Power distribution network double-end traveling wave distance measurement method and system based on CVT and current traveling wave cooperation, product and medium

Abstract

The invention provides a power distribution network double-end traveling wave distance measurement method and system based on CVT and current traveling wave cooperation, a product and a medium, and relates to the technical field of electric variable measurement. CVT secondary voltage traveling waves and TA current traveling waves are synchronously collected, and on the basis of the two paths of signals which are originated from the same physical event (fault) but have independent channels, comparison is carried out by utilizing the wave head time difference of arrival and polarity characteristics of the two paths of signals through a cooperative verification step. Transient interference which is only caused by switching operation or inductive thunder and only shows a remarkable effect on a single physical quantity is filtered out. An actual time difference matrix is constructed by using reliable time data of a plurality of measurement points, and information is improved from a one-dimensional time difference to a multi-dimensional time sequence topology network. Difference calculation and optimization are carried out on an assumed time difference matrix generated near a preliminary positioning result and an actual matrix, and the preliminary positioning result is adjusted by using global and redundant time sequence information, so that deviation introduced by inaccurate single line parameters or small synchronization errors can be smoothed and corrected.

Description

Technical Field

[0001] This application relates to the field of electrical variable measurement technology, and in particular to a method, system, product and medium for measuring travel wave distance at two ends of a distribution network based on the coordination of CVT and current traveling wave. Background Technology

[0002] The timeliness and accuracy of fault location in power distribution networks are crucial for ensuring the supply of electricity to society. Traveling wave ranging is one of the mainstream technologies for rapid fault location. Among them, the double-ended traveling wave method, by comparing the synchronization times of the fault traveling wave arriving at both ends of the line, eliminates the need to identify reflected waves, making it theoretically more reliable than the single-ended method. However, this method places extremely high demands on the precise capture of the traveling wavefront and the accuracy of time synchronization at both ends.

[0003] To capture traveling wave signals, current transformers (CTs) are typically deployed at both ends of the line to collect the current traveling wave. This is because the essence of a fault is a sudden change in current, and the current traveling wave can directly reflect the fault characteristics. However, in practical applications, relying solely on the current traveling wave for distance measurement has many problems. For example, dynamic changes in line parameters and the presence of transition resistance at the fault point can distort the current traveling wave waveform, affecting the accuracy of wavefront arrival time determination.

[0004] The drawback of the aforementioned technologies is that they fail to fully utilize existing measurement resources in the power grid. For example, capacitive voltage transformers (CVTs) are widely distributed in distribution networks, but related technologies argue that their internal structure attenuates high-frequency signals, thus making them unsuitable for traveling wave ranging.

[0005] In summary, the key question is how to collaboratively utilize the traveling waves of CVT voltage and TA current to overcome the problems of low reliability in wavefront detection and susceptibility to interference in positioning accuracy caused by a single signal source. Summary of the Invention

[0006] This application provides a method, system, product, and medium for measuring the distance between two ends of a distribution network based on the coordination of CVT and current traveling wave, which can improve the reliability of wavefront detection and the positioning accuracy.

[0007] Firstly, this application provides a method for determining travel wave distance at both ends of a distribution network based on the coordination of CVT and current traveling wave. This method is applied to distribution network line groups where CVTs are deployed at measurement points at both ends of the line, and the GPS at both ends of the line has a time synchronization mechanism. The method includes: synchronously acquiring the CVT secondary voltage traveling wave and the TA current traveling wave determined by the current transformer under the time synchronization mechanism; the CVT secondary voltage traveling wave is a specific high-frequency oscillation signal generated on the secondary side after a fault traveling wave excites the CVT; wherein the CVT secondary voltage traveling wave is characterized by a CVT high-frequency transient model; and co-verification is performed by comparing the wavefront arrival time difference and polarity characteristics of the CVT secondary voltage traveling wave and the TA current traveling wave. Traveling waves caused by real faults are selected and their arrival times are determined. The arrival time of the traveling waves is the average of the arrival times of traveling waves caused by real faults. The arrival times of two measurement points are selected, the time difference between the two ends is calculated, and the result is substituted into the double-ended traveling wave ranging formula to obtain the preliminary fault distance. Based on the arrival times of multiple measurement points, an actual time difference matrix is ​​constructed. With the preliminary fault distance as the center, a fault search interval is defined on the line model by extending a preset length to both sides. Multiple hypothetical fault points are constructed on the fault search interval, and their corresponding hypothetical time difference matrices are generated. The difference between the actual time difference matrix and each hypothetical time difference matrix is ​​calculated. The hypothetical fault point with the smallest difference is selected as the corrected final fault distance.

[0008] By adopting the above technical solution, the traveling waves of the CVT secondary voltage and the TA current are simultaneously acquired. Based on these two signals originating from the same physical event (fault) but with independent channels, the CVT is utilized, and the CVT secondary voltage traveling wave is considered as the instant the initial traveling wave front arrives. Through a collaborative verification step, their wavefront arrival time difference and polarity characteristics are compared. This filters out transient interference caused only by switching operations or induced lightning, which is significant only in a single physical quantity, ensuring the authenticity of the input data from the source. On this basis, a real time difference matrix is ​​constructed using reliable time data from multiple measurement points, elevating the information from one-dimensional time difference to a multi-dimensional time-series topology network. Finally, by calculating and optimizing the difference between the hypothetical time difference matrix generated near the preliminary positioning result and the actual matrix, and by adjusting the preliminary positioning result using global, redundant time-series information, deviations introduced by inaccurate single line parameters or minor synchronization errors can be effectively smoothed and corrected.

[0009] In conjunction with some embodiments of the first aspect, after the step of defining a fault search interval on the line model with the initial fault distance as the center and extending a preset length to both sides of the line, the method further includes: performing a preset verification on the line based on the actual time difference matrix, wherein the preset verification is to determine that the time difference required for propagation between any two measurement points cannot be greater than the sum of the time differences required for relay propagation through an intermediate measurement point; identifying the lines that fail the verification as contradictory paths and identifying the measurement points on the contradictory paths as suspicious data sources; based on the suspicious data sources as breakpoints, breaking the line model to obtain one or more suspicious line models and one or more other line models, wherein the suspicious line models are line compositions corresponding to measurement points with connection relationships, and the other line models are line compositions corresponding to other measurement points with connection relationships; if the fault search interval is completely covered by other line models, then the step of constructing multiple hypothetical fault points on the fault search interval and generating their corresponding hypothetical time difference matrices is performed.

[0010] By adopting the above technical solution, after obtaining the actual time difference matrix containing time series information from multiple measurement points, instead of directly entering complex optimization calculations, a pre-defined verification step based on physical laws is introduced. Utilizing the common sense of physics—that the direct propagation time of a traveling wave from point A to point C should not be longer than the sum of its propagation times via the intermediate relay point B (A->B->C)—suspicious data sources can be directly identified where serious logical contradictions in the time series data arise due to different propagation speeds. Once such data sources are identified, they are used as breakpoints, decoupling the complete line model into other line models with higher credibility and the suspected line model to be investigated. In this way, subsequent precise positioning can be performed first on a subset of reliable data, reducing computational complexity and the risk of contamination by bad data, thus improving the robustness of the entire positioning algorithm and its reliability in the presence of abnormal data.

[0011] In conjunction with some embodiments of the first aspect, in some embodiments, after the step of breaking the line model based on a suspicious data source as a breakpoint to obtain one or more suspicious line models and one or more other line models, the method further includes: if the fault search interval is not completely covered by other line models, then performing reverse time-series localization on the corresponding other line models and the corresponding suspicious line models to generate sub-candidate fault intervals respectively; combining the sub-candidate fault intervals and updating the fault search interval, and performing the step of constructing multiple hypothetical fault points on the fault search interval and generating their corresponding hypothetical time difference matrices.

