Power transmission line abnormity diagnosis and positioning method based on high-frequency sampling and fault recording

Abstract

The invention discloses a power transmission line abnormity diagnosis and positioning method based on high-frequency sampling and fault recording, and the method comprises the steps: achieving the collection of transient traveling wave currents of a lightning stroke fault and a non-lightning stroke fault of a power transmission line through high-frequency sampling and fault recording; by comparing and analyzing the transient traveling wave current collected during the lightning stroke fault, the identification of lightning stroke shielding failure and back flashover fault properties can be realized; transient traveling wave currents collected during non-lightning stroke faults are compared and analyzed, a fault interval is determined according to the flowing direction of the traveling wave currents, accurate positioning of fault points is carried out in the determined interval through a traveling wave distance measurement method, in addition, for the hidden danger of insulation performance reduction caused by natural aging of a power transmission line, a discharge waveform is detected, and the fault location accuracy of the power transmission line is improved. The method can greatly improve the detection precision of the power transmission line fault monitoring work, effectively reduces the manual workload, reduces the labor cost investment, and facilitates the improvement of the operation safety and stability of the power transmission line and the efficient electric energy transmission capability.

Description

Technical Field

[0001] The present invention relates to the field of power transmission line abnormality diagnosis and positioning methods, and in particular to a power transmission line abnormality diagnosis and positioning method based on high-frequency sampling and fault recording. Background Art

[0002] With the continuous improvement of my country's comprehensive national strength and the rapid development of the power industry, my country's power system has gradually entered the era of smart grid and ubiquitous power Internet of Things. As an important part of the power system, high-voltage transmission lines are responsible for the transmission of electric energy and are the lifeblood of the power system and even the national economic construction. But at the same time, high-voltage transmission lines are also the places where most faults occur in the transmission network. Power lines will inevitably fail. In the thunderstorm season, some areas even have frequent switch tripping. To improve the reliability of transmission line operation, on the one hand, it is necessary to continuously improve and strengthen the insulation capacity of the line and improve the disaster resistance of the line. On the other hand, it is necessary to enhance the line fault perception ability, improve the efficiency of line fault repair, and achieve rapid power restoration. At present, a large number of various fault location devices are installed in the transmission lines. In fault troubleshooting, the fault location device can effectively shorten the scope of fault troubleshooting, which plays an important role in reducing the repair time, improving the repair efficiency, and achieving rapid power restoration. However, there are still some problems in the actual use of various fault location devices, which are mainly manifested in the following aspects:

[0003] (1) The accuracy of fault location still needs to be improved. At present, although there are fault location devices that narrow the scope of fault investigation, the fault location has not yet been achieved within the specific range, and manual search is still required to find the line fault point. When a line fault occurs in a mountainous area, due to the long distance of the transmission line, complex terrain, and harsh climate, the line fault point investigation is time-consuming and labor-intensive, which reduces the efficiency of power repair.

[0004] (2) The efficiency of fault type analysis needs to be improved. The analysis of the specific type of line fault is the key to fault repair. For line faults caused by different factors such as typhoons, thunderstorms, and wildfires, the corresponding fault repair methods are also different. At present, the transmission line fault location device has not yet realized the automatic diagnosis of different types of faults. Due to the complex geographical environment of the transmission line, the existing fault repair method still requires manual on-site confirmation of the fault type and then carry out repairs, which reduces the efficiency of line repairs.

[0005] Moreover, with the continuous improvement of transmission capacity and voltage level and the continuous growth of transmission distance, the failure of high-voltage transmission lines will bring more serious harm and impact to the stable operation of the power system, national economic construction and people's daily life; in recent years, there have been many major power outages caused by transmission line failures at home and abroad. Rapid and accurate transmission line fault classification is the premise of fault location and rapid restoration of power supply to the power grid, and it is also an important part of power system fault analysis. Therefore, the research on the classification method of high-voltage transmission line faults is of great significance to ensure the safety and economy of the system. On the other hand, the geographical environment of the high-voltage transmission corridor is relatively harsh, and line faults often occur in relatively harsh weather conditions; at the same time, the rapid action of the relay protection device makes most of the fault lines have no obvious signs of damage, which makes it extremely difficult to find transmission line faults. Therefore, accurate and rapid high-voltage transmission line fault location can not only shorten the fault repair time, reduce power outage losses, and reduce the inspection work intensity of staff, but also has important significance for improving the power supply reliability and stable operation of the power grid.

[0006] Therefore, it is of great practical significance to develop a transmission line abnormality diagnosis and positioning method based on high-frequency sampling and fault recording, improve the transmission line fault positioning accuracy, realize automatic diagnosis of fault types, establish a maintenance suggestion expert database system, and make maintenance suggestions through automatic fault analysis, which can greatly improve the emergency repair efficiency of operation and maintenance personnel and improve the reliability of power supply of the power grid. Summary of the invention

[0007] In view of the above-mentioned prior art problems, the present invention provides the following solutions to solve the above-mentioned technical problems:

[0008] A method for diagnosing and locating abnormalities in a power transmission line based on high-frequency sampling and fault recording, characterized in that it includes the following two aspects:

[0009] a. Through high-frequency sampling and fault recording, the transient traveling wave current of the transmission line lightning fault and non-lightning fault is collected; by comparing and analyzing the transient traveling wave current collected during lightning fault, the nature of lightning shielding and counter-strike faults can be identified; by comparing and analyzing the transient traveling wave current collected during non-lightning fault, and determining the fault interval through the flow direction of the traveling wave current, the fault point can be accurately located within the determined interval through the traveling wave ranging method;

[0010] b. Aiming at the hidden danger of insulation performance degradation caused by natural aging of transmission lines, the discharge waveform is detected to determine the fault type and actively warn.

