A double-circuit transmission line hybrid fault location method combining impedance and traveling wave

Abstract

The application relates to a double-circuit transmission line hybrid fault positioning method combining impedance and traveling wave, which comprises the following steps: collecting voltage and current time domain signals before and after the occurrence of a fault, and calculating pure fault current; based on the pure fault current, a fault distance estimation value is calculated; a traveling wave signal triggered by the fault is extracted; based on the fault distance estimation value, a self-adaptive time window with a constraint effective wave head is generated; a dynamic correction model of mutual coupling influence coefficients is constructed, and the output is dynamic mutual coupling influence coefficients; based on the dynamic mutual coupling influence coefficients, the self-adaptive time window is adjusted; the wave head time difference between incident waves and reflected waves is calculated; and the fault positioning distance of the double-circuit transmission line is calculated by combining the fault distance estimation value and the wave head time difference. The application solves the problems of low fault positioning precision of a double-circuit transmission line of a common tower under the conditions of strong mutual coupling and high resistance, reduces the wave head misrecognition rate through mutual coupling perception self-adaptive time window and wave head sharpness discrimination, and guarantees positioning stability through iterative optimization and result verification.

Description

Technical Field

[0001] This application relates to the field of power system fault diagnosis technology, specifically to a hybrid fault location method for dual-circuit transmission lines that combines impedance and traveling wave. Background Technology

[0002] Existing transmission line fault location methods are mainly divided into two categories: impedance methods and traveling wave methods. Impedance methods are simple to implement and have low computational complexity, but their location accuracy is easily affected by fault resistance, remote system feed, phase-to-phase coupling in dual-circuit lines, and system non-uniformity, especially in dual-circuit lines where accuracy is significantly reduced. Traveling wave methods have high location accuracy, but single-ended traveling wave methods are susceptible to reflected wave interference and time window selection. Traditional single-ended traveling wave methods rely on high sampling rates and wavelet transforms, while traveling wave correlation methods, although reducing sampling rate requirements, are highly sensitive to time window selection. Dual-ended traveling wave methods require synchronous data and high-speed communication channels at the other end, resulting in high engineering deployment costs and poor economic efficiency. In addition, some existing methods use artificial intelligence technology to achieve fault location, but they suffer from computational complexity and poor real-time performance, making them difficult to adapt to actual engineering application scenarios.

[0003] However, existing technologies have significant drawbacks: 1. Most existing hybrid positioning methods are simply a combination of impedance and traveling wave methods, failing to dynamically constrain the impedance pre-estimation results for traveling wave correlation analysis. They cannot fundamentally suppress reflected waves and noise interference, and lack a dedicated adaptation mechanism for the strong mutual coupling characteristics of double-loop lines, making it difficult to balance positioning accuracy and anti-interference capabilities. 2. They cannot adapt to different fault types, load changes, and line parameter differences. Residual mutual coupling interference can lead to traveling wave signal distortion and wavefront identification deviation, resulting in excessive positioning error, especially under complex interference scenarios. 3. They cannot effectively distinguish between valid wavefronts and noise / clutter interference, easily leading to wavefront misjudgment and missed judgment. Furthermore, there is no closed-loop optimization mechanism for invalid wavefronts; once a wavefront is incorrectly identified, it cannot be corrected by adjusting the time window, making it difficult to guarantee traveling wave positioning accuracy. 4. The accuracy of fault distance prediction is insufficient, failing to effectively offset the inherent errors of a single method. Simultaneously, the impact of mutual coupling interference on the prediction is not considered, resulting in large prediction deviations and an inability to provide precise time window constraints for traveling wave positioning, further affecting overall positioning accuracy.

[0004] Therefore, this invention proposes a hybrid fault location method for dual-circuit transmission lines that combines impedance and traveling wave. Summary of the Invention

[0005] To overcome the shortcomings of the prior art, this application provides a hybrid fault location method for dual-circuit transmission lines that combines impedance and traveling wave, specifically adopting the following technical solution.

[0006] A hybrid fault location method for two-circuit transmission lines that combines impedance and traveling wave includes the following steps.

[0007] S1. Electrical Characteristics Analysis and Measurement Boundary Calibration of Double-Circuit Faults: Analyze the conduction characteristics of double-circuit faults, calibrate the effective signal measurement boundaries, unify the electrical calculation benchmark, and eliminate the positioning deviation caused by mutual coupling interference from the source.

[0008] S2. Local voltage and current signal synchronous acquisition and pure fault component extraction: Synchronously acquire voltage and current signals before and after the fault, separate the pure fault component through difference calculation, eliminate various interferences, and provide clean and reliable input data for subsequent calculations.

[0009] S3. Impedance-based pre-estimation of fault distance: Based on the pure fault component, three improved impedance algorithms are used to calculate the per-unit value of the fault distance. The fault distance prediction value is obtained through adaptive weighted fusion, which limits the range of subsequent traveling wave analysis and improves the robustness of the location.

[0010] S4. Extract the traveling wave signal and calculate the wavefront time difference: Purify, decouple, and enhance the wavefront of the faulty high-frequency transient traveling wave to eliminate mutual coupling interference and obtain a clear and stable traveling wave signal, which prepares for wavefront identification.

[0011] S5. Mutual Coupling Sensing Adaptive Time Window Generation Based on Impedance Prediction: Combining the fault distance prediction value and real-time mutual coupling strength, the center and width of the time window are dynamically corrected to accurately lock the effective wavefront and avoid missed identification and false identification.