[0012] By adopting the above technical solution, if the initial fault search interval falls entirely within other line models composed of reliable data, this high-quality data can be directly used for high-precision matrix matching optimization, avoiding interference from suspicious data. When the search interval intersects with a suspicious data area, reverse time-series localization is performed separately for both reliable other line models and suspicious line models, generating independent sub-candidate fault intervals for each. This step fully leverages the information value within local data clusters. Finally, by combining and updating these independent judgment results, the system can form a more comprehensive and closer-to-the-truth final fault search interval.

[0013] In conjunction with some embodiments of the first aspect, in some embodiments, during the preset verification, two measurement points are connected, and an intermediate measurement point is connected to any measurement point.

[0014] By adopting the above technical solution, the amount of data in the suspicious route model is minimized, while the amount of other route models composed of reliable data is maximized, thereby increasing the possibility of falling into the first scenario.

[0015] In conjunction with some embodiments of the first aspect, in some embodiments, before the step of screening out traveling waves caused by real faults and determining the arrival time of the traveling waves by comparing the arrival time difference and polarity characteristics of the wavefronts of the CVT secondary voltage traveling waves and the TA current traveling waves for collaborative verification, the method further includes: based on the CVT high-frequency transient model, confirming that when the wavefront of the side transient traveling wave arrives, a damped high-frequency oscillation signal with a frequency determined by the parameters of the CVT high-frequency transient model itself is excited on the CVT secondary side; after acquiring the waveform of the damped high-frequency oscillation signal, applying wavelet transform for multi-scale analysis to identify the moment when the maximum value of the signal energy modulus appears at the highest frequency analysis scale; and calibrating the moment as the arrival time of the wavefront of the CVT secondary voltage traveling wave.

[0016] By adopting the above technical solution, instead of attempting to recover the severely distorted primary traveling wave waveform from the secondary side of the CVT, this solution utilizes its inherent physical characteristics: the primary traveling wave will inevitably excite a high-frequency oscillation signal with a fixed characteristic frequency on the secondary side. This solution treats this distortion product as a reliable beacon of the traveling wave's arrival. By applying wavelet transform, a powerful time-frequency analysis tool, the signal can be decomposed at multiple scales. Especially at the highest frequency analysis scale, the oscillation start point excited by the traveling wave will manifest as a modulus maxima of energy, the timing of which can be accurately identified. This transforms the inherent defects of the CVT into usable stable characteristics, successfully upgrading existing CVT equipment deployed in the distribution network into high-precision traveling wave sensors on-site without any modifications. This reduces the deployment cost of the traveling wave ranging system and broadens its applicability.

[0017] In conjunction with some embodiments of the first aspect, in some embodiments, before the step of screening out traveling waves caused by real faults and determining the arrival time of the traveling waves by comparing the wavefront arrival time difference and polarity characteristics of the CVT secondary voltage traveling wave and the TA current traveling wave for joint verification, the method further includes: performing phase-mode separation on the TA current traveling wave to extract the line-mode component; wherein the phase-mode separation parameters are: In the formula, For zero modulus components, , For line mode components, , , , These are the three-phase currents of the TA current traveling wave; the modulus maxima method is applied to the line mode component to detect the moment when the amplitude first changes abruptly, and this moment is determined as the wavefront arrival time of the TA current traveling wave.

[0018] By employing the above technical solution, when processing traditional current traveling waves (CTs), the original three-phase A, B, and C currents are decoupled, and the line-mode components that most directly reflect the transient process of phase-to-phase or ground faults are extracted. This actively suppresses zero-sequence components and background noise interference generated by system imbalance or non-faulty phase induction. This makes the transient characteristics of the fault pure in the processed single-line-mode signal. Based on this, the modulus maxima method is then applied to detect the first abrupt change in amplitude, improving accuracy and anti-interference capability compared to direct detection on the original coupled signal.

[0019] In conjunction with some embodiments of the first aspect, in some embodiments, the step of co-verifying the wavefront arrival time difference and polarity characteristics of the CVT secondary voltage traveling wave and the TA current traveling wave specifically includes: comparing the wavefront arrival time of the CVT secondary voltage traveling wave with the wavefront arrival time of the TA current traveling wave; when the time difference is less than a preset time threshold, determining the polarity relationship of the initial half-wave of the CVT secondary voltage traveling wave and the TA current traveling wave, and comparing it with a pre-established fault event polarity feature library; excluding traveling waves whose polarity feature and fault characteristic difference is greater than a preset characteristic threshold.

[0020] By employing the aforementioned technical solution, and by determining whether the arrival time difference between the wavefronts of the CVT voltage traveling wave and the TA current traveling wave is less than a preset threshold, the solution leverages the physical principle that a real fault will inevitably generate disturbances in both voltage and current simultaneously, instantly eliminating interference caused by a single physical quantity. Subsequently, for candidate events that pass the time verification, the solution delves further into the physical level by comparing the polarity relationship of the initial half-waves of voltage and current. Different types of events (such as forward faults, reverse faults, and switch closing) exhibit distinctly different electromagnetic transient responses. By comparing these responses with a pre-established fault event polarity feature library, non-fault events such as switch operations can be accurately eliminated.

[0021] Secondly, this application provides a CVT and current traveling wave coordinated dual-end traveling wave ranging system for distribution networks. The CVT and current traveling wave coordinated dual-end traveling wave ranging system for distribution networks includes: one or more processors and a memory; the memory is coupled to one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and one or more processors call the computer instructions to cause the CVT and current traveling wave coordinated dual-end traveling wave ranging system for distribution networks to perform the method described in the first aspect and any possible implementation thereof.

[0022] Thirdly, this application provides a computer program product containing instructions that, when the computer program product is run on a CVT-and-current traveling wave coordinated distribution network dual-end traveling wave ranging system, causes the CVT-and-current traveling wave coordinated distribution network dual-end traveling wave ranging system to execute the method described in the first aspect and any possible implementation thereof.

[0023] Fourthly, this application provides a computer-readable storage medium including instructions that, when executed on a CVT-and-current traveling wave-based distribution network dual-end traveling wave ranging system, cause the CVT-and-current traveling wave-based distribution network dual-end traveling wave ranging system to perform the method described in the first aspect and any possible implementation thereof.

[0024] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. Simultaneously acquire the traveling waves of the CVT secondary voltage and the TA current. Based on these two signals originating from the same physical event (fault) but with independent channels, utilize the CVT and treat the CVT secondary voltage traveling wave as the instant the initial traveling wave front arrives. Through a collaborative verification step, compare their wavefront arrival time difference and polarity characteristics. This filters out transient interference caused only by switching operations or induced lightning, which is significant only in a single physical quantity, ensuring the authenticity of the input data from the source. On this basis, construct an actual time difference matrix using reliable time data from multiple measurement points, elevating the information from one-dimensional time difference to a multi-dimensional time-series topology network. Finally, by calculating and optimizing the difference between the hypothetical time difference matrix generated near the initial location result and the actual matrix, and by adjusting the initial location result using global, redundant time-series information, effectively smoothing and correcting deviations introduced by inaccurate single line parameters or minor synchronization errors.

[0025] 2. After obtaining the actual time difference matrix containing time series information from multiple measurement points, instead of directly entering complex optimization calculations, a pre-defined verification step based on physical laws was introduced. Utilizing common physics, namely that the direct propagation time of a traveling wave from point A to point C should not be longer than the sum of its propagation times via the intermediate relay point B (A->B->C), suspicious data sources can be directly identified where serious logical contradictions in the time series data arise due to different propagation speeds. Once such data sources are identified, they are used as breakpoints, decoupling the complete line model into other, more reliable line models and the suspicious line models to be investigated. In this way, subsequent precise positioning can be performed on a reliable subset of data first, reducing computational complexity and the risk of contamination by bad data, thus improving the robustness of the entire positioning algorithm and its reliability in the presence of abnormal data.