[0011] Preferably, the high-frequency sampling and fault recording measures the AC current by directly putting a Rogowski coil on the conductor to be measured, thereby restoring the original waveform of the line to the greatest extent, and then collecting the short-circuit fault current and lightning current traveling wave waveform of the overhead line, and using Pix2Pix with a conditional generative adversarial network (cGAN, Conditional Generative Adversarial Nets) as the framework as a translation algorithm for the traveling wave image.

[0012] Preferably, the Rogowski coil is an AC current sensor, which is a hollow annular coil and has two types: flexible and rigid.

[0013] Preferably, the traveling wave ranging method is divided into two types: double-end traveling wave ranging and single-end traveling wave ranging, wherein double-end traveling wave ranging is the main method. Double-end traveling wave ranging refers to installing traveling wave ranging equipment at both ends of the line, and calculating the fault point location by detecting the time difference of the first traveling wave arriving at both ends;

[0014] Single-ended traveling wave ranging is supplemented by single-ended traveling wave ranging, which means installing a measuring device at one end of the line and using the time difference between the first traveling wave measured when the line fails and the reflected second traveling wave to calculate the distance between the measuring point and the fault point.

[0015] Preferably, the traveling wave ranging method relies on a high-precision system clock, and the clock timing accuracy needs to reach the nanosecond level, wherein the timing accuracy relies on the GPS timing system.

[0016] Preferably, the GPS timing system uses the high-precision atomic clock carried by GPS satellites to generate reference signals and time standards, and provides global time services with a timing accuracy of up to 2 billionths of a second. The GPS timing system mainly uses the characteristics of GPS precise timing to achieve unified timing of the system. The GPS receiver can simultaneously receive signals from 4 to 8 satellites within its field of view at any time, and extract and output two time signals from them after decoding:

[0017] a. The pulse signal PPS with a time interval of 1s, its pulse leading edge is consistent with the international standard time (Greenwich Mean Time).

[0018] The synchronization error of Weizhi time) does not exceed 1us;

[0019] b. The international standard time and date code corresponding to the leading edge of the PPS pulse output through the serial port.

[0020] Preferably, the system time synchronization process includes the following three steps:

[0021] a. The device acquires satellite clock data, including basic time information and second pulse and signal;

[0022] b. The first time synchronization system has its own clock based on satellite data;

[0023] c. Compare the pulse-per-second signal with the system's own one-second pulse signal to perform real-time compensation on the system clock.

[0024] Preferably, there are three main time synchronization methods: hard time synchronization (pulse time synchronization), soft time synchronization (using communication messages to synchronize time synchronization) and coded time synchronization (IRIG-B time synchronization).

[0025] a. Hard synchronization generally uses second pulse synchronization. This format of time signal outputs one pulse per second or per minute. The clock pulse output does not contain specific time information, but it can provide the device with a stable and accurate clock reference to the greatest extent, and is not affected by external environments such as temperature and magnetic fields. As long as it can receive satellite signals,

[0026] No., it will take effect;

[0027] b. Soft time synchronization uses communication messages to transmit complete time including year, month, day, hour, minute, second and millisecond;

[0028] c. The coded time synchronization adopts IRIG-B time synchronization, which is divided into two types: modulation and non-modulation. IRIG-B is a comprehensive time synchronization solution. Its message contains time information such as seconds, minutes, hours, and date. At the same time, each frame can be divided into 100 code elements, which is equivalent to a pulse synchronization signal.

[0029] Technical effects and advantages of the present invention:

[0030] The present invention can provide a more precise fault location algorithm, so that the acquisition system can accurately identify transmission line faults such as single-phase grounding fault distance, phase-to-phase short circuit fault distance, circuit breaker fault distance, instantaneous fault distance, lightning strike fault, shielding failure fault, etc. with an accuracy rate of more than 99%, and the error of the measurement position can be controlled within 100 meters.

[0031] The present invention has a strong anti-interference ability, and its measurement accuracy is not affected by factors such as fault resistance, line parameter asymmetry, mutual inductor error, line distributed capacitance, noise, etc. It can accurately measure and record line fault information within 30 kilometers without interruption and without recording dead zones, and can calmly deal with measurement problems under various special network topologies.

[0032] The present invention can achieve the purpose of high-speed sampling of traveling waves and nanosecond time synchronization through an ultra-low power consumption ARM processor with a sampling rate of 2MHz and a high-precision second pulse timing and a constant temperature compensation crystal oscillator of a satellite, and can utilize the second pulse to correct the system clock coefficient in real time.

[0033] The present invention can effectively separate the power frequency and traveling wave signals through the self-developed Rogowski coil lightning waveform integration module, and deeply restore the real current situation of the transmission line while ensuring that the separated waveform is consistent with the original current waveform of the line.

