] k1, k2, and k3 are the correction coefficients of the characteristic first wave polarity P1, energy polarity P2, and maximum peak polarity P3, respectively.

[0039]

[0040]

[0041]

[0042] t p1 The time of the first peak and trough in the discrete waveform sequence f(n), f(n A ) is the amplitude of the maximum peak;

[0043] By weighting each characteristic coefficient, the fault waveform polarity coefficient P is calculated to determine the polarity direction of the fault waveform.

[0044] Furthermore, the fault interval judgment algorithm is:

[0045] First, determine the faulty cross-connected cable loop by checking whether the polarity of the three-phase waveforms at the direct grounding location is the same. Perform the following calculations:

[0046]

[0047]

[0048]

[0049]

[0050]

[0051]

[0052] P 1a 、P 1b 、P 1c They represent the polarity of the high-frequency fault waveform detected at the three-phase grounding wire positions A, B, and C of the first grounding box (direct grounding) of the cross-connection section, where the positive polarity is assigned a value of 1 and the negative polarity is assigned a value of 0; 2a 、P 2b、P 2c They represent the polarity of the high-frequency fault waveform detected at the three-phase grounding wire positions of A, B, and C in the second grounding box (cross-connection) of the cross-connection section, where the positive polarity is assigned a value of 1 and the negative polarity is assigned a value of 0; 3a 、P 3b 、P 3c They represent the polarity of the high-frequency fault waveform detected at the three-phase grounding wire positions A, B, and C of the third grounding box (cross-connection) in the cross-connection section, where the positive polarity is assigned a value of 1 and the negative polarity is assigned a value of 0; 4a 、P 4b 、P 4c They represent the polarity of the high-frequency fault waveform detected at the three-phase grounding wire positions of A, B, and C in the fourth grounding box (cross-connection) of the cross-connection section, where the positive polarity is assigned a value of 1 and the negative polarity is assigned a value of 0; D 1a 、D 1b 、D 1c Respectively represent the comparison results of the high-frequency fault waveform detected at the three-phase grounding position of A, B, and C in the first grounding box (direct grounding) of the cross-connection section with the polarity of the other two phases. If the polarities are opposite, the value is assigned to 1, otherwise it is 0; D 1a 、D 1b 、D 1c Respectively represent the comparison results of the high-frequency fault waveform detected at the third grounding line position of the fourth grounding box (direct grounding) of the cross-connection section, A, B, and C, with the polarity of the other two phases. If the polarities are opposite, the value is assigned to 1, otherwise it is 0;

[0053] In the A1-B2-C3, B1-C2-A3, and C1-A2-B3 cross-connection structures, the fault loop L is determined as:

[0054]

[0055] In the A1-C2-B3, B1-A2-C3, and C1-B2-A3 cross-connection structures, the fault loop L is determined as:

[0056]

[0057] In the determined fault loop L, the polarity of the fault waveforms of adjacent monitoring points is compared, and the interval segment with opposite polarity of adjacent measuring points is found in loop L, which is determined as the fault interval L on loop L. D .

[0058] A method for online identification and diagnosis of a breakdown fault in a cross-interconnected grounded cable line is provided, using the device for online identification and diagnosis of a breakdown fault in a cross-interconnected grounded cable line. The method specifically comprises:

[0059] A data acquisition unit is installed at each of the four grounding boxes in the same cable cross-connection section. The data acquisition unit synchronously collects high-frequency fault waveforms on the grounding box interconnection row or single-core grounding wire when the cable breaks down. The three high-frequency Rogowski coil sensors of the data acquisition unit are installed in the same direction. Each data acquisition unit is connected to the analysis and diagnosis unit via a communication optical fiber.

[0060] The analysis and diagnosis unit summarizes the collected data of each data acquisition unit in the complete cross-interconnection section, performs polarity analysis and calculation on the collected high-frequency fault waveform, identifies the positive and negative polarity of the fault waveform, and summarizes the polarity of the fault waveform collected by each data acquisition unit in the cross-interconnection section. Based on the principle that the polarity of the traveling waves at both ends of the fault point is opposite, the fault circuit is first determined from the polarity of the fault waveform at the direct grounding points at both ends, and then the fault interval is determined from the fault circuit, so as to achieve rapid interval determination of the breakdown fault waveform in the cross-interconnection section.