[0026] 3. If the initial fault search interval falls entirely within other line models composed of reliable data, this high-quality data can be directly used for high-precision matrix matching optimization, avoiding interference from suspicious data. When the search interval intersects with a suspicious data area, reverse time-series localization is performed separately for both reliable other line models and suspicious line models, generating independent sub-candidate fault intervals for each. This step fully leverages the information value within local data clusters. Finally, by combining and updating these independent judgment results, the system can form a more comprehensive and closer-to-the-truth final fault search interval. Attached Figure Description

[0027] Figure 1 This is a flowchart illustrating the dual-end traveling wave ranging method for distribution networks based on CVT and current traveling wave coordination in the embodiments of this application. Figure 2 This is the simulated CVT primary side voltage waveform after a fault. Figure 3 This is a simulated CVT secondary voltage waveform after a fault. Figure 4 yes Figure 3 A magnified view of a portion of the image; Figure 5 This is a schematic diagram of the structure of a CVT high-frequency transient model; Figure 6 This is a schematic diagram of the wave trap structure; Figure 7 This is a schematic diagram of a CVT structure; Figure 8 This is a schematic diagram of another type of CVT structure; Figure 9 This is another flowchart illustrating the dual-end traveling wave ranging method for distribution networks based on CVT and current traveling wave coordination in the embodiments of this application; Figure 10 This is an exemplary hardware structure diagram of a distribution network dual-end traveling wave ranging system based on CVT and current traveling wave coordination in the embodiments of this application. Detailed Implementation

[0028] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions a, an, said, the above, and this are intended to also include the plural expressions unless the context clearly indicates otherwise. It should also be understood that the terms used in this application refer to and / or include any or all possible combinations of one or more of the listed items.

[0029] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0030] Please see Figure 1 , Figure 1 This is a flowchart illustrating the dual-end traveling wave ranging method for distribution networks based on CVT and current traveling wave coordination in the embodiments of this application. A dual-end traveling wave ranging method for distribution networks based on CVT and current traveling wave coordination is applied to distribution network line groups. CVTs are deployed at measurement points at both ends of the line group, and the GPS devices at both ends of the line have a time synchronization mechanism. The method includes: S101. Under the time synchronization mechanism, the CVT secondary voltage traveling wave and the TA current traveling wave determined by the current transformer are synchronously acquired; the CVT secondary voltage traveling wave is a specific high-frequency oscillation signal generated on the secondary side after the fault traveling wave excites the CVT; wherein, the CVT secondary voltage traveling wave is characterized by the CVT high-frequency transient model. The time synchronization mechanism refers to using the Global Positioning System (GPS) or other methods to provide a unified time reference with nanosecond-level accuracy for all measurement points along the line, ensuring that the timestamps of signals collected at different locations can be compared.

[0031] It should be noted that, given the inherent defects of CVTs revealed in the background technology, the raw data generated in actual operation is not only enormous in quantity, but also generally suffers from problems such as low signal-to-noise ratio and a large amount of non-fault interference, resulting in inaccurate signals.

[0032] Therefore, in this technical solution, the CVT secondary voltage traveling wave is characterized by the CVT high-frequency transient model. Instead, it is a signal identification and reconstruction process based on model matching. That is, a piece of actual data collected from the CVT is only confirmed as a valid CVT secondary voltage traveling wave if its waveform characteristics are highly consistent with or similar to the theoretical waveform characteristics predicted by the CVT high-frequency transient model.

[0033] This screening process can be implemented in the following two ways: The theoretical waveform generated by the CVT high-frequency transient model is used as a feature template to continuously monitor the real-time data stream output by the CVT. When a waveform segment highly similar to the template appears in the data stream, the system captures it immediately as a valid traveling wave event.

[0034] First, a segment of raw CVT data is pre-acquired and stored. Then, the stored data is scanned and compared using the aforementioned feature templates, and all waveform segments that meet the characteristics are selected and extracted as the CVT secondary voltage traveling wave.

[0035] It should be noted that the above comparison or matching does not require a 100% perfect match. In practical applications, the similarity between the actual data segment and the theoretical waveform is usually calculated. When the similarity exceeds a preset threshold, the match is considered successful.

[0036] Furthermore, for CVT secondary voltage traveling waves that have been successfully matched but have some missing data or severe noise pollution, the corresponding theoretical waveform generated by the CVT high-frequency transient model can be used for interpolation or signal reconstruction to obtain a more complete and clean waveform for subsequent high-precision analysis.

[0037] The CVT high-frequency transient model is a simulation model of the circuit structure. In an exemplary embodiment, two CVT devices are connected to both ends of the line. (See reference...) Figure 5 , Figure 5 This is a schematic diagram of the structure of a CVT high-frequency transient model; where Figure 5 The wave traps in the middle can be referenced Figure 6 , Figure 6 This is a schematic diagram of the wave trap structure; Figure 5 The CVT in the middle can be referenced Figure 7 and Figure 8 , Figure 7 This is a schematic diagram of a CVT structure; Figure 8 This is a schematic diagram of another type of CVT.

[0038] The theoretical basis of this application is that the secondary side waveform of the CVT can reflect the instant when the initial traveling wave front is reached. The specific simulation and derivation are as follows: The CVT secondary-side transmission was calculated using ATP software under various fault types, fault locations, and fault phase angles. Simulation results show that the CVT primary and secondary-side signals exhibit some common characteristics under various fault conditions.

[0039] Depend on Figure 2 , 3 As can be seen from 4, Figure 2 This is the simulated CVT primary side voltage waveform after a fault. Figure 3 This is a simulated CVT secondary voltage waveform after a fault. Figure 4 yes Figure 3 The enlarged view shows that the waveforms of each wavefront arriving on the primary side of the CVT are clearly visible, while the primary wavefronts, after transmission, manifest as a decaying high-frequency oscillation on the secondary side, indicating a fundamental change in the signal. The high-frequency oscillation on the CVT secondary side is caused by the charging and discharging of the energy storage elements within the CVT's equivalent circuit. Ideally, the charging and discharging of these elements does not result in energy loss; however, the resistive elements in the circuit consume energy during oscillation, leading to its decay.

[0040] In traveling wave localization, besides understanding the variation characteristics of the CVT secondary traveling wave signal, another problem to be solved is determining the characteristic parameters of the secondary traveling wave signal. Based on the simulation results above, after the primary traveling wave signal passes through the CVT, the secondary signal becomes a high-frequency oscillating signal, with its main characteristic parameter being the oscillation frequency of the main frequency component. Therefore, in the simulation calculations, the oscillation frequency of the secondary traveling wave signal under different fault conditions was further analyzed and calculated. The specific results are shown in Table 1 below: Table 1. Main frequencies of CVT secondary side oscillation under different fault conditions The above simulation calculations simulated various fault types, and each fault type was simulated at nine different fault points. Therefore, the simulation results are universal.

[0041] Theoretically, the traveling wavefront can be approximated as a step signal, and the traveling wavefront can be clearly detected on the primary side (please refer to...). Figure 4 The high-frequency oscillation on the secondary side is the step response of the CVT.

[0042] The calculation formula is: in, This is the input voltage for the CVT. This refers to the CVT output voltage. The pass function for CVT.

[0043] Because the CVT input voltage (wavefront) is a step function, after the wavefront arrives, the above equation can be rewritten as: Where A is the intensity of the wavefront.

[0044] Therefore, the wavefront on the primary side of a CVT generally only affects the oscillation amplitude of the secondary side signal, while the frequency and decay rate of the secondary side oscillation are mainly determined by the CVT's transfer function. As long as the structure and component parameters of the CVT model remain unchanged, the frequency of its secondary side high-frequency oscillation signal should, in principle, be the same.

[0045] Therefore, under the 27 different simulation conditions mentioned above, the main frequency components of the high-frequency oscillation on the secondary side of the CVT should theoretically be completely equal. In the simulation calculations, a sampling rate of 5MHz was used, and the ATP time scale was accurate to 1µs. Due to the precision limitations, the obtained main frequency component results were not entirely identical, resulting in some errors. Based on the causes of these errors, to minimize them, the average of all obtained frequency results was taken, yielding a frequency of 1.22MHz.