[0034] The present invention can provide a brand-new fault location algorithm, can widely collect on-site waveforms, establish a sample library of power frequency, traveling wave, and hidden danger current, and perform fault simulation and simulated ranging, which can greatly improve the intelligence and accuracy of the fault algorithm in fault simulation drills. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a technical flow chart of the present invention;

[0036] Figure 2 is a traveling wave propagation path diagram in the present invention;

[0037] Figure 3 is a traveling wave reflection and refraction diagram in the present invention;

[0038] Figure 4 Schematic diagram of single-ended traveling wave ranging in the present invention;

[0039] Figure 5 It is a schematic diagram of double-end traveling wave ranging in the present invention;

[0040] Figure 6 It is a schematic diagram of the Rogowski coil in the present invention;

[0041] Figure 7 This is the training image of the Pix2Pix traveling wave image translation algorithm in the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of 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.

[0043] When a transmission line fails, the fault point will propagate transient voltage and current traveling waves to both ends of the line, and the traveling waves contain rich fault information. Under the determined traveling wave propagation speed, the traveling wave information of the fault point is measured, the transmission time of the reflected wave to the detection point is analyzed, and the distance between the fault point and the detection point is determined based on time and speed. At present, the distance measurement of traveling wave ranging for realizing fault location at home and abroad is mainly divided into early traveling wave method and modern traveling wave method.

[0044] (1) Early traveling wave method

[0045] According to different fault location theories, the early traveling wave method can be divided into three categories: A, B, and C, and different methods have different characteristics.

[0046] ① Type A fault distance measuring device uses the time difference and the speed of the traveling wave after the traveling wave generated by the fault point reaches the bus end and then reflects to the fault point, and then determines the distance of the fault point. However, this method does not solve the problem of distinguishing the reflected wave from the fault point and the reflected wave from the opposite bus end at the fault point, which is difficult to implement.

[0047] ② Type B fault distance measuring device records the time when the traveling wave reaches both ends of the line, and then uses two-end communication to achieve fault distance measurement. This distance measuring device uses the information of the first traveling wave that reaches the bus end after the fault occurs. There is no problem of distinguishing the reflected wave at the fault point and the reflected wave at the opposite bus end at the fault point. The principle is relatively simple, but this method requires communication between the two ends of the line and has high requirements for the time scale.

[0048] ③ After a fault occurs, the C-type fault distance measuring device emits high-voltage high-frequency or DC pulses, and measures the distance based on the time it takes for the high-frequency pulse to travel from the device to the fault point. This type of distance measuring device has a simple principle and high accuracy, but it requires additional components such as a high-frequency pulse signal generator, which is relatively expensive and complicated. In addition, external interference has a great influence on the reflected pulses at the fault point, requiring high-frequency signal processing and carrier channel equipment for all three phases of the transmission line, which has a high investment cost.

[0049] Comparison of the three ranging principles: Type A and Type C ranging principles belong to single-end ranging, which do not require communication between the two ends of the line; while Type B ranging principle belongs to double-end communication, which requires double-end information. Type A and Type B ranging principles are applicable to transient and permanent faults, while Type C ranging principle is only applicable to permanent faults.

[0050] (2) Modern traveling wave method

[0051] Since the mid-1960s, people have conducted a lot of in-depth research on the traveling wave transmission theory of transmission lines proposed in 1926, and have done a lot of work in phase mode transformation, parameter frequency change and transient numerical calculation. This has further deepened the understanding of traveling wave ranging and many related factors. Based on the early type A traveling wave method, the modern traveling wave method has been gradually improved and developed, which mainly includes two aspects: traveling wave correlation method and high-frequency traveling wave method.

[0052] ①Traveling wave correlation method

[0053] The traveling wave correlation method uses the similar waveform and opposite polarity characteristics of the forward voltage traveling wave of the fault point and the reverse voltage traveling wave returned from the fault point to record the time delay △t of the traveling wave in the round trip between the bus and the fault point, thereby realizing traveling wave ranging.

[0054] This method also has the problem of distinguishing the reflected wave from the fault point and the transmitted wave from the opposite busbar end at the fault point. Because in some fault situations there are transmitted waves from the opposite end, they will overlap with the reflected wave from the fault point, which brings great difficulties to the correlation method of ranging.

[0055] ② High frequency traveling wave method

[0056] The high-frequency traveling wave method is different from other traveling wave methods in that it extracts the high-frequency traveling wave component of the voltage or current, then performs digital signal processing, and then performs fault location based on the traveling wave method. This method successfully distinguishes the reflected wave at the fault point and the transmitted wave at the opposite bus end at the fault point based on the reflection characteristics of the bus end at high frequency.

[0057] Table 1 Comparison of different traveling wave ranging methods

[0058]

[0059] (3) Traveling wave ranging method combined with artificial intelligence technology

[0060] Since the traditional traveling wave ranging method still has some problems in practical applications, researchers at home and abroad have been trying to use different methods to improve the reliability and accuracy of traveling wave ranging. In recent years, with the continuous development of intelligent algorithms, more and more scholars have begun to combine intelligent algorithms with traditional traveling wave ranging methods and try to apply them to transmission line fault ranging. Artificial neural networks simulate the neural structure of the human brain and can perform distributed parallel processing on fault traveling wave signals. The reasonable application of neural networks to transmission line fault ranging can greatly simplify the ranging procedure and improve the ranging accuracy.