[0061] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0062] (1) The sensor for measuring cable fault signals of the present invention adopts a high-frequency Rogowski coil sensor, which can detect the large current signal generated at the moment of cable breakdown. The Rogowski coil has a relatively large deformation, which ensures the safety of signal detection. Moreover, the Rogowski coil is a flexible coil, which is easy to install and can be widely used in grounding boxes.

[0063] (2) The present invention monitors the fault signals at multiple points of a complete cable loop without GPS timing at both ends. In the long cable cross-connected grounding mode, by identifying the polarity direction of the high-frequency fault waveform, based on the basic principle that the polarity of the fault waveform is opposite at both ends of the breakdown point, the fault cross-connected cable loop is first determined, and the breakdown fault is further determined in the cable section between the grounding boxes. The source of the fault signal can be determined, and the fault phase and fault section can be located. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is a schematic diagram of an online identification and diagnosis device for breakdown faults in cross-connected grounded cable lines according to the present invention;

[0065] Figure 2 This is a workflow diagram of an online identification and diagnosis method for a cross-connected grounded cable line breakdown fault according to the present invention;

[0066] Figure 3 This is a workflow diagram of an online identification and diagnosis method for a cross-connected grounded cable line breakdown fault according to the present invention;

[0067] Figure 4 This is a schematic diagram of the installation points of data acquisition sensors in a complete cross-connected and grounded cable line according to an embodiment of the present invention;

[0068] Figure 5 This is a schematic diagram of the signal directions of each line segment in an example of a single-point failure in a cross-connected grounded cable line according to an embodiment of the present invention. DETAILED DESCRIPTION

[0069] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0070] On the one hand, the present invention provides an online identification and diagnosis device for a cross-connected grounded cable line breakdown fault, such as Figure 1 As shown, the system consists of a data acquisition unit and an analysis and diagnosis unit. The data acquisition unit collects high-frequency fault waveforms at each grounding location on the cross-connected cable line. The analysis and diagnosis module synchronizes the timing of each data acquisition unit through remote synchronization to ensure that the collected fault waveform data is from the same time period. The analysis and diagnosis unit collects data from each acquisition unit and uses calculations and analysis to determine the interval where the cable fault occurred.

[0071] The data acquisition unit includes: three data acquisition sensors, an isolation protection circuit, a signal conditioning circuit, a high-speed acquisition circuit, a signal synchronization circuit, a control circuit, a communication circuit and a synchronization circuit. The three data acquisition sensors utilize high-frequency Rogowski coils with a response bandwidth greater than 5MHz. They are installed on the three-phase grounding wires at the monitoring point to sense the traveling waves generated by cable breakdown faults. The isolation protection circuit, located at the front end of the acquisition process, uses transformer and capacitor coupling technology to isolate the input terminal from the grounding system of the subsequent circuits, blocking high voltage surges and high common-mode voltages, thereby protecting the chip and circuits. The signal conditioning circuit includes de-noising and amplification, de-jittering the double-ended differential signal to remove jitter noise and amplifying the processed signal to the matching range of the high-speed acquisition circuit, thereby improving measurement accuracy and sensitivity. The high-speed acquisition circuit digitizes the analog signal through high-speed sampling and sends the digital signal to the control circuit. The signal synchronization circuit is used to synchronize the acquisition of the three high-speed acquisition circuits. The control circuit performs preliminary processing on the digital signal, packages the data according to the communication protocol, and sends the data packets to the communication circuit. The communication circuit establishes a communication connection with the analysis and diagnosis unit according to the TCP / IP protocol and transmits the data packets. The synchronization circuit ensures that the breakdown fault waveforms collected by each data acquisition unit are at the same time.

[0072] The analysis and diagnosis unit includes a communication module, a polarity calculation module, an interval judgment module and a data display module.