[0046] To verify the simulation results, a CVT high-frequency transient model was used (please refer to...). Figure 5 and Figure 8 The transfer function of the circuit was theoretically analyzed for the frequency of the dominant frequency component. To reduce computational complexity, the circuit structure was appropriately simplified during the derivation of the transfer function, omitting branches with minor impact.

[0047] According to modern control theory, the transient oscillation frequency can be deduced from the poles of the transfer function. The poles of the transfer function are generated when the denominator is zero. Therefore, when studying the poles of the transfer function, it is only necessary to derive the expression for the denominator. The denominator M of the CVT transfer function is as follows: in When M=0, the transfer function has poles, and the roots of the equation when M=0 are: Therefore, the main frequency of the CVT secondary side oscillation can be calculated to be 1.151MHz. The frequency of the high-frequency oscillation is basically the same as the frequency calculated in the simulation, indicating that the simulation result is correct.

[0048] S102. By comparing the arrival time difference and polarity characteristics of the CVT secondary voltage traveling wave and the TA current traveling wave, the traveling waves caused by real faults are screened out and the arrival time of the traveling waves is determined. The arrival time of the traveling waves is the average of the arrival times of the traveling waves caused by real faults. Among them, the wavefront arrival time difference refers to the time difference between the characteristic moment extracted from the CVT secondary voltage traveling wave and the characteristic moment extracted from the TA current traveling wave at the same measurement point; the polarity characteristic refers to the amplitude direction (positive or negative) of the initial half-wave of the traveling wave signal, which reflects the physical relationship between voltage drop and current surge when a fault occurs.

[0049] Specifically, for each pair of CVT and TA waveforms acquired in S101, a time consistency check is first performed: the start time of the transient event is extracted from the voltage and current waveforms respectively, and the time difference between the two is calculated. Since a real line fault will inevitably cause a sudden change in voltage and current simultaneously, these two moments should be very close (on the order of microseconds). If the time difference exceeds a preset small threshold, the event is considered to be likely caused by interference that only affects a single electrical quantity (such as lightning induction, which mainly affects voltage) and should be discarded. Secondly, for events that pass the time check, the algorithm further analyzes the consistency of their physical characteristics. For example, for a single-phase ground fault, the fault phase voltage will drop (initial polarity is negative), while the fault current will surge (initial polarity is positive or negative, depending on the current reference direction). The algorithm compares the observed voltage-current polarity combination with a pre-established polarity feature library containing various typical fault physical characteristics. Only when the observed polarity relationship matches the real fault characteristics defined in the library will the event be finally confirmed as a traveling wave caused by a real fault. After passing through this double firewall, the arrival times of the two signals are averaged to obtain a more reliable final traveling wave arrival time that smooths out minor measurement errors.

[0050] In some embodiments, step S102 specifically includes: S1021. Compare the arrival time of the wavefront of the CVT secondary voltage traveling wave with the arrival time of the wavefront of the TA current traveling wave. S1022. When the time difference is less than the preset time threshold, determine the polarity relationship between the CVT secondary voltage traveling wave and the initial half-wave of the TA current traveling wave, and compare it with the pre-established fault event polarity feature library. Among them, the time difference refers to Δt calculated by S1021; the preset time threshold is a very small time window (e.g., several microseconds) set according to actual system errors (such as differences in sensor response delay, sampling clock synchronization errors, etc.) to tolerate reasonable small differences between t_v and t_i; the polarity relationship of the initial half-wave refers to the combination relationship formed by the direction (positive or negative) of the first peak or trough of the voltage signal and the current signal after the traveling wave arrives, for example (voltage is negative, current is positive); the fault event polarity feature library is a database established in advance through theoretical analysis or simulation, which stores the initial polarity relationship of standard voltage and current traveling waves corresponding to different fault types (such as single-phase grounding, phase-to-phase short circuit, etc.) and fault locations (upstream or downstream).

[0051] Specifically, once a strong temporal correlation is confirmed, the algorithm proceeds to a deeper level of physical characteristic analysis. According to traveling wave theory, the location of the fault point relative to the measurement point, as well as the type of fault, determines the specific relationship between the initial polarities of the voltage and current traveling waves observed at the measurement point. For example, for a single-phase ground fault occurring downstream of the measurement point, the voltage traveling wave typically exhibits negative polarity (voltage drop), while the current traveling wave is positive polarity (current surge). The algorithm returns to the original acquired waveform data and analyzes the initial change directions of the CVT secondary voltage signal (oscillation envelope) and the TA line-mode current signal near times t_v and t_i, thereby obtaining a measured polarity pair, such as P_measured = (-, +). Then, the algorithm queries a fault event polarity feature library, which stores standard patterns such as P_fault1 = (-, +) and P_fault2 = (+, -). By matching P_measured with various P_faults in the library, it determines whether the characteristics of the current event conform to the physical model of a known fault.

[0052] By effectively distinguishing between real faults and non-fault transient events, the method can accurately identify events that appear to be faults but are actually normal operations or non-fault disturbances, thereby improving the startup reliability of the traveling wave ranging method and avoiding the trouble caused to scheduling operations by erroneous actions.

[0053] S1023. Exclude traveling waves whose polarity characteristics and fault characteristics differ from a preset characteristic threshold.

[0054] As can be seen, by determining whether the arrival time difference between the wavefronts of the CVT voltage traveling wave and the TA current traveling wave is less than a preset threshold, the system leverages the physical principle that a real fault will inevitably cause disturbances in both voltage and current simultaneously, instantly eliminating interference caused by a single physical quantity. Next, for candidate events that pass the time verification, the scheme delves further into the physical level by comparing the polarity relationship of the initial half-waves of voltage and current. Different types of events (such as forward faults, reverse faults, and switch closing) exhibit distinctly different electromagnetic transient responses. By comparing these responses with a pre-established fault event polarity feature library, non-fault events such as switch operations can be accurately eliminated.

[0055] In some embodiments, before step S102, the method further includes: S201, based on the CVT high-frequency transient model, confirming that when the transient traveling wave front arrives on the primary side, a damped high-frequency oscillation signal with a frequency determined by the parameters of the CVT high-frequency transient model itself is excited on the secondary side of the CVT. The primary side refers to the end of the CVT connected to the high-voltage transmission line; the secondary side refers to the end of the CVT that outputs a low-voltage signal and is connected to the measurement and protection device; the attenuated high-frequency oscillation signal refers to a specific waveform that has a large amplitude in the initial stage, then the amplitude decreases rapidly according to an exponential law, and oscillates at a certain high frequency.

[0056] Specifically, when a line fault occurs, generating a steep traveling wave front (which can be imagined as a very short voltage pulse) that propagates to the measurement point and impacts the primary side of the CVT, the CVT cannot perfectly transmit this pulse pattern to the secondary side like an ideal transformer. This is because the internal structure of the CVT (including the capacitive voltage divider, intermediate transformer, compensating reactor, etc.) behaves as a complex RLC oscillating circuit at high frequencies, as described by the CVT high-frequency transient model. This steep traveling wave front, like a hammer striking a tuning fork, provides an impulsive excitation to this oscillating circuit. As a result, the secondary side of the CVT does not output a clean pulse, but instead generates a decaying high-frequency oscillating signal with a specific inherent frequency, determined by the CVT's own structural characteristics (i.e., the L and C parameters of its high-frequency transient model). The start time of this signal precisely corresponds to the moment when the original traveling wave front arrives at the primary side. Therefore, to accurately capture the arrival time of the voltage traveling wave, one should not look for the severely distorted original wave front, but rather capture the start time of this characteristic oscillating signal.