[0012] S6. Intra-window constraint cross-correlation calculation and determination of effective wavefront time difference: Traveling wave cross-correlation analysis is performed within an adaptive time window. The effective wavefront and pseudo wavefront are distinguished by the sharpness of the correlation peak, and the time difference between the incident wave and the reflected wave is accurately obtained.

[0013] S7. Calculate the fault distance: Combine the traveling wave propagation speed and wavefront time difference to calculate the basic traveling wave positioning distance. After correction by reflection at the opposite end, it is initially integrated with the impedance estimate to lay the foundation for accurate positioning.

[0014] S8. Mutual Coupling Compensation, Confidence-Weighted Iterative Optimization and Final Fault Distance Calculation: Mutual coupling residual compensation is performed on the preliminary fusion results, and weighted fusion is performed in combination with wavehead recognition confidence. After iterative optimization and result verification, a high-precision and stable final fault location result is obtained.

[0015] The technical solution of this application has achieved the following beneficial effects.

[0016] Signal distortion is eliminated through triple mutual coupling suppression measures. Combined with an improved Erikson method and multi-algorithm adaptive fusion, it is adaptable to various complex fault conditions and lines ranging from 50 to 300 km. By using mutual coupling sensing adaptive time windows and wavefront sharpness discrimination, the false wavefront recognition rate is reduced. Iterative optimization and result verification ensure stable positioning. No additional measurement equipment is required; it boasts high computational efficiency and strong real-time performance. It achieves deep synergy between impedance and traveling wave methods, overcoming the limitations of traditional hybrid positioning, significantly improving positioning accuracy and reliability, reducing application costs, facilitating rapid fault diagnosis, minimizing power outage time, and ensuring the safe and stable operation of the power system. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the hybrid fault location method for dual-circuit transmission lines that combines impedance and traveling wave in the embodiments of this application.

[0018] Figure 2 This is a schematic diagram illustrating the final fault distance calculation process of the hybrid fault location method for dual-circuit transmission lines that combines impedance and traveling wave in the embodiments of this application. Detailed Implementation

[0019] The present application will now be further described with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present application and should not be construed as limiting the scope of protection of the present application. It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present application.

[0020] like Figure 1 and Figure 2 As shown, this invention discloses a hybrid fault location method for dual-circuit transmission lines that combines impedance and traveling wave, comprising the following steps.

[0021] S1. Electrical Characteristics Analysis and Measurement Boundary Calibration of Double-Circuit Faults: In response to the common problem that traditional double-circuit fault location methods generally ignore phase-to-phase mutual coupling, line-to-line electromagnetic coupling, and system non-uniformity, resulting in fault signal distortion, component extraction distortion, and large location errors, this application completes the sorting out of the conduction characteristics of double-circuit faults and the calibration of effective signal boundaries. This provides a unified electrical reference for subsequent accurate extraction of pure fault components, impedance pre-estimation, and traveling wave analysis, eliminating the systematic deviation caused by coupling interference from the source.

[0022] S2. Local voltage and current signal synchronous acquisition and pure fault component extraction: Separate the electrical components that are only related to the fault from the time-domain synchronous acquisition signal, eliminate the interference of load fluctuation, normal system operation status and power frequency steady-state components, and obtain pure fault data without steady-state offset and unaffected by load dynamics, so as to provide a clean and reliable input signal for subsequent impedance pre-estimation.

[0023] S3. Impedance-based Fault Distance Prediction: Based on pure fault components, three improved impedance algorithms adapted to dual-loop coupling scenarios are used to calculate the per-unit value of the fault distance. A stable and reliable fault distance prediction is obtained through adaptive weighted fusion, which is used to limit the effective search range of subsequent traveling wave analysis. This solves the problems of traditional traveling wave methods being susceptible to clutter, multiple reflected wave interference, and wavefront misidentification in full-line search, improving the efficiency and anti-interference capability of traveling wave localization. Through redundant calculations using multiple algorithms and adaptive weight allocation, the failure of a single impedance algorithm in high fault resistance, strong mutual coupling, and near-end fault scenarios is avoided, balancing the robustness and accuracy of coarse localization and providing a reliable pre-constraint for adaptive time window generation.

[0024] S4. Extract traveling wave signal and calculate wavefront time difference: For high-frequency transient traveling wave signal triggered by fault, complete signal purification, coupling suppression, and wavefront feature enhancement to eliminate waveform distortion and inter-mode crosstalk caused by strong mutual coupling of double loops, providing a clear, stable, and interference-free traveling wave signal for subsequent in-window correlation analysis.

[0025] S5: Adaptive Time Window Generation Based on Impedance Prediction: Based on the obtained fault distance prediction, the theoretical arrival time of the reflected traveling wave at the fault point is calculated. Combined with the dynamic correction of the window center and window width by the real-time mutual coupling strength of the double loop, an adaptive time window for constraining the effective wavefront is generated. This completely solves the problem that the traditional fixed window width is prone to missing the effective wavefront or containing too many interference reflected waves and noise, and realizes wavefront locking in the mutual coupling scenario.

[0026] S6. Intra-window constrained cross-correlation calculation and determination of effective wavefront time difference: Within the adaptive time window, forward and backward traveling wave cross-correlation analysis is carried out. The effective wavefront and pseudo wavefront are distinguished by the sharpness of the correlation peak, and the time difference between the incident wave and the reflected wave is accurately identified, eliminating misidentification caused by noise, clutter, and multiple reflections.