[0073] Communication module, used to establish communication connection with each data acquisition unit, receive and analyze fault waveform data packets, and continuously perform synchronous timing;

[0074] The polarity calculation module uses the established fault waveform curve model and feature extraction algorithm to perform polarity identification and determine the polarity direction of the fault waveform of each data acquisition unit;

[0075] The interval judgment module determines the polarity of the synchronous trigger fault waveform of 12 measuring points in the same cross-connected section based on the polarity calculation module. According to the fault interval judgment algorithm, it further realizes the interval judgment of the cable line breakdown fault in the cable section between the grounding boxes.

[0076] The data display module is used to display the fault waveform and fault interval diagnosis results of the data acquisition unit.

[0077] Generally speaking, the analysis and diagnosis unit receives the fault waveform data packet sent by the data acquisition unit, parses and processes the data packet, uses the established fault waveform model and algorithm to identify the polarity, and determines the direction of the fault current at each monitoring point; comprehensively analyzes the information detected at multiple points, and for two different cross-interconnection structures, locates the fault interval based on the fault interval determination method, and determines whether the location of the fault is in this section of the cross-interconnection cable. If it is in this section of the cross-interconnection cable, it further quickly determines in which cable interval the fault occurs.

[0078] The analysis and diagnosis unit is connected to each data acquisition unit through an optical fiber communication line. The connection structure is parallel communication to achieve synchronous timing.

[0079] According to another aspect of the present invention, a method for locating a fault interval using the cross-connected grounded cable line breakdown fault online identification and diagnosis device is proposed, such as Figure 2 As shown, the following steps are included:

[0080] Step 1, data collection

[0081] Four of the aforementioned online cable breakdown fault identification and diagnosis devices were installed in the same cable cross-connection section to collect data on the fault waveform, with one device installed at each grounding box, for a total of four monitoring points. Within the cross-connection grounding system, the cables within each cross-connection section were divided into three equal-length segments. Two sets of insulating joints were installed within each segment, and the metal sheath at the insulating joints was led to the cross-connection grounding box via coaxial cable for transposition. The metal sheath was then grounded via a cable sheath protector. A set of insulating joints was installed between the segments, with both ends of the insulating joints directly grounded.

[0082] At a single monitoring point, three acquisition sensors are installed on three interconnected ground bars or single-core ground wires. The three phases are independently triggered for real-time data acquisition. When the threshold is not triggered, real-time waveforms are uploaded at equal intervals of 100ms. After the threshold is triggered, waveforms are collected simultaneously for 100ms for each of the three phases. The collected signals are transmitted to the analysis and diagnosis unit via the data acquisition unit. The acquisition sensors use high-frequency Rogowski coil sensors. Due to their directional nature, at the same monitoring point, when the acquisition sensors are installed in the same direction, the collected signals are in the same direction. When the sensors are installed in the opposite direction, the collected signals are in the opposite direction.

[0083] At the next detection point, the above steps are repeated synchronously to complete the monitoring within a cable loop, and the analysis data is transmitted to the background diagnosis system via the communication network.

[0084] Step 2: Calculate the waveform polarity

[0085] The analysis and diagnosis unit aggregates and identifies the signal waveforms at each point. Within each cross-connection section, in the absence of a fault, the induced currents generated by each small section of the metal sheath are equal in amplitude and 120 degrees out of phase, ensuring a nearly balanced three-phase system. When a single-phase cable breaks down, a transient large pulse signal is emitted. A fault waveform curve model is established, from which characteristic parameters such as the first wave polarity, energy density, and maximum peak value are extracted. Training samples are used to develop a fault waveform polarity correction formula to determine the polarity and direction of the fault waveform.

[0086] In detail,

[0087] 1) After obtaining the actual waveform sequence and performing a 10MHz low-pass filter, the fault waveform curve model is constructed as follows:

[0088]

[0089] f(t) is the curve of the waveform amplitude changing with time after 10MHz low-pass filtering, t is the time from 0 to infinity in microseconds, F0 is the peak value of the damped oscillation waveform, β is the underdamping coefficient, ω is the oscillation frequency of the waveform without the damping coefficient, is the modified oscillation frequency with a damping coefficient, Ф is the phase angle of the curve; A0sin(γt1) is a randomly appearing spike signal, which can be represented by a half-cycle ultra-narrowband sinusoidal signal, where γ is the frequency of the ultra-narrowband sinusoidal signal, which is much larger than ω, and t0 represents the time when the signal appears. is the duration of the signal, i.e., half a cycle. A0sin(γt1) is generally in the first half cycle of the curve model. g(t) is the randomly occurring glitch interference below 10 MHz, with an amplitude much smaller than F0.