[0061] At present, with the in-depth research of traveling wave ranging algorithms and the development of electronic technology, traveling wave ranging devices have realized functions such as signal acquisition and high-precision timing, and have been widely used in power systems. However, there is still room for further improvement in ranging reliability and improving ranging accuracy. At present, the further development of traveling wave ranging is mainly affected by the following technical links:

[0062] 1) Accurate acquisition of traveling wave components

[0063] The frequency bandwidth of fault transient traveling waves is distributed in the frequency range of several thousand hertz to several megahertz. In order to accurately capture the complete fault traveling wave information, high requirements are placed on the traveling wave signal measurement and acquisition devices.

[0064] There are many different methods for transient traveling wave acquisition at home and abroad, which can be roughly divided into indirect and direct methods. The indirect method uses a specific transient signal coupler to indirectly obtain high-frequency traveling wave fault signals. Zeng Xiangjun of Changsha University of Science and Technology installed a special traveling wave signal sensor on the capacitive voltage transformer (CVT) grounding wire to obtain the traveling wave transient current signal on the grounding wire, which indirectly reflects the transient traveling wave voltage signal of the transmission line. BC Hydro of Canada uses a small reactor in series on the CVT grounding wire to obtain transient traveling wave signals. Zhang Qingrui of Kunming University of Science and Technology added a data acquisition module on the high-voltage side of the transformer to eliminate the limitation that the sampling rate of the electronic transformer does not exceed 10kHz. The indirect method can use existing devices, has low cost, and has a wide range of application scenarios, but there are problems such as signal distortion and noise interference. The direct method can effectively retain the traveling wave model, but it places high demands on the insulation and acquisition performance of the measuring device.

[0065] 2) Determination of traveling wave velocity

[0066] The selection of traveling wave velocity will have a direct impact on the fault location result. Due to the frequency-varying characteristics of transmission line parameters, the transient traveling wave velocity does not remain constant at a fixed value, but is closely related to the frequency of the traveling wave component. Different frequency traveling waves correspond to different traveling wave velocities.

[0067] In current engineering applications, the speed of traveling waves is basically the speed of light or a fixed value close to the speed of light. This treatment will obviously cause a large error in the ranging results. In order to eliminate the error caused by the uncertainty of the wave speed, industry researchers proposed the use of an online wave speed measurement method. When there is an out-of-area fault, the traveling wave transmission speed can be calculated in real time by extracting the initial traveling wave arrival time on both sides of the non-fault line, but the installation point of the ranging device has a certain influence on the wave speed determination. According to the positive and negative nature of the fault traveling wave, the industry researchers extracted the initial traveling wave and the reflected wave from the mixed fault traveling wave signal, and based on this, proposed a single-end traveling wave fault ranging method whose ranging accuracy is not affected by the wave speed. However, this method is too complicated, the amount of calculation is huge, and it is not easy to identify various reflected waves, so it is not very practical. Industry researchers use the time difference of the fault traveling wave, the reflected wave at the fault point, and the reflected wave of the opposite busbar collected by the substation ranging device to calculate the traveling wave speed, so that the selection of the wave speed has a certain basis, thus avoiding blindness. Therefore, choosing a reasonable traveling wave speed is of great significance to improving positioning accuracy.

[0068] 3) Accurate calibration of the moment when the fault traveling wave reaches the measurement end

[0069] After a transmission line fault occurs, the fault voltage and current traveling wave signals generated will have key points such as amplitude points and frequency extreme points. These key points correspond to the time when the traveling wave signal arrives at the ranging device. The calibration of the arrival time of the fault traveling wave head depends to a large extent on whether the key point detection is accurate and reliable.

[0070] Domestic and foreign researchers have proposed the use of wavelet transform, Hilbert-Huang transform (HHT), mathematical morphology, local feature scale decomposition (LMD) and other methods to determine the arrival time of the traveling wave head. Since the wavelet modulus maximum can correspond to the singular point of the fault traveling wave signal, the arrival time of the traveling wave head can be determined by wavelet transform. However, the wavelet transform fault analysis result is closely related to the type of wavelet basis, sampling rate and decomposition scale. If it cannot be reasonably selected, it is difficult to obtain an ideal ranging result. Some scholars have also tried to introduce HHT into the transmission line traveling wave fault ranging to perform empirical mode decomposition (EMD) on the traveling wave signal, and then use the instantaneous frequency of the first intrinsic mode component (IMF) in the Hilbert transform decomposition result to find the first mutation point of the instantaneous spectrum to improve the accuracy of wave head detection. MoravejZ compared and analyzed the wavelet transform modulus maximum method and the HHT method. The ranging results show that the latter is significantly better than the former, but the endpoint effect of the HHT method will have an adverse effect on the ranging results.