[0027] S7. Calculate the fault distance: based on the wavefront time difference Δt and the traveling wave propagation speed. By combining the impedance prediction results and the traveling wave time difference, the accurate fault distance is calculated, and the accuracy is further improved by the reflection correction at the opposite end, laying the foundation for subsequent mutual coupling compensation and final location solution.

[0028] S8. Mutual Coupling Compensation, Confidence-Weighted Iterative Optimization and Final Fault Distance Solution: Distance compensation is performed to address the residual mutual coupling effect of the double loop. Dynamic weighted fusion is completed based on wavefront identification confidence. Iterative window optimization is used to further improve accuracy, ultimately obtaining a stable, high-precision fault location result with strong anti-interference capability.

[0029] Example 1.

[0030] This embodiment 1 discloses a hybrid fault location method for dual-circuit transmission lines that combines impedance and traveling wave. In step S1, for dual-circuit transmission lines with a common tower, it is clear that strong magnetic coupling between phases and between lines is the core cause of fault signal distortion and decreased location accuracy. The coupling effect directly changes the transmission path, amplitude distribution and timing characteristics of the fault transient signal, resulting in poor adaptability, failure of high-impedance faults and misidentification of wavefronts when conventional single-circuit line location algorithms are directly applied.

[0031] Define unified electrical constraint boundaries across the entire line: calibrate the electrical influence range of the local measurement end and the remote system, clarify the start and end conditions for fault-related additional states, and determine the effective measurement frequency band, synchronous acquisition requirements, and fault trigger thresholds for voltage and current signals.

[0032] Example 2.

[0033] This embodiment 2 discloses a hybrid fault location method for dual-circuit transmission lines that combines impedance and traveling wave. In step S2, due to the strong mutual coupling effect of the shared-tower double-circuit line in the dual-circuit line, signal distortion is amplified, resulting in a high degree of overlap between fault characteristic signals and normal operation signals. Conventional methods that directly extract the signal after the fault cannot eliminate the influence of load and system operating mode. Therefore, this step, based on the principle of linear superposition, decomposes the system electrical state into a pre-fault steady-state operating network and a post-fault additional state network. By using difference calculation, the additional fault components are completely isolated, eliminating the negative impact of load fluctuations and changes in system operating mode on the location results from the root cause.

[0034] Specific operation: A high-precision measuring device is installed locally to synchronously acquire the time-domain signals of three-phase voltages (Va, Vb, Vc) and three-phase currents (Ia, Ib, Ic) before and after the fault occurs. The sampling frequency is selected from 1kHz to 5kHz to balance measurement accuracy and computational efficiency. Based on the superposition principle, the difference between the post-fault signal and the pre-fault signal is calculated to extract the pure fault component that is only related to the fault. The period before the fault occurs is the steady-state operation stage, and the period after the fault occurs is the fault transient and steady-state stage.

[0035] The formula for calculating the change in pure fault current is as follows.

[0036] In the formula, This represents the change in pure fault current; Indicates fault current; This indicates the pre-fault current.

[0037] At the same time, the same principle is used to calculate the pure fault voltage change: .in, This represents the change in voltage during a pure fault. For the voltage phasor after the fault, This is the pre-fault voltage phasor.

[0038] By measuring signals and performing differential calculations, pure fault electrical data that is completely independent of load dynamics and system steady-state operation is obtained, eliminating steady-state component interference caused by mutual coupling, and providing accurate and offset-free input data for subsequent fault distance prediction based on impedance algorithms.

[0039] This technical solution calculates the pure fault current change and the pure fault voltage change, and completely isolates the interference of load fluctuations and system operating mode changes on the fault signal through difference calculation. No additional measuring equipment is required; only existing measuring devices are used to collect steady-state and fault-state signals before and after the fault, allowing the extraction of pure fault components. This provides offset-free and interference-free input data for subsequent impedance calculations and traveling wave analysis, solving the problems of fault signal aliasing and low positioning accuracy in traditional methods, while simultaneously ensuring measurement efficiency and data accuracy.

[0040] Example 3.

[0041] This embodiment 3 discloses a hybrid fault location method for dual-circuit transmission lines that combines impedance and traveling wave, wherein step S3 includes the following steps.

[0042] S31. Calculate the per-unit value of fault distance using the Takagi method. Based on the pure fault current component optimization algorithm, the influence of load current on the positioning results is eliminated, and the robustness of coarse positioning is improved.

[0043] .

[0044] In the formula, It is the conjugate phasor of the pure fault current variation component; ( ) indicates the imaginary part; Indicates the local fault voltage; Indicates the local fault current; This indicates the positive sequence impedance of the line.

[0045] S32. Calculate the per-unit value of fault distance using the zero-sequence improved Takagi method. The method calculates fault distance based on zero-sequence fault components, eliminating the need for pre-fault steady-state data acquisition. It is suitable for engineering scenarios where there is no pre-fault waveform recording or historical steady-state data, significantly improving the method's field applicability.

[0046] .

[0047] In the formula, For zero-sequence current phasors, .

[0048] S33. Establish per-unit values ​​for fault distance using the Eriksson method. The equations take into account the influence of source impedance and fault resistance, further improving the coarse positioning accuracy and adapting to high fault resistance scenarios.

[0049] .

[0050] ; ; .