[0090] Based on the fault waveform curve model, the actual fault waveform discrete sequence f(n)=(f(1),f(n2),…f(n)) is obtained, as shown in Figure 3 The figure shows a schematic diagram of the actual fault waveform sequence obtained. n ranges from 0 to N, where each waveform segment is 100ms and N is the total number of waveform points (2500000). The actual parameters of the curve (F0, β, ω, Ф) are calculated to facilitate the subsequent calculation of the characteristic quantities of the curve. The steps are as follows:

[0091] (1) Scan the discrete sequence of f(n) waveforms with a window, find the n sequence whose amplitude in f(n) is in the interval (-0.005, 0.005), and output them in ascending order (n0, n1, n i …n m ), to compare data, first set the zero-crossing sequence, include n0 in the sequence, so that i = 1, such as n i+1 -n i Much larger than n i -n i-1 , then we think n i+1 and n i is the same zero-crossing point, otherwise, it is considered to be a new zero-crossing point, and n i Input to the zero-crossing sequence, compare in sequence, and get the zero-crossing sequence (o1, o2...o j ), j is often greater than 7 according to experience.

[0092] (2) Since A0sin(γt1) is generally in the first half cycle of the curve model, β is calculated from the second half cycle (i.e., starting from the second zero-crossing point p2).

[0093] First, calculate the position of the second peak and trough of the curve in the discrete sequence f(n) Index f(n) can be obtained from f(p2), at time f(t) Here f is the high-speed sampling rate, which is 25MHz in this method; calculate the position of the third peak and valley of the curve in the discrete sequence f(n) Index f(n) can get f(p3), at time f(t) t g2 and t g3 Substitute into the model formula respectively

[0094]

[0095] get:

[0096]

[0097] Since A0sin(γt1) acts on the first half cycle of the curve, at t p2 and t p3 When A0sin(γt1)=0; g(t p2 ) and g(t p3 ) is noise, which is much smaller than Can be ignored; t p2 and t p3 The positions of adjacent peaks and troughs:

[0098]

[0099] The above formula can be simplified to obtain:

[0100]

[0101] The curve f(t) can be calculated

[0102]

[0103] (3) Calculate the average cycle time of the f(t) curve From the formula of the typical model, we know Right now

[0104]

[0105] Given β, where ω is positive, we can calculate

[0106]

[0107] (4) Through the known point coordinates (t p2 ,f(t p2 ))Substitute into the model formula,

[0108]

[0109] t p2 is the peak and trough, according to the underdamped oscillation characteristics, F0≥0, the simplified formula can be obtained

[0110]

[0111] From the model, we can see that

[0112]

[0113]

[0114] but,

[0115]

[0116] Therefore, based on the above calculations, the curve model of the actually collected waveform sequence can be obtained.

[0117] 2) Extract the first wave polarity, energy density, and maximum peak value from the model, and use the training samples to form a fault waveform polarity correction formula to determine the polarity direction of the fault waveform. The calculation steps are as follows:

[0118] (1) Calculation of first wave polarity

[0119] Substitute the model formula into the first wave peak and trough position If f(t p1 )>0, the first wave is positive polarity; f(t p1 )≤0, the first wave is negative polarity. Here, the first characteristic value first wave polarity is defined as

[0120]

[0121] (2) Calculation of energy polarity

[0122] Calculate the positive and negative waveform energy values ​​of the actual collected waveform discrete sequence f(n). First, scan and classify f(n) with 0 as the boundary, that is,

[0123]

[0124] Calculate the total energy of the amplitude normalized for the positive half-cycle and negative half-cycle waveform sequences respectively and

[0125]

[0126]

[0127] Compare and The energy polarity is

[0128]

[0129] The average energy density of the polar curve is

[0130]

[0131] Energy polarity can also be expressed as

[0132]

[0133] Right now:

[0134]

[0135] (3) Calculation of maximum peak polarity

[0136] Calculate the maximum peak amplitude A, maximum peak polarity P3 and position n of the actual collected discrete waveform sequence f(n) A .