[0071] In summary, the transmission line fault location method based on traveling wave ranging has made great progress and can basically realize the transmission line fault location, but there is still room for improvement in the accuracy of transmission line fault location. In the traveling wave acquisition link, it is still necessary to improve the ability to accurately obtain the traveling wave components and accurately calibrate the arrival time of the traveling wave. In the traveling wave analysis link, it is still necessary to continuously improve the positioning algorithm to improve the accuracy, applicability and reliability of traveling wave ranging.

[0072] An American electric power company has developed a two-terminal distance measuring device using voltage traveling waves, which requires the installation of a special voltage traveling wave measurement and conversion device. It does not have the function of collecting and recording traveling waves and does not have the function of single-terminal traveling wave fault distance measuring.

[0073] An American instrument company has developed a fault distance measurement device using current traveling waves. Although it has been successfully put into trial operation in Scotland, it also lacks the function of collecting and recording traveling waves and does not have the function of single-ended traveling wave fault distance measurement.

[0074] A Canadian hydropower company has developed a double-terminal traveling wave fault location system and installed it in multiple substations in British Columbia. The system indirectly obtains the fault voltage traveling wave signal through a specially developed hybrid device, but the device does not have a waveform recording function.

[0075] In the late 1960s and early 1970s, China began to develop transmission line distance measurement devices. Although some progress has been made in traveling wave fault location technology at home and abroad, there are still some problems in the accuracy and reliability of data processing that have not been solved and need to be improved urgently:

[0076] (1) Interference from various noises on site

[0077] Noise is an inevitable influencing factor on site and an important reason that easily causes ranging failure. The traveling wave fault ranging device is subject to more interference on site. At the same time, the sampling frequency of the traveling wave fault ranging device is relatively high, which makes it easier to receive interference from various types of noise, mainly white noise. The existence of white noise will affect the accurate calibration of the singularity of the traveling wave signal, especially for traveling wave signals with small amplitudes. White noise may submerge the singularity of the traveling wave signal or even the entire traveling wave process, causing the singularity to be unmeasurable or causing large errors.

[0078] How to minimize the interference of noise under strong noise interference and how to accurately calibrate the arrival time of traveling waves under noise interference are important links to ensure the development of traveling wave fault ranging devices towards high reliability and high precision.

[0079] (2) Accurate determination of singular points of weak traveling wave signals

[0080] When the voltage near the fault point crosses zero or a high-resistance ground fault occurs, the amplitude of the traveling wave signal is low and the singularity is weak. Affected by factors such as line dispersion, the amplitude and singularity will be further reduced when it reaches the measurement end. At the same time, the initial mutation point of the wave head is interfered by noise, and the traditional detection method may fail at this time. The accuracy of weak traveling wave signal singular point detection is still an important problem that plagues traveling wave fault ranging. How to strengthen the understanding of this type of traveling wave and improve its recognition and processing capabilities still needs further research.

[0081] (3) Impact of wave velocity uncertainty

[0082] The traveling wave component will be dispersed and attenuated when propagating in the transmission line, and the higher the frequency of the traveling wave component, the faster it will attenuate, while the lower the frequency of the traveling wave component, the slower it will attenuate. After the fault traveling wave is generated, the wave velocity of the traveling wave component that reaches the bus measurement end is strictly speaking the wave velocity of the traveling wave component with the highest frequency and whose amplitude can still be detected after reaching the measurement end. Since this frequency component is directly related to the initial traveling wave spectrum range of the fault and the attenuation on the transmission line, its frequency range is not easy to determine, so the value of the wave velocity is uncertain.

[0083] At present, the speed of traveling waves in transmission lines can be obtained mainly through the following methods: parameter calculation, online measurement and empirical parameters. However, the parameter calculation method relies on the accuracy of the actual line parameters and the determination of the center frequency band of the traveling wave. The online measurement method requires the assistance of the traveling wave signal of the out-of-area fault or reclosing switch. The empirical parameter method often lacks a theoretical basis. Therefore, the use of these methods is not conducive to obtaining accurate wave velocity values. How to develop new ideas and eliminate the influence of wave velocity parameters on traveling wave ranging has important research value.

[0084] (4) Traveling wave fault location problem under special network topology

[0085] The fault traveling waves of T-type lines have more complex refraction and reflection phenomena than conventional lines. The identification of fault branches and the location of fault points need to make full use of the fault traveling wave propagation characteristics unique to the network topology. When the three busbar measurement ends of the T-type line can provide traveling wave recording data, the traditional method will use two-two measurement ends to realize double-end ranging to locate the fault point. Obviously, the algorithm is cumbersome and the comprehensive utilization of resources is low. Strengthening the algorithm research on the comprehensive utilization of the three busbar measurement end traveling wave data will be of great significance to improving the reliability of fault branch identification and the accuracy of fault point location; and when only one measurement end can provide traveling wave recording data, how to realize single-end traveling wave fault ranging under the T-type structure is also a problem worth considering.

[0086] Example:

[0087] ① Traveling wave characteristics

[0088] If the transmission loss of the transmission line is ignored (distributed resistance and ground conductivity are ignored), it can be considered to be composed of a large number of distributed inductances and capacitors. Assuming that the beginning of a line is M and the end is N, and a ground fault occurs at a point P in the middle of the line, it is equivalent to connecting an equivalent power supply at point P, whose voltage is equal to the voltage before the fault at this point, but opposite in direction.