[0051] In the formula, , , It is a constant, determined by the local terminal voltage. Current Positive sequence impedance of the line Source impedance To be determined jointly; This is the fault resistor.

[0052] S34. Adaptive weighted fusion to obtain the comprehensive fault distance prediction value: Construct a comprehensive fault distance prediction value model, and adaptively allocate weights based on the variance of the calculation results of each algorithm. The smaller the variance and the stronger the algorithm stability, the higher the corresponding weight. By fusion, the shortcomings of the single algorithm in adapting to the working conditions are eliminated, and a stable and reliable comprehensive fault distance prediction value is obtained. The comprehensive fault distance prediction value model is as follows.

[0053] ; .

[0054] ; ; .

[0055] In the formula: This is the per-unit value for the comprehensive fault distance; , and The per-unit values ​​of fault distance calculated by the Takagi method, the zero-sequence improved Takagi method, and the Eriksson method are respectively. The distance to the fault is the estimated value; L is the total length of the dual-circuit transmission line; , and They are respectively , and Adaptive weighting coefficients, , , They are respectively , and The smaller the variance, the more stable the algorithm's calculation results, and the larger the weights. Variance , , The calculation is based on the pure fault components in step S2. For each algorithm, 3 to 7 sets of per-unit fault distance values ​​are calculated continuously, obtained through rolling variance calculation, balancing computational efficiency and result stability without complex iterative calculations. Algorithm failure handling is as follows: when a single algorithm fails, such as when the calculation result is out of bounds or there is no solution, its weight is removed and evenly distributed to the effective algorithms; when two algorithms fail, the results of the remaining effective algorithms are directly used.

[0056] This technical solution employs a multi-algorithm collaborative coarse localization approach, utilizing three complementary impedance algorithms to address the insufficient adaptability of a single algorithm. Specifically, the zero-order improved Takagi method is adapted to scenarios without pre-fault data, the Eriksson method is adapted to high-impedance fault scenarios, and adaptive weight allocation effectively avoids the failure of a single algorithm under specific operating conditions. Furthermore, an algorithm failure handling mechanism is implemented to ensure the stability and reliability of the coarse localization results, defining a reasonable range for subsequent traveling wave precise localization and improving the versatility and robustness of the localization.

[0057] This invention improves pre-location accuracy and adapts to different fault scenarios by dynamically allocating weights through a comprehensive fault distance prediction model. It calculates per-unit fault distance values ​​using a weighted fusion of the Takagi method, the zero-order improved Takagi method, and the Eriksson method, thereby enhancing the reliability and accuracy of coarse-grained location and addressing the issues of single impedance algorithms being susceptible to failure due to fault resistance and mutual coupling interference, as well as insufficient accuracy.

[0058] Example 4.

[0059] This embodiment 4 discloses a hybrid fault location method for dual-circuit transmission lines that combines impedance and traveling wave, wherein step S4 includes the following steps.

[0060] S41. Description of Traveling Wave Propagation Characteristics: When a fault occurs, a high-frequency transient traveling wave is generated, which propagates bidirectionally along the line. Reflection and transmission occur at impedance discontinuities such as the fault point and the opposite end of the line. The fault distance can be accurately calculated by using the timing difference between the incident traveling wave and the reflected traveling wave at the fault point. For double-circuit scenarios, the transmission line telegraph equation is used to describe the propagation law of the traveling wave along the line.

[0061] The voltage distribution along the line is expressed by the following formula, and the propagation characteristics of the traveling wave are described by the forward and backward traveling wave functions.

[0062] .

[0063] In the formula, This represents the voltage at position x and time t. Represents the forward and backward functions; Indicates the speed of propagation; Indicates time.

[0064] The current distribution along the line is calculated using the following formula, which, together with the voltage distribution function, describes the propagation of the traveling wave.

[0065] .

[0066] In the formula, This represents the current at position x and time t. This represents the characteristic impedance of the line.

[0067] S42. Traveling Wave Signal Mode Decoupling (Suppressing Mutual Coupling Interference): To address the strong mutual coupling characteristics of dual circuit lines, Park mode transformation is introduced to separate the strongly coupled three-phase signals into independent mode components, including zero-axis, direct-axis, and quadrature-axis components. This avoids mutual coupling interference to the traveling wave signal, ensures the purity of the traveling wave signal, and provides a reliable signal foundation for subsequent wavefront identification and correlation analysis.

[0068] The voltage components after Park transformation are calculated according to the following formula, and the transformation is performed using three-phase voltage signals.

[0069] .

[0070] In the formula, Represents zero, direct-axis, and quadrature-axis voltages; Indicates the phase angle; This represents the three-phase voltage. Park mode transformation is used to decouple the three-phase signals, eliminating phase-to-phase and line-to-line mutual coupling interference. The zero-axis component helps determine the fault type, while the direct-axis and quadrature-axis components serve as the main criteria for wavefront identification, ensuring the reliability of subsequent traveling wave analysis.

[0071] S43. Construction of forward and backward traveling wave components: Based on the decoupled pure fault incremental signal, construct forward incident traveling wave and backward reflected traveling wave components to highlight the wavefront timing characteristics.

[0072] The backward traveling wave signal is calculated using the following formula. .

[0073] .

[0074] In the formula, Indicates a backward traveling wave; This represents the incremental voltage fault signal after decoupling; Indicates characteristic impedance; This represents the current fault increment signal after decoupling.

[0075] The forward traveling wave signal is calculated using the following formula. .

[0076] In the formula, It indicates a forward traveling wave.