[0057] S202. After acquiring the waveform of the attenuated high-frequency oscillation signal, wavelet transform is applied to perform multi-scale analysis to identify the moment when the signal energy modulus maximum occurs at the highest frequency analysis scale. Among them, wavelet transform is an advanced time-frequency analysis tool for signals. It can provide information about signals in both time and frequency, and is particularly good at detecting instantaneous and abrupt components in signals. Multi-scale analysis is the core feature of wavelet transform, which refers to decomposing the signal into a series of detailed levels of different frequencies (scales) for observation and analysis. High frequencies correspond to small scales, and low frequencies correspond to large scales. The highest frequency analysis scale is the finest scale in wavelet analysis, which is most sensitive to the fastest and steepest changes in the signal. The signal energy modulus maxima refers to the point in the wavelet transform result where the absolute value (modulus) of the wavelet coefficients reaches a local peak. Mathematically, it precisely corresponds to the location of singular points (such as abrupt changes and spikes) in the original signal.

[0058] Specifically, after acquiring the waveform containing attenuated high-frequency oscillations, this step performs the following operations: The algorithm first performs a wavelet transform on this digital waveform signal. This transform process decomposes the original one-dimensional time-series signal into a two-dimensional time-frequency graph, showing the intensity of different frequency components at different time points. Since the oscillation signal's onset point is a mathematical singularity point transitioning from stationary to violent oscillation, this abrupt change produces the strongest response at the highest frequency analysis scale of the wavelet transform. At this scale, the signal energy (characterized by the modulus of the wavelet coefficients) forms a very sharp local maximum at this abrupt change point, i.e., a modulus maxima. Therefore, the algorithm only needs to scan the wavelet coefficient modulus sequence at the highest frequency analysis scale to find the first significant peak point. The time coordinate corresponding to this peak point is the precise onset time of the oscillation signal.

[0059] S203. The time is calibrated as the arrival time of the wavefront of the CVT secondary voltage traveling wave.

[0060] Specifically, this step is a simple assignment operation, but it is of great significance. It formally and definitively defines the precise moment (the moment when the modulus maxima appears) obtained through complex wavelet analysis in S202 as the arrival time of the wavefront of the CVT secondary voltage traveling wave. This means that, through the mechanism analysis in S201 and the signal processing in S202, the true arrival time of the original primary side traveling wave has been successfully retrieved from the signal heavily contaminated by the CVT. This timestamp will then be sent to step S102 for comparison and co-verification with the arrival time of the current traveling wave obtained through other methods, thus forming a key source of information for the voltage channel in the entire ranging method.

[0061] As can be seen, instead of attempting to recover the severely distorted primary traveling wave waveform from the secondary side of the CVT, this approach utilizes its inherent physical properties: the primary traveling wave will inevitably excite a high-frequency oscillation signal with a fixed characteristic frequency on the secondary side. This scheme treats this distortion product as a reliable beacon of the traveling wave's arrival. By applying wavelet transform, a powerful time-frequency analysis tool, the signal can be decomposed at multiple scales. Especially at the highest frequency analysis scale, the oscillation start point excited by the traveling wave will manifest as a modulus maxima of energy, the timing of which can be accurately identified. This transforms the inherent defects of the CVT into usable stable characteristics, successfully upgrading existing CVT equipment deployed in the distribution network into high-precision traveling wave sensors without any modifications. This reduces the deployment cost of the traveling wave ranging system and broadens its applicability.

[0062] In some embodiments, before step S102, in step S301, phase mode separation is performed on the traveling wave of the TA current to extract the line mode component. In the formula, For zero modulus components, , For line mode components, , , , These are the three-phase currents of the traveling wave of the current transformer (TA). Phase-mode separation, also known as the symmetrical component method or modulus transformation, is a power system analysis technique used to decompose asymmetrical three-phase electrical quantities (such as three-phase currents a, b, and c) into several independent, symmetrical components with clear physical meaning; zero-mode components ( The term "line modulus component" refers to the component of the three-phase current that is in the same direction and has the same amplitude. It is mainly related to grounding faults, propagates along the phase lines, and forms a loop through the earth or overhead ground wire; the line modulus component ( , The term "traveling wave" refers to the traveling wave components that mainly propagate between phase conductors. They have different phase sequences, their propagation speed is close to the speed of light, they are less affected by the ground, and their waveform distortion is also small.

[0063] Specifically, traveling waves generated when a power line fault occurs contain multiple propagation modes (i.e., moduli). Zero-mode traveling waves, needing to return through the ground, propagate slowly and are greatly affected by unstable factors such as soil resistivity, resulting in severe waveform attenuation and distortion, making them unsuitable for precise timing. In contrast, line-mode traveling waves primarily propagate between conductors, exhibiting a stable path, constant speed (approaching the speed of light), and good waveform preservation. Therefore, to obtain the clearest and most reliable wavefront signal, the mixed original three-phase currents must be processed to remove interfering zero-mode components, retaining only the high-quality line-mode components for subsequent analysis. This step involves applying the given Clarke transform (or a similar phase-mode transformation matrix) to perform a linear combination operation on the three-phase current waveforms acquired in the time domain at each sampling point, thereby calculating the corresponding zero-mode component and two line-mode components. After the transformation, subsequent steps can focus only on these two signals containing purer traveling wavefront information.

[0064] S302. Apply the modulus maxima method to the linear modulus component to detect the moment when the amplitude first changes abruptly, and determine it as the arrival time of the wavefront of the TA current traveling wave.

[0065] The application of the modulus maxima method here is similar to the principle in S202. It usually refers to the modulus maxima detection method based on wavelet transform, which is used to accurately locate the abrupt change point of the signal. The moment when the amplitude first changes abruptly refers to the first time point in the online modulus component signal waveform that is caused by the traveling wave front and jumps sharply from the stable or power frequency background.

[0066] Specifically, after obtaining one or two line-mode component waveforms in S301, the goal of this step is to find the precise moment of arrival of the traveling wave in this waveform. Although the waveform of the line-mode component is clearer than that of the original phase current, it may still be affected by background noise and power frequency fluctuations. Therefore, this step employs an advanced technique similar to that used in S202 for processing CVT signals—the modulus maxima method based on wavelet transform. The algorithm performs wavelet transform on the line-mode component signal and searches for the maxima of the wavelet coefficient modulus at the highest frequency analysis scale. Since the traveling wave front is a typical abrupt change (singularity), it will inevitably produce a significant modulus maxima at the highest frequency scale of the wavelet transform. The algorithm searches along the time axis and determines the moment corresponding to the first (i.e., the earliest) modulus maxima detected as the arrival time of the TA current traveling wave front. This moment is the final judgment result of the current channel on the arrival time of the fault traveling wave.

[0067] It is evident that in processing traditional current traveling waves (CTs), by decoupling the original A, B, and C phase currents and extracting the line-mode components that most directly reflect the transient process of phase-to-phase or ground faults, the zero-sequence components and background noise interference induced by system imbalance or non-faulty phases are actively suppressed. This makes the transient characteristics of the fault pure in the processed single-line-mode signal. Based on this, applying the modulus maxima method to detect the first abrupt change in amplitude improves accuracy and anti-interference capability compared to direct detection on the original coupled signal.

[0068] S103. Select the arrival times of two measurement points, calculate the time difference between the two ends and substitute it into the double-end traveling wave ranging formula to obtain the preliminary fault distance; Among them, the two measurement points usually refer to the measurement points at the two ends of the line section to be measured, such as the substation at the beginning of the line and the switching station at the end; the double-end traveling wave ranging formula is a classic positioning algorithm, the basic form of which is D=(Lv*Δt) / 2, where D is the distance from the fault point to one of the measurement points, L is the total length of the line between the two measurement points, v is the traveling wave propagation speed, and Δt is the time difference between the traveling wave arriving at the two measurement points.

[0069] Specifically, after S102 filters and determines the arrival times of traveling waves caused by real faults at multiple measurement points on the line, the two most critical points are selected first, usually the beginning and end of the line under test (denoted as points M and N). Let the time for the traveling wave to arrive at point M be t_M, and the time to arrive at point N be t_N. The system calculates the time difference between the two ends Δt = t_M - t_N. Simultaneously, the system reads the total line length L and the average propagation velocity v of the traveling wave from the pre-stored line parameter database. Then, these three parameters (L, v, Δt) are substituted into the standard two-end traveling wave ranging formula to calculate a preliminary fault distance value. For example, the distance from the fault point to point M is calculated. This result is considered a preliminary distance because it relies on idealized parameters such as the average wave velocity and may contain some error. Its main purpose is to provide an initial anchor point for subsequent refined positioning.