[0137] A=Max|f(n)|

[0138] n A =f -1 (A)

[0139]

[0140] (4) The training sample forms a fault waveform polarity correction formula to determine the polarity direction of the fault waveform

[0141] By weighting each characteristic coefficient, the fault waveform polarity coefficient P is calculated:

[0142]

[0143] k1, k2, and k3 are the correction coefficients of the first wave polarity, energy polarity, and maximum peak polarity of the characteristic quantities, respectively. The correction coefficients are numerically trained by simulating breakdown fault samples in the laboratory. Different sample types include changing the test breakdown point position, changing the applied breakdown voltage, and changing the test breakdown type until the correction coefficient P reaches the required recognition accuracy. New test samples are input, and the recognition accuracy is set to 100% in the initial stage.

[0144] Step 3: Determine the fault range

[0145] When a single-phase cable breaks down, it emits a transient large pulse signal. Based on the fundamental principle that waveforms at both ends of the cable near the breakdown point have opposite polarities, the cable cross-connection structure characteristics, and the current flow characteristics, the polarity of the synchronously triggered fault waveforms at 12 measuring points in the same cross-connection section is determined using a polarity identification formula. The following fault loop determination algorithm is established to further determine the cable line breakdown fault in the interval between grounding boxes. The specific method is as follows:

[0146] 1) Extract the waveform of each monitoring point. The installation diagram of the monitoring point is as follows: Figure 4 As shown, based on the polarity identification formula mentioned above, the positive and negative polarity of the waveform is judged. If it is positive, the value is assigned to 1, and if it is negative, the value is assigned to 0. The polarity matrix of the monitoring points in the same cross-connection segment is established:

[0147]

[0148] Positions 1 and 4 are direct grounding points, and positions 2 and 3 are cross-connected grounding points.

[0149] 2) Based on the basic principle that the polarity of the waveforms at both ends of the breakdown point is opposite, first determine whether the polarity of the three-phase waveforms at positions 1 and 4 is the same, and perform the following calculations:

[0150]

[0151]

[0152]

[0153]

[0154]

[0155]

[0156] Comparison of D under two cross-connection structures 1a 、D 1b 、D 1c 、D 4a 、D 4b 、D 4c , judge the faulty circuit based on its numerical rules.

[0157] (1) Under the first A1-B2-C3, B1-C2-A3 and C1-A2-B3 cross-connection structures:

[0158] Such as D 1a 、D 1b 、D 1c 、D 4a 、D 4b 、D 4c If both are 0, the fault section is not in this cross-connection cable; 1a 、D 1b 、D 1c 、D 4a 、D 4b 、D 4c If only one of them is 1 and the others are 0, or if only two of them are 1 and the others are 0, the fault circuit L is determined as follows:

[0159]

[0160] (2) Under the second A1-C2-B3, B1-A2-C3 and C1-B2-A3 cross-connection structures:

[0161] Such as D 1a 、D 1b 、D 1c 、D 4a 、D 4b 、D 4cIf both are 0, the fault section is not in this cross-connection cable; 1a 、D 1b 、D 1c 、D 4a 、D 4b 、D 4c If only one of them is 1 and the others are 0, or if only two of them are 1 and the others are 0, the fault circuit L is determined as follows:

[0162]

[0163] (3) In the fault loop L, the polarity of the fault waveforms of adjacent monitoring points is compared, and the interval segment with opposite polarity of adjacent measuring points is found in the loop, which is the fault interval L on the loop. D .