[0077] S44. Traveling wave signal filtering and purification: Since the actual acquired traveling wave signal contains a large amount of low-frequency noise, which will interfere with wavefront identification, a Butterworth high-pass filter is used to filter the traveling wave signal, isolate high-frequency transient traveling wave signals, and improve wavefront clarity.

[0078] The Butterworth filter transfer function is defined as follows.

[0079] .

[0080] In the formula, Represents the transfer function; Represents the Laplace variable; Indicates the cutoff angular frequency; Indicates the filter order.

[0081] The filtered signal is calculated using the following formula. The original traveling wave signal is then processed by applying the filter's impulse response. and It can effectively isolate high-frequency transient traveling wave signals, improve the robustness of the signal in noisy environments, and provide a clear signal basis for subsequent wavefront identification.

[0082] .

[0083] In the formula, This represents the filtered signal; Indicates impulse response; This represents convolution.

[0084] Traveling wave signals consist of forward and backward traveling wave signals. This invention filters out the effective traveling wave signals related to the fault from both signals, using them as input for fault location and protection criteria. The forward and backward traveling wave signals are the two core components of a traveling wave signal; the effective traveling wave signal is the useful portion filtered from these two components, relating only to the fault. The forward traveling wave signal is the incident wave triggered by the fault, propagating along the line towards the opposite end, and is far from the local measurement end. The backward traveling wave signal is the incident wave reflected from the fault point and the opposite end of the line, propagating along the line towards the local measurement end.

[0085] This technical solution constructs forward-incident and backward-reflected traveling wave components based on the decoupled pure fault incremental signal, highlighting the wavefront timing characteristics and solving the problems of signal coupling, noise interference, and difficulty in wavefront identification in traditional traveling wave positioning. It provides high-quality traveling wave signals for subsequent wavefront identification and time difference calculation, offering reliable support for accurate positioning.

[0086] Example 5.

[0087] This embodiment discloses a hybrid fault location method for dual-circuit transmission lines that combines impedance and traveling wave, wherein step S5 includes the following steps.

[0088] S51, Calculation of mode-domain traveling wave propagation speed: The mode-domain wave speed is calculated by combining mutual inductance and self-inductance parameters, which fits the actual transmission characteristics of the double-loop line and eliminates the inherent positioning error caused by wave speed calculation.

[0089] Linear mode principal component wave velocity : In the formula, Indicates the self-inductance per unit length of the line; Indicates mutual inductance per unit length of the line; This represents capacitance per unit length.

[0090] S52, Calculate the center time of the time window Based on the theoretical arrival time of the reflected wave, which is based on impedance pre-estimation, it is ensured that the window can accurately cover the effective reflected wavefront.

[0091] .

[0092] In the formula: At the initial moment of the fault, it is identified by the abrupt change characteristics of the fault components; Estimated distance to fault; η is the traveling wave propagation speed; η is the dynamic mutual coupling effect coefficient; the center time is corrected according to the mutual coupling effect to offset the traveling wave propagation time deviation caused by mutual coupling and ensure that the window accurately covers the reflected wavefront.

[0093] S53. Dynamic Coupling Coefficient Correction and Adaptive Window Width Calculation: Construct a dynamic correction model for the mutual coupling influence coefficient, perform dynamic correction of the mutual coupling influence coefficient, combine the real-time load rate of the line and the fault type, dynamically calculate the mutual coupling influence coefficient, fit the real-time operating status of the line, and adaptively adjust the window width according to the mutual coupling strength: the stronger the mutual coupling and the greater the waveform distortion, the wider the window width should be appropriately increased to avoid missing effective wavefronts.

[0094] The dynamic correction model for the mutual coupling influence coefficient is as follows.

[0095] ; . ; . .

[0096] In the formula, W is the adaptive window width; Based on the basic window width, L is the total length of the two-circuit transmission line; η is the traveling wave propagation speed; η is the dynamic mutual coupling influence coefficient; For double-loop mutual impedance; λ is the positive sequence impedance of the line; λ is the mutual coupling correction coefficient, and α is the weighting coefficient, α∈[0.4,0.6], which is calibrated according to the actual structure of the line, and is defaulted to 0.5; This is the load factor correction factor, where S is the current actual load on the line. The load is the line's rated load; the larger the load, the stronger the mutual coupling effect. The closer to 1; This is a fault type correction factor for ground faults. =1.2, phase-to-phase fault =0.9, the mutual coupling effect is more significant during ground faults, requiring greater correction. A mutual coupling correction coefficient is introduced, based on fault type and load rate, to accurately adapt to complex operating scenarios with dual-circuit lines. This is the wavefront ambiguity correction factor, calibrated based on the amplitude of the fault voltage mutation. This is the maximum amplitude of the voltage signal change when the fault occurs, i.e., the pure fault voltage increment. The peak value; This is the rated voltage of the transmission line. The value range is [0.1, 0.5]. The larger the voltage jump amplitude, the clearer the wavefront. The smaller the value, the smaller the window width adjustment range; conversely, the larger the value, the smaller the voltage fluctuation amplitude, and the more blurred the wavefront. The larger the value, the wider the window should be to ensure that no valid waveheads are missed.