[0070] S104. Construct the actual time difference matrix based on the arrival times of multiple measurement points; Among them, multiple measurement points refer to measurement points that may be deployed in the middle of the line in addition to the two ends of the line, with a total number of greater than or equal to 2; the actual time difference matrix (denoted as ΔT_real) is a square matrix whose dimension is equal to the total number of measurement points N. The element ΔT_real(i,j) in the matrix represents the difference between the actual measurement time of the traveling wave arriving at measurement point i and measurement point j, i.e. t_i-t_j.

[0071] Specifically, assume that N measurement points (P1, P2, ..., PN) are deployed on the distribution network line, and the reliable arrival time of the traveling wave (t1, t2, ..., tN) for each measurement point has been obtained through S102. The goal of this step is to transform this time information into a mathematical object that reflects the global temporal topology. The system creates an N×N matrix, namely the actual time difference matrix ΔT_real. The diagonal elements ΔT_real(i, i) of this matrix are all 0. For the off-diagonal elements ΔT_real(i, j), their values ​​are calculated as ti-tj. For example, ΔT_real(1, 3) represents the time difference between the traveling wave arriving at P1 earlier or later than arriving at P3. This matrix completely encapsulates all available temporal information of the fault traveling wave propagating to the entire monitoring network. It includes not only the time difference between the beginning and end points, but also the crossover time differences between all intermediate nodes and between intermediate nodes and endpoints. This matrix becomes the core input data for the subsequent precise location algorithm.

[0072] S105. Using the initial fault distance as the center, and extending a preset length to both sides of the line, define the fault search interval on the line model. The line model refers to a digital representation of the line topology and parameters, including information such as the line length, direction, branching, and electrical parameters (such as wave speed) of each segment; the fault search interval refers to a finite length of line defined on the line model, and subsequent precise location calculations will only be performed within this interval, rather than the entire line.

[0073] Specifically, after calculating the preliminary fault distance (e.g., distance from point M, D_preliminary kilometers) in S103, the system recognizes that this value may contain errors. To ensure that the actual fault point is included without having to search the entire line, this step adopts a compromise strategy. It extends a predetermined length (e.g., ±500 meters) to both sides of the digitized line model, centered on this preliminary location point. This creates a clearly defined, finite-length fault search interval on the line. The length of this interval is empirically determined; it needs to be large enough to cover the maximum possible error of the preliminary location, but as small as possible to reduce the amount of subsequent calculations. All subsequent precise location operations will focus on this significantly reduced interval.

[0074] S106. Construct multiple hypothetical fault points on the fault search interval and generate their corresponding hypothetical time difference matrix; Specifically, the system first discretizes the fault search interval defined in S105 with a very small step size (e.g., 1 meter), generating hundreds or thousands of hypothetical fault points. For each hypothetical fault point (e.g., the k-th point, located at x_k), the system performs a forward simulation calculation: assuming the fault occurs at point x_k, based on the precise distances from each measurement point in the line model to x_k and the wave speed of each line segment, the theoretical time t_i_hyp(k) required for the traveling wave to propagate from x_k to each measurement point Pi can be calculated. Subsequently, using these theoretically calculated arrival times, following the exact same method as in S104, an N×N theoretical time difference matrix uniquely corresponding to the hypothetical fault point x_k is constructed, i.e., the hypothetical time difference matrix ΔT_hyp(k). This process is repeated for every hypothetical point within the search interval, ultimately generating a large set of hypothetical time difference matrices, each matrix representing a probability of a fault location.

[0075] S107. Calculate the difference between the actual time difference matrix and each assumed time difference matrix; and select the assumed fault point with the smallest difference as the corrected final fault distance.

[0076] The dissimilarity measure is a scalar value used to quantify the similarity between two matrices (the actual time difference matrix and a hypothetical time difference matrix). It can be calculated as the sum of squared differences between corresponding elements of the two matrices (i.e., the square of the Frobenius norm), or other more complex matrix distance metrics. A smaller dissimilarity measure indicates that the two matrices are closer.

[0077] Specifically, the system iterates through all hypothetical time difference matrices ΔT_hyp(k) generated in S106. For each ΔT_hyp(k), the system compares it with ΔT_real, constructed in S104, which represents the actual measurement result, and calculates the difference between the two. For example, it calculates ||ΔT_real - ΔT_hyp(k)||_F^2. This process yields a sequence of differences, where each value corresponds to a hypothetical fault point within the search interval. This difference value can be understood as the gap between theory and reality. Theoretically, if the line model is perfectly accurate and the measurement has no errors, then when the hypothetical fault point coincides with the actual fault point, its corresponding hypothetical time difference matrix will be exactly equal to the actual time difference matrix, with a difference of zero. In practice, the point with the smallest difference represents the point where the theoretical time series relationship best matches the actual measured time series relationship. Therefore, the system ultimately selects the hypothetical fault point with the smallest difference, uses its location coordinates as the final fault distance with the highest accuracy after global information correction, and outputs this result.

[0078] As can be seen, synchronously acquiring the CVT secondary voltage traveling wave and TA current traveling wave, based on these two signals originating from the same physical event (fault) but with independent channels, utilizes the CVT and treats the CVT secondary voltage traveling wave as the instant the initial traveling wave front arrives. Through a collaborative verification step, their wavefront arrival time difference and polarity characteristics are compared. This filters out transient interference caused only by switching operations or induced lightning, which is significant only in a single physical quantity, ensuring the authenticity of the input data from the source. On this basis, a real time difference matrix is ​​constructed using reliable time data from multiple measurement points, elevating the information from one-dimensional time difference to a multi-dimensional time-series topology network. Finally, by calculating and optimizing the difference between the hypothetical time difference matrix generated near the preliminary location result and the actual matrix, and by adjusting the preliminary location result using global, redundant time-series information, deviations introduced by inaccurate single line parameters or minor synchronization errors can be effectively smoothed and corrected.

[0079] In practical applications, a special case exists in step S105: the fault search range includes lines with different media. For example, a short section of the line may be an underground cable, where the propagation speed of traveling waves is much lower than that of overhead lines. The dual-end ranging formula uses an average wave velocity. Therefore, the above embodiments are based on the assumption that the line medium is homogeneous or that the wave velocity is known and accurate. In this common scenario of mixed lines, the time difference matrix verification in the second stage is affected by the erroneous results of the first stage, meaning the optimal solution found is within the error range, resulting in lower accuracy of the optimal solution.

[0080] Therefore, in some other embodiments, after step S105, the method further includes: Please see Figure 9 , Figure 9 This is another flowchart illustrating the dual-end traveling wave ranging method for distribution networks based on CVT and current traveling wave coordination in the embodiments of this application; S401. Perform a preset check on the line based on the actual time difference matrix. The preset check is to determine that the time difference required for propagation between any two measurement points cannot be greater than the total time difference required for relay propagation through an intermediate measurement point. Among them, the preset verification refers to an algorithm that checks the inherent logical consistency of measurement data based on physical laws (the principle of the shortest propagation path for traveling waves). It does not require external references and can determine whether there is a contradiction based solely on the interrelationship of the data itself. Any two measurement points indicate that the verification process will traverse all possible pairs of measurement points in the time difference matrix (such as P_i, P_j). The intermediate measurement point refers to the third measurement point P_k introduced when considering the relationship between P_i and P_j. The total time difference required for relay propagation refers to the sum of the absolute value of the time difference between the traveling wave propagating from P_i to P_k and the absolute value of the time difference between the traveling wave propagating from P_k to P_j, i.e., |t_i-t_k|+|t_k-t_j|.