[0089] like Figure 2 As shown in the figure, an electric field is established around the conductor and the adjacent capacitors are charged. The distributed capacitance of the line is charged in turn. This process is like a voltage wave propagating along the line at a certain speed. At the same time, as the capacitors charge and discharge, current will flow through the distributed inductance of the line, and a current wave will also propagate along the line. Therefore, through the above analysis, after a line fault occurs, a voltage wave and a current wave will propagate from the fault point to both ends of the line.

[0090] The speed of the traveling wave is related to the characteristics of the line itself, where L and C are the inductance and capacitance per unit length of the line. The speed of the traveling wave of the line is only related to the properties of its insulating medium, and has nothing to do with the material and cross-sectional area of ​​the conductor. For example, the speed of the traveling wave of the overhead line is 294km / ms, the speed of the traveling wave of the paper insulated cable line is 160km / ms, and the speed of the traveling wave of the cross-linked polyethylene cable is 170km / ms. The speed formula is as follows:

[0091]

[0092] like Figure 3 As shown, the traveling wave will be reflected and refracted at the boundary point where the wave impedance changes. For example, if N in the figure is a busbar and N has several outgoing lines, reflection and refraction will occur at N.

[0093] ②Single-ended traveling wave ranging

[0094] Single-ended traveling wave ranging is to install a measuring device at one end of the line and use the time difference between the first traveling wave measured when the line fails and the reflected second traveling wave to calculate the distance between the measuring point and the fault point. Figure 4 , install the traveling wave measuring equipment at point M. The first traveling wave measured at point M is i1. i1 is reflected twice at point M and the fault point F and is measured again at point M. Then the time difference is twice the distance MF. Therefore, the fault point location is as follows: Figure 4 As shown:

[0095]

[0096] The above consideration is the case where the fault point is relatively close to point M. In fact, if point F is relatively close to point N, the second traveling wave measured should be i5. Assuming that the distance of MN is L, it can be calculated that if the distance between point F and point M is less than L / 2, the second traveling wave is i3, otherwise the second traveling wave is i5. The calculation formulas for these two fault distances are different.

[0097] If there is a line on the upper end of M, i1 will be transmitted, and the transmitted traveling wave will be reflected back at the far end to form i4, which may be measured by the M-end device earlier than i3. As can be seen from the above, the traveling wave measured by single end is more complicated. In some cases, it is difficult to identify the type of the second traveling wave, and the ranging result is wrong.

[0098] ③Dual-end traveling wave ranging

[0099] Double-end traveling wave ranging is to install traveling wave ranging equipment at both ends of the line, and calculate the fault point location by detecting the time difference between the first traveling wave arriving at both ends. Figure 5 As shown in the figure, a fault occurs at point K in the line. The first traveling wave moves from point K to the M and N ends respectively. Two traveling wave ranging devices are installed at M and N to record the time point when the first traveling wave is received respectively. The position of the fault point can be calculated by the following formula:

[0100]

[0101] Where L is the length of line MN, the schematic diagram of double-terminal traveling wave ranging is as follows Figure 5 shown.

[0102] As can be seen from the above, dual-end traveling wave ranging only needs to identify the first traveling wave. The principle is simple and easy to implement. However, this method requires high-precision real-time clocks at both ends. For example, for overhead lines, the wave speed is about 300km / ms, so the time difference of each us will cause an error of 150m. Using GPS timing can reduce the time error to less than 1us, making this algorithm feasible. In addition, dual-end traveling wave ranging needs to communicate with each other or send time information to the same master station to calculate the fault location. In the actual application of current power systems, dual-end traveling wave ranging is generally used as the main method, and single-end traveling wave ranging is used as an auxiliary method.

[0103] ④ Comparison of distance measurement methods

[0104] Through the principle analysis of the two methods, the advantages and disadvantages of the two methods are summarized as follows:

[0105] Single-ended traveling wave ranging method: The advantages are low cost, only a high-stability real-time clock is required, and GPS synchronization is not required. If two traveling waves can be identified, the error is only affected by the accuracy of the clock and is easy to control. The disadvantage is that there are defects in the principle. In some cases, the second traveling wave cannot be identified, resulting in ranging errors.

[0106] Double-end traveling wave ranging method: The advantages are simple principle, high ranging accuracy, general error less than 500m, and wide adaptability; the disadvantages are high cost, and both ends of the device need GPS synchronization; the sampling channel and the CT amplitude-frequency and phase-frequency characteristics require high consistency, which will directly affect the measurement error, and both require communication.

[0107] The difference between the two methods lies in the number of PMUs used. The single-ended method uses only one PMU, and the double-ended method uses two PMUs, which are located at both ends of the transmission line. In addition, whether it is a single-ended or double-ended method, accurate wave head calibration is based on the high sampling frequency of the measuring device. The higher the sampling frequency, the higher the positioning accuracy. From the perspective of the precise requirements for fault location, it is generally required that the fault location error of the transmission line between two transmission towers should not exceed 300-400m. According to the speed v of the traveling wave propagation in the transmission line of 289942.31km / s, its fault location error should be within 300m, so the sampling frequency should be greater than 106Hz (1MHz).