[0097] Traditional mutual coupling coefficients only consider inherent line parameters, neglecting the impact of load variations and fault types on the mutual coupling effect during actual operation. For example, under heavy load, the double-circuit current increases, enhancing the mutual coupling effect; during ground faults, the presence of zero-sequence current further amplifies the mutual coupling effect. This invention uses a dynamically corrected coefficient λ to make the mutual coupling effect coefficient more closely match actual operating conditions, solving the positioning deviation problem caused by traditional fixed mutual coupling coefficients. Engineering adaptability: load factor It can be calculated in real time using local terminal current signals, and the fault type can be quickly identified by the characteristics of zero-sequence current and phase current abrupt changes, without the need for additional measuring equipment.

[0098] S54: Determine the final adaptive time window: Based on the corrected center time and the adaptive window width, determine the effective wavefront search interval.

[0099] The window range is determined by the following formula, achieving precise centering and adaptive width, which is different from the rough design of traditional fixed windows.

[0100] .

[0101] In the formula: The center moment of the time window; , These represent the start and end times of the time window.

[0102] In this technical solution, a dynamic correction model for the mutual coupling influence coefficient is constructed to dynamically correct the mutual coupling influence coefficient. Combining the real-time load rate and fault type of the line, the mutual coupling influence coefficient is dynamically calculated to match the real-time operating status of the line. The window width is adaptively adjusted according to the mutual coupling strength to avoid missing effective wavefronts. Simultaneously, the dynamic mutual coupling coefficient correction model, closely matching the real-time operating status of the line, achieves precise time window adaptation without the need for additional measurement equipment, balancing real-time positioning and accuracy, and avoiding missed or misidentified wavefronts.

[0103] Example 6.

[0104] This embodiment discloses a hybrid fault location method for dual-circuit transmission lines combining impedance and traveling wave. In step S6, based on the obtained fault distance estimate, adaptive time window constrained correlation analysis is used to first calculate the in-window cross-correlation function, identify the wavefront positions of the reflected wave and the incident wave, and then obtain the time difference. The following formula is used to calculate the in-window cross-correlation function, which is used to quantify the similarity of the forward and backward traveling wave signals and identify the wavefront of the reflected wave.

[0105] .

[0106] In the formula, Represents the relevant function; Indicates a time lag; Indicates the sampling interval; Indicates the number of samples; This represents the mean of the forward and backward traveling wave signals. Through traveling wave signal extraction, decoupling, filtering, and correlation analysis, the foundation is laid for subsequent calculations of wavefront time difference and accurate fault distance, while also addressing issues of mutual coupling and noise interference. Specifically, it consists of the following two sub-steps:

[0107] S61: Intra-window cross-correlation function calculation: within a defined adaptive time window and Inside, calculate the cross-correlation function. Focus on the effective traveling wave head and reduce irrelevant interference.

[0108] . .

[0109] In the formula: For the in-window cross-correlation function; Indicates time lag; N represents the number of samples within the adaptive time window; , These are the sampled values ​​of the filtered forward and backward traveling wave signals at the corresponding times; , These represent the mean values ​​of the forward and backward traveling wave signals within the window after filtering, respectively; The sampling interval is represented by . Through cross-correlation calculation with in-window constraints, noise and clutter interference outside the time window can be effectively shielded, ensuring that the correlation analysis is only for the valid traveling wave signal; m represents the number of lag sampling points, which is an integer.

[0110] S62: Wavehead Identification and Time Difference Determination: Finding the Cross-Correlation Function The time lag corresponding to the maximum value The correlation peak sharpness correction factor δ is calculated to quantify the correlation peak sharpness, distinguish between effective wavefronts and pseudo wavefronts, and avoid misidentification caused by noise interference.

[0111] .

[0112] If δ ≥ 0.6, where 0.6 is an empirical threshold, it can be determined as a valid wavefront based on line parameter calibration. That is, the time difference Δt between the arrival of the incident wave and the reflected wave, i.e.: .

[0113] If δ < 0.6, it is determined to be a false wavefront, and the process returns to step S5 to readjust the time window until a valid wavefront is identified.

[0114] An effective wavefront is defined as: the wavefront signal of an incident traveling wave or a traveling wave reflected from a fault point that propagates along a double-circuit transmission line after a fault is triggered. It has clear waveform abrupt change characteristics, and after cross-correlation analysis, the correlation peak sharpness correction factor δ≥0.6. It can accurately reflect the traveling wave propagation timing information related to the fault location and can be used for wavefront time difference calculation. A false wavefront is defined as a false wavefront signal caused by line clutter, multiple reflection interference, electromagnetic coupling interference, or measurement noise. Its waveform changes abruptly and becomes blurred. After cross-correlation analysis, the correlation peak sharpness correction factor δ < 0.6. It cannot reflect the traveling wave propagation timing information related to the fault location and cannot be used for wavefront time difference calculation.

[0115] This step optimizes the window constraint and sharpness, ensuring that the correlation analysis focuses only on the effective wavefront region. This effectively compensates for reflected wave interference even at a kHz sampling rate, improving the accuracy and reliability of wavefront identification.

[0116] This technical solution distinguishes between valid and spurious wavefronts by constructing a correlation peak sharpness correction factor, avoiding misidentification caused by noise interference. Combined with correlation peak sharpness discrimination, it solves the problem of wavefront misidentification caused by noise and clutter interference in traditional traveling wave analysis. The correlation peak sharpness correction factor accurately distinguishes between valid and spurious wavefronts, avoiding time difference calculation errors caused by noise interference. Simultaneously, a spurious wavefront feedback adjustment mechanism is set up to further improve the accuracy of wavefront identification, providing a reliable time difference basis for subsequent accurate fault distance calculation.