[0081] Specifically, for any three measurement points P_i, P_j, and P_k on the line, the travel wave originating from the fault point arrives at them at times t_i, t_j, and t_k, respectively. The direct propagation time difference between the travel wave and P_i and P_j is numerically expressed as |t_i - t_j|. The total time difference of the travel wave along the path via P_k is |t_i - t_k| + |t_k - t_j|. According to physical laws, the time consumption of the direct path should not exceed the time consumption of any detour path; therefore, |t_i - t_j| <= |t_i - t_k| + |t_k - t_j| must be satisfied. This step systematically traverses all combinations of measurement points in the actual time difference matrix, performing the above inequality check on each possible triple (P_i, P_j, P_k). Once any set of data is found to violate this basic physical constraint, it indicates that there is a logical contradiction within that set of data.

[0082] In some embodiments, during the preset verification, two measurement points are connected, and an intermediate measurement point is connected to any measurement point.

[0083] Specifically, S201 mentions that any two measurement points and an intermediate measurement point need to be verified. If all possible triples are traversed in a complex network with N measurement points, the computational load will be huge (number of combinations C(N, 3)), and many combinations have no physical meaning for verification.

[0084] This step precisely defines the preconditions for performing the triangle inequality check: |t_i-t_j|<=|t_i-t_k|+|t_k-t_j| First, the two measurement points (P_i and P_j) selected as verification endpoints must have a clear connection relationship. This means that the algorithm will only focus on measurement point pairs that are directly or significantly related in the topology, such as the two ends of a backbone or two adjacent nodes in a ring network.

[0085] Secondly, the selected intermediate measurement point P_k must also be connected to at least one of P_i or P_j. This ensures that P_k is a relevant point located on the path from P_i to P_j or on its immediate neighboring branch, rather than a completely unrelated remote point in the network.

[0086] For example, if the line topology is a straight line like P1-P2-P3-P4, the algorithm will choose (P1, P3) as two measurement points and P2 as an intermediate measurement point for verification, because P1 and P3 define a path, and P2 happens to be on this path. However, the algorithm will avoid choosing a combination like (P1, P4) and P_other (a point on a distant branch) for verification, because P_other is unrelated to the path from P1 to P4, so verifying it has no physical meaning and cannot effectively expose contradictions in the P1, P2, P3, P4 data link.

[0087] As can be seen, the above scheme minimizes the amount of data in the suspicious route model and maximizes the amount of other route models composed of reliable data, thereby increasing the likelihood of falling into the first scenario.

[0088] S402. Identify the lines that fail the verification as conflicting paths and identify the measurement points on the conflicting paths as suspicious data sources. Among them, a failed verification path refers to a path in which at least one triplet (P_i, P_j, P_k) is found in the S401 verification such that the inequality |t_i-t_j|<=|t_i-t_k|+|t_k-t_j| is not true; a contradictory path is the name given to this failed verification path consisting of three measurement points, indicating that its timing relationship violates common sense physics; a suspicious data source refers to the measurement points (i.e., P_i, P_j, P_k) that constitute this contradictory path, because the contradiction in this path must be caused by an error in the timing data of at least one of these three measurement points.

[0089] S403. Based on the suspicious data source, the line model is broken as the break point to obtain one or more suspicious line models and one or more other line models. The suspicious line model is the line composition corresponding to the measurement points with connection relationship, and the other line model is the line composition corresponding to other measurement points with connection relationship. In this context, "breakpoint" refers to treating the location of a suspicious data source as a logical disconnection point in the digitized line model; "breaking the line model" means dividing the originally connected and complete line topology network into several independent and unconnected subnetworks based on the breakpoint; "suspicious line model" refers to the subnetwork formed after the division that contains the suspicious data source; and "other line models" refer to the pure subnetwork formed after the division where all members are not suspicious data sources and the data is self-consistent.

[0090] Specifically, after S402 marks measurement points P1, P5, P8, etc., as suspicious data sources, this step operates on the digitized circuit model. It locates the positions of these suspicious measurement points in the model and treats them as breakpoints. This means logically severing all circuits directly connected to these points. For example, if P5 is located between P4 and P6, the system will disconnect the connections between P4 and P5 and between P5 and P6. As a result, the originally complete and continuous circuit network is divided into several independent segments. The circuit segments containing the suspicious data sources P1, P5, P8, and their direct neighbors are combined into one or more suspicious circuit models. The other parts of the network, whose measurement points are proven to be temporally consistent, constitute other circuit models. In this way, the system successfully decouples a complex problem containing a mixture of high-confidence and low-confidence data into several sub-problems with different levels of purity that require differentiated treatment.

[0091] S404. If the fault search range is completely covered by other line models, then execute S106.

[0092] The fault search range being completely covered by other line models means that the initial fault range defined in S105 (e.g., from line 10.5 km to 11.5 km) is geographically located entirely within the range of other line models composed of reliable data sources as defined in S403, and has no intersection with any suspicious line models.

[0093] As can be seen, after obtaining the actual time difference matrix containing time series information from multiple measurement points, the algorithm does not directly proceed to complex optimization calculations. Instead, it introduces a pre-defined verification step based on physical laws. Utilizing the common sense physics that the direct propagation time of a traveling wave from point A to point C should not be longer than the sum of its propagation times via the intermediate relay point B (A->B->C), it can directly identify suspicious data sources where serious logical contradictions in the time series data arise due to different propagation speeds. Once such data sources are identified, they are used as breakpoints, decoupling the complete line model into other, more reliable line models and the suspected line models to be investigated. In this way, subsequent precise positioning can be performed on a subset of reliable data first, reducing computational complexity and the risk of contamination by bad data, thus improving the robustness of the entire positioning algorithm and its reliability in the presence of abnormal data.

[0094] S405. If the fault search interval is not completely covered by other line models, reverse time-series localization will be performed on the corresponding other line models and the corresponding suspected line models to generate sub-candidate fault intervals respectively. Among them, "fault search range not completely covered by other line models" means that the preliminary fault search range determined in S105 partially or even entirely falls within the suspicious line models containing suspicious data sources that were segmented in S403; "reverse temporal sequence localization" refers to a localization approach opposite to the forward method in S106. It does not assume fault points, but rather calculates the set of all possible fault source locations that can generate this set of time sequences based on the known arrival times of the measurement points within each sub-model. This set is usually a continuous line segment; "sub-candidate fault range" refers to the fault location range obtained after performing reverse temporal sequence localization on each independent sub-line model (whether it is other or suspicious).

[0095] Specifically, when the system detects an overlap between the initial fault search interval defined in S105 and a suspected line model, it means the fault is likely located near this unreliable data area. At this point, the original high-precision matrix matching method (S106-S107) cannot be used directly due to inherent contradictions in the input data (actual time difference matrix). Therefore, the algorithm switches to a more robust divide-and-conquer strategy. It processes each segmented sub-model separately: For other line models (with reliable data): The algorithm uses the reliable arrival times of all measurement points within the model to perform a reverse time-series localization. For example, based on the arrival times of P2, P3, and P4, it calculates that the only fault source that can simultaneously satisfy these time constraints must be located within the [a, b] segment of the line. This [a, b] segment is the sub-candidate fault interval generated by the model.

[0096] For suspicious route models (unreliable data): the algorithm also performs reverse temporal localization on this model. Although the data of this model contradicts other models, there may be some internal consistency in the data (e.g., the overall GPS time of the entire substation has shifted). Based on the data logic within this model, the algorithm calculates another possible fault source range [c, d]. This [c, d] segment is the sub-candidate fault interval contributed by the suspicious model. In this way, the system decomposes a challenging global localization problem into multiple independent local localization problems solvable within their respective data systems.

[0097] S406. Combine the sub-candidate fault intervals and update the fault search interval, then execute S106.