[0108] ⑤GPS timing and synchronization theory

[0109] From the above analysis, it can be seen that traveling wave ranging relies on a high-precision system clock. The clock accuracy needs to reach the nanosecond level, and this timing accuracy must rely on the GPS timing system.

[0110] GPS timing uses the high-precision atomic clocks carried by GPS satellites to generate reference signals and time standards, providing global time services with a timing accuracy of up to 2 billionths of a second. The GPS timing system mainly uses the precise timing characteristics of GPS to achieve unified timing of the system. At any time, the GPS receiver can simultaneously receive signals from 4 to 8 satellites within its operating range, and extract and output two time signals after decoding:

[0111] A) The synchronization error between the pulse leading edge of the pulse signal PPS with a time interval of 1s and the international standard time (Greenwich Mean Time) shall not exceed 1us;

[0112] B) The international standard time and date code corresponding to the leading edge of the PPS pulse output via the serial port.

[0113] There are currently three main methods of time synchronization: hard time synchronization (pulse time synchronization), soft time synchronization (communication message time synchronization) and coded time synchronization (widely used IRIG-B time synchronization).

[0114] Hard time synchronization generally uses second pulse time synchronization. This format of time signal outputs one pulse per second or per minute. Obviously, the clock pulse output does not contain specific time information. However, it can provide the device with a stable and accurate clock reference to the greatest extent, and is not affected by external environments such as temperature and magnetic fields. As long as the satellite signal can be received, it will be effective.

[0115] Soft time synchronization uses communication messages to transmit complete time including year, month, day, hour, minute, second and millisecond. This method has inherent propagation delay error, so it cannot meet the requirements in situations where high accuracy is required.

[0116] The coded time synchronization uses IRIG-B time synchronization, which is divided into two types: modulation and non-modulation. IRIG-B is actually a comprehensive time synchronization solution, because its message contains time information such as seconds, minutes, hours, and dates, and each frame can be divided into 100 code elements, which is equivalent to a pulse synchronization signal.

[0117] In fact, the GPS timing system provides a high-precision time reference in various forms. How to synchronize the system time with the UTC time is an important link that affects the final timing accuracy. The system timing process can be divided into the following three steps:

[0118] A) The device acquires satellite clock data, including basic time information and second pulse and signal.

[0119] B) The first time synchronization system has its own clock based on satellite data.

[0120] C) Compare the pulse-per-second signal with the system's own one-second pulse signal to perform real-time compensation on the system clock.

[0121] ⑥ Rogowski coil sampling theory

[0122] The acquisition of traveling wave waveforms and power frequency waveforms of overhead lines requires the restoration of the original waveforms of the lines to the greatest extent possible, which is a prerequisite for waveform analysis. However, the short-circuit fault current or lightning current of overhead lines has the characteristics of high frequency and large current. Traditional current transformers have low sensitivity, small frequency band, and are easy to saturate. Figure 6 As shown in the figure, the Rogowski coil is an AC current sensor, a hollow ring-shaped coil, available in two types: flexible and rigid. It can be directly put on the conductor to measure the AC current. The Rogowski coil is suitable for measuring AC current in a wide frequency range, has no special requirements for the conductor and size, has a fast instantaneous response capability, and is widely used in situations where traditional current measurement devices such as current transformers cannot be used, for current measurement, especially high-frequency and large current measurement.

[0123] The theoretical basis for measuring current with Rogowski coil is Faraday's law of electromagnetic induction and Ampere's loop law. When the measured current passes through the center of Rogowski coil along the axis, a corresponding magnetic field with intensity H is generated in the volume surrounded by the annular winding. According to Ampere's loop law, we get:

[0124]

[0125] From B=μH, e(t)=d(Φ) / dt, Φ=N∫B·dS, e(t)=M·d i / d t ,have to:

[0126] When the cross section is rectangular, the mutual inductance coefficient M and self-inductance coefficient L are:

[0127] M=μ0Nhln(b / a)2π

[0128] L=μ0N 2 hln(b / a)2π

[0129] In the above formula, H is the magnetic field strength inside the coil, B is the magnetic induction strength inside the coil, μ0 is the vacuum magnetic permeability, N is the number of coil turns, e(t) is the induced voltage at both ends of the coil, a and b are the inner and outer diameters of the coil cross section, respectively, and h is the cross-sectional height.

[0130] It can be seen that when the coil is fixed, M is a fixed value, and the output voltage of the coil is related to d i / d t In other words, the output voltage of the Rogowski coil is proportional to the differential of the measured current. As long as the output passes through the integrator, an output voltage proportional to the primary current can be obtained.

[0131] ⑦Traveling wave image translation algorithm

[0132] In the application of power systems, as a general solution for image-to-image conversion, GANs that are not clear about image conditions still cannot achieve good image translation effects and reach the condition of producing reasonable results. Therefore, this paper chooses Pix2Pix, which is based on the conditional generative adversarial network (cGAN, Conditional Generative Adversarial Nets) framework, as the translation algorithm for traveling wave images. The first main reason for choosing Pix2ix is ​​that Pix2Pix proposes a simple framework that can achieve good results. The second main reason for choosing Pix2ix is ​​that Pix2Pix can produce reasonable results on a variety of problems.