[0164] As an implementation method, the following takes the fault interval judgment under a cross-connection structure as an example. Under the premise that the fault waveform polarity has been calculated, Figure 5 As shown in the figure, the fault interval determination method is as follows:

[0165] 1. Determine the ground loop:

[0166] 1) If the direction of the fault signal collected on the direct grounding line of phase A of the #3 insulated joint is opposite to that of the other two phases, the fault point exists in the A1-A2-B3-B4-C5-C6 loop, which is recorded as grounding loop A;

[0167] 2) If the direction of the fault signal collected on the direct grounding line of phase B of the #3 insulated connector is opposite to that of the other two phases, the fault point exists in the B1-B2-C3-C4-A5-A6 loop, which is recorded as grounding loop B;

[0168] 3) If the direction of the fault signal collected on the direct grounding line of phase C of the #3 insulated connector is opposite to that of the other two phases, the fault point exists in the C1-C2-A3-A4-B5-B6 loop, which is recorded as grounding loop C.

[0169] 2. Determine where the fault point is in the grounding circuit

[0170] 1) If there is a fault in ground loop A, compare the polarity directions of the fault waveforms at the four monitoring points A1-L1, A2-B3, B4-C5, and C6-L2 in sequence.

[0171] 2) If the direction of A1-L1 is opposite to that of other monitoring points, the fault point is determined to be in the A1-A2 section;

[0172] 3) If the directions of A1-L1 and A2-B3 are the same and opposite to those of B4-C5 and C6-L2, the fault point is determined to be in the B3-B4 section;

[0173] 4) If the directions of A1-L1, A2-B3, and B4-C5 are the same and opposite to the direction of C6-L2, the fault point is determined to be in the C5-C6 section;

[0174] 5) If faults occur in the other two phases, the fault finding method is the same as above.

[0175] Therefore, by monitoring a complete cross-connected grounding system and comprehensively analyzing the polarity and direction of the fault signal waveforms at multiple monitoring points, the fault section can be quickly located.

[0176] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention's description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An online identification and diagnosis device for breakdown faults in cross-connected grounded cable lines, characterized in that: include: Data acquisition unit and analysis and diagnosis unit; The data acquisition unit is used to synchronously collect high-frequency fault waveforms on the grounding box interconnection row or single-core grounding wire when the cable breaks down; The analysis and diagnosis unit is used to summarize the collected data of each data collection unit in the complete cross-connection section, perform polarity analysis and calculation on the collected high-frequency fault waveform, identify the positive and negative polarity of the fault waveform, and summarize the polarity of the fault waveform collected by each data collection unit in the cross-connection section. Based on the principle that the polarity of the traveling waves at both ends of the fault point is opposite, the fault circuit is first determined from the polarity of the fault waveform at the direct grounding of both ends, and then the fault interval is determined from the fault circuit, so as to achieve rapid interval determination of the breakdown fault waveform in the cross-connection section; The analysis and diagnosis unit includes: The communication module is used to establish a communication connection and synchronous timing with the data acquisition unit, receive and analyze the high-frequency fault waveform data packet transmitted by the data acquisition unit, and transmit the high-frequency fault waveform to the polarity calculation module; The polarity calculation module is used to receive the high-frequency fault waveform analyzed from the communication module, and use the established fault waveform curve model and feature extraction algorithm to identify the waveform polarity. The fault waveform curve model simulates the high-frequency fault waveform as an underdamped oscillation motion of an object with a correction coefficient, which conforms to the law of propagation and attenuation of the breakdown fault signal in the cable. By inputting the coordinate information of the second and third peaks and troughs of the analyzed high-frequency fault waveform into the fault waveform curve model, the damped oscillation waveform onset peak value, underdamping coefficient, oscillation frequency and curve phase angle under no damping coefficient in the curve model are calculated to obtain the high-frequency fault waveform. The feature extraction algorithm calculates the time period of underdamped oscillation from the fitting curve of the high-frequency fault waveform from back to front, removes the influence of high-frequency spike pulse interference and establishes three unrelated feature quantities, and uses weighted coefficients to calculate the values ​​of the three feature quantities to form a fault waveform polarity correction formula. The above weighted coefficients are obtained by neural network training based on laboratory samples, and the fault waveform polarity correction formula is used to calculate the polarity coefficient, so as to judge the polarity direction of the fault waveform of each data acquisition unit respectively, and collect the fault waveform polarity of the data acquisition units in the same cross-connected segment and send it to the interval judgment module; The interval judgment module is used to determine the cable line breakdown fault interval between the grounding boxes based on the synchronous triggering fault waveform polarity of the 12 data acquisition unit measurement points in the same cross-connected section collected by the polarity calculation module. Based on the fault interval judgment algorithm and the basic principle that the fault waveforms at both ends of the breakdown fault point have opposite polarities, the module further determines the cable line breakdown fault interval between the grounding boxes. The data display module is used to receive data from the polarity calculation module and the interval judgment module, and display the fault waveform and fault interval diagnosis results.