[0117] Example 7.

[0118] This embodiment discloses a hybrid fault location method for dual-circuit transmission lines that combines impedance and traveling wave, wherein step S7 includes the following steps.

[0119] S71: Basic Traveling Wave Positioning Distance Calculation: Combining Traveling Wave Propagation Speed The positioning distance of the base traveling wave is calculated using the following formula, based on the wavefront time difference Δt. This distance represents the preliminary positioning result of the traveling wave method, demonstrating its high-precision advantage. .

[0120] In the formula, The basic traveling wave positioning distance; Δt is the propagation speed of the traveling wave; Δt is the arrival time difference between the incident wave and the reflected wave.

[0121] S72: End-to-End Reflection Correction: Improves positioning accuracy and suppresses interference from far-end reflected waves. Considering that reflected waves at the end of the line may interfere with the positioning distance of the basic traveling wave, an end-to-end reflection correction factor model is constructed, which, combined with the total line length L, corrects the reflection at the far end. After making corrections, we obtain the corrected precise value of the traveling wave. The model for the opposite-end reflection correction factor is as follows.

[0122] .

[0123] In the formula, The corrected traveling wave value is given; L is the total length of the two-circuit transmission line. This is the basic traveling wave positioning distance. Correction logic: When the basic traveling wave positioning distance exceeds the total line length, it is determined to be a deviation caused by far-end reflected wave interference, and corrected accordingly. The corrections ensure that the positioning results are within a reasonable range, further improving the reliability of traveling wave positioning.

[0124] S73: Preliminary Impedance-Traveling Wave Fusion Localization: Constructing a preliminary impedance-traveling wave fusion localization model, combined with the obtained impedance prediction value. And traveling wave precise valuation First, preliminary fusion is performed to provide a foundation for subsequent confidence-weighted fusion and mutual coupling compensation. The preliminary impedance-traveling wave fusion positioning is as follows.

[0125] .

[0126] In the formula, For initial fusion of positioning distances; Estimated distance to fault; This is the revised precise value for traveling wave. Weighting logic: The precise value for traveling wave is more accurate and is given a higher weight of 0.6, while the impedance prediction is more robust and is given a lower weight of 0.4, achieving an initial combination of the advantages of both. Further optimization will be achieved through confidence-weighted fusion in the future.

[0127] This technical solution addresses the problems of traditional traveling wave positioning being susceptible to interference from far-end reflections and having poor synergy with impedance positioning by constructing a peer-end reflection correction factor model and using impedance-traveling wave preliminary fusion positioning. Peer-end reflection correction eliminates positioning errors caused by reflected waves at the end of the line, improving the stability of the positioning results.

[0128] Example 8.

[0129] This embodiment discloses a hybrid fault location method for dual-circuit transmission lines that combines impedance and traveling wave, wherein step S8 includes the following steps.

[0130] S81: Theoretical Mutual Coupling Residual Error Compensation: Abandoning fixed empirical coefficients without theoretical basis, a dynamic compensation model based on the inherent mutual coupling parameters of the line is adopted to eliminate residual coupling error throughout the process and form dual mutual coupling suppression by decoupling from the front-end phase mode.

[0131] By combining the dynamic mutual coupling influence coefficient η, mutual coupling compensation is performed on the initial fusion positioning distance to eliminate the positioning deviation caused by the mutual coupling effect. The mutual coupling compensation calculation model is as follows.

[0132] .

[0133] In the formula, This is the positioning distance after mutual coupling compensation; The initial fusion positioning distance is defined by η, the dynamic mutual coupling influence coefficient, and k, the line compensation coefficient. Calibrated based on the mutual coupling characteristics of dual-circuit lines, it is suitable for lines ranging from 50 to 300 km and can be fine-tuned according to actual line parameters. The compensation logic is as follows: the stronger the mutual coupling effect, the larger η, and the greater the compensation amplitude. This offsets the traveling wave propagation deviation and impedance calculation deviation caused by mutual coupling, ensuring that the positioning distance is not affected by the residual mutual coupling effect, further improving positioning accuracy. Combined with modal decoupling, this forms a dual mutual coupling suppression mechanism, fully demonstrating its adaptability advantages to dual-circuit mutual coupling scenarios.

[0134] S82: Wavehead Identification Confidence Calculation: Construct a correlation peak sharpness correction factor model. Based on the identified effective waveheads, combine the correlation peak sharpness correction factor δ to quantify the correlation peak confidence γ, providing a reliable basis for subsequent weighted fusion. The confidence calculation is linked with the sharpness factor to form a closed-loop optimization. The correlation peak sharpness correction factor model is as follows.

[0135] .

[0136] In the formula, δ represents the confidence level of the correlation peak (value range [0,1]); δ is the correlation peak sharpness correction factor. The maximum value of the cross-correlation function corresponds to the effective wavefront position; This represents the maximum value of the cross-correlation function within the entire adaptive time window. Logical explanation: Taking into account both the sharpness and relative intensity of the wave head, The closer it is to 1, the more accurate and reliable the wave head identification is, and the stronger the reliability of the traveling wave precise value. The closer it is to 0, the greater the noise interference to the wavefront, and the greater the weight of the impedance prediction needs to be to ensure positioning robustness.