[0133] As shown in the attached picture Figure 7 As shown in Figure 1, after Pix2Pix conversion, the position of the first local modulus maximum in the 1-scale detail component will become more accurate in the image. Through the conversion of pictures and data, this more accurate positioning can lead to a more accurate arrival time of the traveling wave head generated by the fault. Therefore, Pix2Pix can improve the low

[0134] The sampling frequency of the PMU is accurate in locating traveling wave faults.

[0135] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for diagnosing and locating abnormalities in power transmission lines based on high-frequency sampling and fault recording, characterized in that: It includes the following two aspects: a. Through high-frequency sampling and fault recording, the transient traveling wave current of the transmission line lightning fault and non-lightning fault is collected; by comparing and analyzing the transient traveling wave current collected during lightning fault, the nature of lightning shielding and counter-strike faults can be identified; by comparing and analyzing the transient traveling wave current collected during non-lightning fault, and determining the fault interval through the flow direction of the traveling wave current, the fault point can be accurately located within the determined interval through the traveling wave ranging method; b. Aiming at the hidden danger of insulation performance degradation caused by natural aging of transmission lines, the discharge waveform is detected to determine the fault type and actively warn.

2. A method for diagnosing and locating abnormalities in power transmission lines based on high-frequency sampling and fault recording according to claim 1, characterized in that: The high-frequency sampling and fault recording described herein measures the AC current by directly placing a Rogowski coil on the conductor to be measured, thereby restoring the original waveform of the line to the greatest extent, and then collecting the short-circuit fault current and lightning current traveling wave waveform of the overhead line, and using Pix2Pix with a conditional generative adversarial network (cGAN, Conditional Generative Adversarial Nets) as the framework as a translation algorithm for the traveling wave image.

3. A method for diagnosing and locating abnormalities in power transmission lines based on high-frequency sampling and fault recording according to claim 2, characterized in that: The Rogowski coil is an alternating current sensor, a hollow annular coil, and has two types: flexible and rigid.

4. The method for diagnosing and locating abnormalities in power transmission lines based on high-frequency sampling and fault recording according to claim 1, characterized in that: The traveling wave ranging method includes double-end traveling wave ranging and single-end traveling wave ranging, among which double-end traveling wave ranging is the main method. Double-end traveling wave ranging refers to installing traveling wave ranging equipment at both ends of the line, and calculating the fault point location by detecting the time difference of the first traveling wave arriving at both ends; Single-ended traveling wave ranging is supplemented by single-ended traveling wave ranging, which means installing a measuring device at one end of the line and using the time difference between the first traveling wave measured when the line fails and the reflected second traveling wave to calculate the distance between the measuring point and the fault point.

5. A method for diagnosing and locating abnormalities in power transmission lines based on high-frequency sampling and fault recording according to claim 4, characterized in that: The traveling wave ranging method described above relies on a high-precision system clock, and the clock timing accuracy needs to reach the nanosecond level, wherein the timing accuracy relies on the GPS timing system.

6. A method for diagnosing and locating abnormalities in power transmission lines based on high-frequency sampling and fault recording according to claim 5, characterized in that: The GPS timing system uses the high-precision atomic clock carried by GPS satellites to generate reference signals and time standards, providing global time services with a timing accuracy of up to 2 billionths of a second. The GPS timing system mainly uses the precise timing characteristics of GPS to achieve unified timing of the system. The GPS receiver can simultaneously receive signals from 4 to 8 satellites within its field of view at any time, and extract and output two time signals from them after decoding: a. The synchronization error between the pulse leading edge of the pulse signal PPS with a time interval of 1s and the international standard time (Greenwich Mean Time) shall not exceed 1us; b. The international standard time and date code corresponding to the leading edge of the PPS pulse output through the serial port.

7. The method for diagnosing and locating abnormalities in power transmission lines based on high-frequency sampling and fault recording according to claim 5, characterized in that: The system timing process includes the following three steps: a. The device acquires satellite clock data, including basic time information and second pulse and signal; b. The first time synchronization system has its own clock based on satellite data; c. Compare the pulse-per-second signal with the system's own one-second pulse signal to perform real-time compensation on the system clock.

8. The method for diagnosing and locating abnormalities of power transmission lines based on high-frequency sampling and fault recording according to claim 6, characterized in that: There are three main methods of timing: hard timing (pulse timing), soft timing (using communication messages to synchronize timing) and coded timing (IRIG-B timing). a. Hard synchronization generally uses second pulse synchronization. This format of time signal outputs one pulse per second or per minute. The clock pulse output does not contain specific time information, but it can provide the device with a stable and accurate clock reference to the greatest extent. It is not affected by external environments such as temperature and magnetic field. As long as the satellite signal can be received, it will be effective. b. Soft time synchronization uses communication messages to transmit complete time including year, month, day, hour, minute, second and millisecond; c. The coded time synchronization adopts IRIG-B time synchronization, which is divided into two types: modulation and non-modulation. IRIG-B is a comprehensive time synchronization solution. Its message contains time information such as seconds, minutes, hours, and date. At the same time, each frame can be divided into 100 code elements, which is equivalent to a pulse synchronization signal.

Images