2. The device for online identification and diagnosis of breakdown faults in cross-connected grounded cables according to claim 1, characterized in that: The data acquisition unit includes: Three data acquisition sensors are installed on the three-phase grounding wires to sense the high-frequency traveling waves generated when the cable breaks down; The isolation protection circuit is used to divide and isolate the high-frequency traveling wave sensed by the data acquisition sensor, eliminate the high voltage surge and common mode voltage in the high-frequency traveling wave, and send it to the signal conditioning circuit; The signal conditioning circuit is used to de-jitter, de-noise and amplify the amplitude of the high-frequency traveling wave processed by the isolation protection circuit and send it to the signal synchronization circuit; The signal synchronization circuit is used to synchronize the time of the three high-frequency traveling waves that have completed signal conditioning, and transmit the synchronized signals to the high-speed acquisition circuit; Three high-speed acquisition circuits are used to sample the waveforms of the synchronized high-speed traveling wave signals, convert analog-to-digital signals into digital signals, and transmit the digital signals to the control circuit; The control circuit is used to perform preliminary processing on the digital signal, package the data according to the communication protocol to obtain a high-frequency fault waveform data packet, and transmit the high-frequency fault waveform data packet to the communication circuit; The communication circuit establishes a communication connection with the analysis and diagnosis unit according to the TCP / IP protocol, and is used to send the data packets processed by the control circuit to the analysis and diagnosis unit; The synchronous timing circuit is remotely synchronized by the analysis and diagnosis unit to ensure that the breakdown fault waveforms collected by each data acquisition unit are at the same time.

3. The device for online identification and diagnosis of breakdown faults in cross-connected grounded cables according to claim 2, characterized in that: The three data acquisition sensors use directional high-frequency Rogowski coil sensors with a 3dB response bandwidth of 5MHz. At the same detection point, the high-frequency Rogowski coil sensors are respectively installed on three mutually connected ground bars or single-core ground wires with the same installation direction.

4. The device for online identification and diagnosis of breakdown faults in cross-connected grounded cables according to claim 1, characterized in that: The analysis and diagnosis unit is connected to each data acquisition unit via an optical fiber communication line to achieve synchronous timing.

5. The device for online identification and diagnosis of breakdown faults in cross-connected grounded cables according to claim 1, characterized in that: The fault waveform curve model is: ，t≥0， ≤ ≤ ； is a curve showing the change of waveform amplitude over time, t is the time in microseconds from 0 to infinity, is the peak value of the damped oscillation waveform, is the underdamping coefficient, is the oscillation frequency of the waveform without damping coefficient, is the modified oscillation frequency with damping coefficient, is the phase angle of the curve; is a randomly appearing spike signal, represented by a half-cycle ultra-narrowband sinusoidal signal, where is the frequency of the ultra-narrowband sinusoidal signal, which is much larger than , Indicates the time when the signal appears. is the duration of the signal, i.e. half a cycle, In the first half cycle of the curve model; It is a random glitch interference below 10MHz, with an amplitude much smaller than ; By obtaining a discrete sequence of high-frequency fault waveforms =( , ,… ), calculate the actual parameters of the curve ( ), thereby calculating the waveform characteristics, where the actual parameters of the curve ( ) is calculated using the following formula: ； ； ； ； in Waveform discrete sequence The second and third peak-trough amplitudes, Waveform discrete sequence The time of the second and third peaks and troughs, for The zero-crossing position sequence, is the sampling interval.