[0098] As can be seen, in some embodiments, after the step of breaking the line model based on a suspicious data source as a breakpoint to obtain one or more suspicious line models and one or more other line models, the method further includes: if the fault search interval is not completely covered by other line models, then reverse time-series localization is performed on the corresponding other line models and the corresponding suspicious line models to generate sub-candidate fault intervals respectively; the sub-candidate fault intervals are combined and the fault search interval is updated, and the step of constructing multiple hypothetical fault points on the fault search interval and generating their corresponding hypothetical time difference matrices is performed.

[0099] The following describes an exemplary dual-end traveling wave ranging system 500 for power distribution networks based on CVT and current traveling wave coordination, provided by an embodiment of this application. Figure 10 This is an exemplary hardware structure diagram of a dual-end traveling wave ranging system 500 for power distribution networks based on CVT and current traveling wave coordination, provided in an embodiment of this application.

[0100] In some embodiments, the CVT-based and current traveling wave coordinated distribution network dual-end traveling wave ranging system 500 is a computer device or includes a computer device. The computer device includes a processor, memory, and network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores data. The network interface communicates with other external terminals or servers via a network connection. In some embodiments, the network interface can be a wired network interface; in some embodiments, it can also be a wireless network interface. When the computer program is executed by the processor, it implements the methods described in the embodiments of this application.

[0101] Those skilled in the art will understand that Figure 10 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0102] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0103] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as meaning if... or after... or in response to determining... or in response to detecting... Similarly, depending on the context, the phrase "when determining... or if (the stated condition or event) is interpreted as meaning if determining... or in response to determining... or in response to detecting (the stated condition or event)" or in response to detecting (the stated condition or event).

[0104] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0105] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A power distribution network double-ended traveling wave distance measurement method based on CVT and current traveling wave cooperation, characterized in that, The method is applied to a power distribution network line group, CVTs are arranged at measuring points at both ends of the line, and GPSs at both ends of the line have a time synchronization mechanism, and the method comprises the following steps: Under the time synchronization mechanism, a CVT secondary voltage traveling wave and a TA current traveling wave determined by a current transformer are synchronously collected; the CVT secondary voltage traveling wave is a specific high-frequency oscillation signal generated at the secondary side after a fault traveling wave excites the CVT; wherein the CVT secondary voltage traveling wave is characterized by a CVT high-frequency transient model; Wave head arrival time differences and polarity characteristics of the CVT secondary voltage traveling wave and the TA current traveling wave are compared and cooperatively verified to screen out a traveling wave caused by a real fault and determine a traveling wave arrival time, wherein the traveling wave arrival time is the average of the arrival times of the traveling wave caused by the real fault; The arrival times of the two measuring points are selected, a time difference between the two ends is calculated and substituted into a double-end traveling wave distance measurement formula to obtain a preliminary fault distance; An actual time difference matrix is constructed according to the arrival times of multiple measuring points; A fault search interval is demarcated on a line model with the preliminary fault distance as the center and a preset length extended to both sides of the line; A plurality of hypothetical fault points are constructed on the fault search interval to generate a corresponding hypothetical time difference matrix; A difference degree between the actual time difference matrix and each hypothetical time difference matrix is calculated, and a hypothetical fault point with the smallest difference degree is selected as a corrected final fault distance.

2. The method of claim 1, wherein, After the step of demarcating the fault search interval on the line model with the preliminary fault distance as the center and a preset length extended to both sides of the line, the method further comprises the following steps: The line is preset verified according to the actual time difference matrix, wherein the preset verification is to determine that a time difference required for propagation between any two measuring points cannot be greater than a total time difference required for relay propagation through an intermediate measuring point; A line with failed verification is determined as a contradictory path, and the measuring points on the contradictory path are identified as suspicious data sources; Based on the suspicious data sources, the line model is broken at the suspicious data sources as breaking points to obtain one or several suspicious line models and one or several other line models, wherein the suspicious line model is composed of lines corresponding to the measuring points with a connection relationship, and the other line model is composed of lines corresponding to other measuring points with a connection relationship; If the fault search interval is completely covered by the other line model, the step of constructing a plurality of hypothetical fault points on the fault search interval and generating a corresponding hypothetical time difference matrix is executed.

3. The method of claim 2, wherein, After the step of obtaining one or several suspicious line models and one or several other line models based on the suspicious data sources and breaking the line model at the suspicious data sources as breaking points, the method further comprises the following steps: If the fault search interval is not completely covered by the other line model, the corresponding other line model and the corresponding suspicious line model are both subjected to reverse time sequence positioning to respectively generate a sub-candidate fault interval; The sub-candidate fault intervals are combined, the fault search interval is updated, and the step of constructing a plurality of hypothetical fault points on the fault search interval and generating a corresponding hypothetical time difference matrix is executed.

4. The method of claim 2, wherein, The two measurement points have a connection relationship, and the intermediate measurement point has a connection relationship with any measurement point.

5. The method of claim 1, wherein, Before the step of performing collaborative verification by comparing the wave head arrival time difference and polarity characteristics of the CVT secondary voltage traveling wave and the TA current traveling wave, filtering out the traveling wave caused by a real fault, and determining the traveling wave arrival time, the method further comprises: Based on the CVT high-frequency transient model, when the side transient traveling wave wave head arrives, an attenuated high-frequency oscillation signal is excited at the CVT secondary side, and the frequency of the attenuated high-frequency oscillation signal is determined by the parameters of the CVT high-frequency transient model itself; After collecting the waveform of the attenuated high-frequency oscillation signal, wavelet transform is applied for multi-scale analysis, and the time when the signal energy modulus maximum value appears under the highest frequency analysis scale is identified. The time is marked as the wave head arrival time of the CVT secondary voltage traveling wave.

6. The method of claim 1 or 2, wherein, Before the step of performing collaborative verification by comparing the wave head arrival time difference and polarity characteristics of the CVT secondary voltage traveling wave and the TA current traveling wave, filtering out the traveling wave caused by a real fault, and determining the traveling wave arrival time, the method further comprises: The TA current traveling wave is subjected to phase-mode separation to extract a line-mode component; wherein the phase-mode separation parameters are: wherein is the zero mode component, , is the line mode component, , , , are the three-phase currents of the TA current traveling wave, respectively; The line-mode component is subjected to a modulus maximum value method, the time when the amplitude first changes abruptly is detected, and is determined as the wave head arrival time of the TA current traveling wave.

7. The method of claim 1, wherein, The step of performing collaborative verification by comparing the wave head arrival time difference and polarity characteristics of the CVT secondary voltage traveling wave and the TA current traveling wave specifically comprises: Comparing the wave head arrival time of the CVT secondary voltage traveling wave with the wave head arrival time of the TA current traveling wave; When the time difference is less than a preset time threshold, the polarity relationship of the initial half wave of the CVT secondary voltage traveling wave and the TA current traveling wave is determined, and is compared with a pre-established fault event polarity characteristic library; Excluding the traveling wave whose polarity characteristic and fault characteristic difference is greater than a preset characteristic threshold.

8. A power distribution network double-ended traveling wave distance measurement system based on CVT and current traveling wave coordination, characterized in that, The power distribution network double-end traveling wave distance measurement system based on CVT and current traveling wave collaboration comprises one or more processors and a memory; the memory is coupled with the one or more processors, the memory is used to store computer program code, the computer program code comprises computer instructions, and the one or more processors invoke the computer instructions to enable the power distribution network double-end traveling wave distance measurement system based on CVT and current traveling wave collaboration to perform the method in any one of claims 1-7.

9. A computer program product comprising instructions, characterized in that, When the computer program product runs on the power distribution network double-end traveling wave distance measurement system based on CVT and current traveling wave collaboration, the power distribution network double-end traveling wave distance measurement system based on CVT and current traveling wave collaboration is enabled to perform the method in any one of claims 1-7.

10. A computer-readable storage medium comprising instructions, characterized in that, When the instructions run on the power distribution network double-end traveling wave distance measurement system based on CVT and current traveling wave collaboration, the power distribution network double-end traveling wave distance measurement system based on CVT and current traveling wave collaboration is enabled to perform the method in any one of claims 1-7.

Images