6. The device for online identification and diagnosis of breakdown faults in cross-connected grounded cables according to claim 1, characterized in that: The fault waveform polarity correction formula is: ； They are the first wave polarity of the characteristic quantity , energy polarity and maximum peak polarity The correction factor is ； ； ； is a discrete sequence of waveforms The time of the first peak and trough in is the amplitude of the maximum peak; By weighting each characteristic coefficient, the fault waveform polarity coefficient P is calculated to determine the polarity direction of the fault waveform.

7. The device for online identification and diagnosis of breakdown faults in cross-connected grounded cables according to claim 1, characterized in that: The fault interval judgment algorithm is: First, determine the faulty cross-connected cable loop by checking whether the polarity of the three-phase waveforms at the direct grounding location is the same. Perform the following calculations: ； ； ； ； ； ； They represent the polarity of the high-frequency fault waveform detected at the three-phase grounding wire positions of A, B, and C in the first grounding box of the cross-connection section, respectively. The positive polarity is assigned a value of 1, and the negative polarity is assigned a value of 0. They represent the polarity of the high-frequency fault waveform detected at the three-phase grounding wire positions A, B, and C of the second grounding box in the cross-connection section, respectively. The positive polarity is assigned a value of 1, and the negative polarity is assigned a value of 0. They represent the polarity of the high-frequency fault waveform detected at the three-phase grounding wire positions of A, B, and C in the third grounding box of the cross-connection section, respectively. The positive polarity is assigned a value of 1, and the negative polarity is assigned a value of 0. They represent the polarity of the high-frequency fault waveform detected at the three-phase grounding wire positions of the fourth grounding box A, B, and C in the cross-connection section, respectively. The positive polarity is assigned a value of 1, and the negative polarity is assigned a value of 0. 、 、 Respectively represent the comparison results of the high-frequency fault waveform detected at the three-phase grounding wire position of A, B, and C in the first grounding box of the cross-connection section with the polarity of the other two phases. If the polarities are opposite, the value is assigned to 1, otherwise it is 0; 、 、 Respectively represent the comparison results of the high-frequency fault waveform detected at the three-phase grounding wire position of the fourth grounding box A, B, and C in the cross-connection section with the polarity of the other two phases. If the polarities are opposite, the value is assigned to 1, otherwise it is 0; In the A1-B2-C3, B1-C2-A3, and C1-A2-B3 cross-connection structures, the fault loop L is determined as: ； In the A1-C2-B3, B1-A2-C3, and C1-B2-A3 cross-connection structures, the fault loop L is determined as: ； In the determined fault loop L, the polarity of the fault waveforms of adjacent monitoring points is compared, and the interval with opposite polarity of adjacent measuring points is found in loop L, which is determined as the fault interval on loop L. .

8. A method for online identification and diagnosis of breakdown faults in a cross-interconnected grounded cable line, performed using the device for online identification and diagnosis of breakdown faults in a cross-interconnected grounded cable line according to any one of claims 1 to 7, characterized in that: The method specifically includes: A data acquisition unit is installed at each of the four grounding boxes in the same cable cross-connection section. The data acquisition unit synchronously collects high-frequency fault waveforms on the grounding box interconnection row or single-core grounding wire when the cable breaks down. The three high-frequency Rogowski coil sensors of the data acquisition unit are installed in the same direction. Each data acquisition unit is connected to the analysis and diagnosis unit via a communication optical fiber. The analysis and diagnosis unit summarizes the collected data of each data acquisition unit in the complete cross-interconnection section, performs polarity analysis and calculation on the collected high-frequency fault waveform, identifies the positive and negative polarity of the fault waveform, and summarizes the polarity of the fault waveform collected by each data acquisition unit in the cross-interconnection section. Based on the principle that the polarity of the traveling waves at both ends of the fault point is opposite, the fault circuit is first determined from the polarity of the fault waveform at the direct grounding points at both ends, and then the fault interval is determined from the fault circuit, so as to achieve rapid interval determination of the breakdown fault waveform in the cross-interconnection section.

Images