[0137] S83: Iterative window shrinking optimization to adapt to complex interference scenarios: If the confidence level γ of the relevant peak is less than 0.8, it indicates that interference still exists in the wavefront. Based on the current optimal distance result, the window is shrunk and the wavefront is re-locked, balancing real-time positioning and accuracy. The maximum number of iterations is set to 3 to avoid over-computation. 0.8 is an empirical threshold that can be calibrated according to the line parameters. The iteration process is as follows.

[0138] (1) Substitute the current traveling wave precise valuation The dynamic mutual coupling influence coefficient η and the confidence level of the correlation peak γ are used to calculate the new window width after shrinkage. And the new window center moment .

[0139] .

[0140] (2) Based on the new time window [ , Repeat step S6 to recalculate the cross-correlation function, correlation peak sharpness correction factor δ, and correlation peak confidence γ, and update the traveling wave precise estimate. dtw_final.

[0141] (3) Repeat the above iterative process until γ≥0.8 or the number of iterations reaches 3. Avoid excessive iterations that increase the amount of computation, and balance real-time performance and accuracy. After the iteration is completed, the optimized traveling wave precise estimate is obtained. and related peak confidence .

[0142] S84: Confidence-weighted fusion: Positioning distance combined with mutual coupling compensation Optimized Traveling Wave Precision Valuation When there is no iteration, directly use Fault distance estimation And the optimized confidence level of the relevant peaks Without iteration, γ is used for confidence-weighted fusion to obtain the final fault distance. The robustness of the impedance method and the high precision of the traveling wave method are deeply synergistic, and the fusion process is as follows.

[0143] (1) When A value ≥0.8 indicates that the traveling wave precision estimate has high reliability, and the fusion result emphasizes the high precision of the traveling wave method.

[0144] .

[0145] (2) When 0.6≤ When the value is less than 0.8, it indicates that there is some interference in the precise value of the traveling wave. To balance robustness and accuracy, the fusion result is corrected by introducing a mutual coupling compensation coefficient to further improve the fusion accuracy.

[0146] .

[0147] In the formula, This is the final fault distance; The optimized confidence level of the relevant peak; For the optimized traveling wave precise value; Estimated distance to fault; 0.1 is the positioning distance after mutual coupling compensation; 0.1 is the compensation weight, used to balance the impact of mutual coupling compensation on the fusion result. The value is based on the measured range of the residual mutual coupling error of the double loop, which is usually 0.05 to 0.15 times the positioning result. Taking the middle value of 0.1 can take into account both the compensation effect and the fusion stability.

[0148] S85. Fusion Result Verification: Calculate the final fault distance. Mutual coupling compensation distance Traveling wave precise valuation Impedance estimation The deviation is given by the following formula.

[0149] ; ; .

[0150] like ≤0.5km ≤0.5km If Δd3 ≤ 1km, the deviation threshold can be calibrated according to the line length and accuracy requirements. If the deviation exceeds the threshold, return to S83 to perform iterative optimization again until the deviation requirements are met, so as to ensure the accuracy and reliability of the final fault distance.

[0151] The core optimization logic for achieving mutual coupling compensation and confidence-weighted fusion solves the problems of poor synergistic effect and failure to eliminate mutual coupling residual interference in traditional hybrid methods by eliminating the residual influence of mutual coupling, dynamically quantifying confidence, optimizing iterative windows, and weighted fusion of multiple results.

[0152] This technical solution addresses the issues of residual interference from mutual coupling and poor fusion synergy in traditional hybrid positioning methods by constructing a mutual coupling compensation calculation model and a correlation peak sharpness correction factor model. Through correlation peak confidence quantification, the reliability of wavefront identification can be assessed; iterative window optimization can dynamically eliminate interference; and confidence-weighted fusion achieves deep synergy between the impedance method and the traveling wave method, thus resolving the problems of residual interference from mutual coupling and unstable results in traditional hybrid positioning.

[0153] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A dual circuit transmission line hybrid fault location method combining impedance and traveling wave, characterized in that, include: Voltage and current time-domain signals are collected before and after the fault occurs. The changes in fault current and fault voltage are calculated based on the voltage and current time-domain signals to obtain the fault current and fault voltage. Based on the fault current, the estimated fault distance is calculated using the Takagi method. Based on the voltage time-domain signal and the current time-domain signal, the traveling wave signal triggered by the fault is extracted; Based on the estimated fault distance, an adaptive time window for constraining the effective wavefront is generated; Within the adaptive time window, the wavefront characteristics of the incident wave and the reflected wave are determined based on the traveling wave signal, and the wavefront time difference between the incident wave and the reflected wave is calculated based on the wavefront characteristics. Based on the estimated fault distance and wavefront time difference, the fault location distance of the double-circuit transmission line is calculated.

2. The dual-circuit transmission line hybrid fault location method combining impedance and traveling wave according to claim 1, characterized in that Based on the phase mode transformation algorithm and the adaptive weighted fusion algorithm, a dynamic correction model for the mutual coupling influence coefficient is constructed. The input of the dynamic correction model for the mutual coupling influence coefficient is the fault current, and the output is the dynamic mutual coupling influence coefficient. The adaptive time window is adjusted based on the dynamic mutual coupling influence coefficient.

3. The dual-circuit transmission line hybrid fault location method combining impedance and traveling wave according to claim 2, characterized in that When generating the adaptive time window, the center time of the time window is calculated. The center time is determined based on the fault distance estimate and adjusted in conjunction with the dynamic mutual coupling influence coefficient.

4. The dual-circuit transmission line hybrid fault location method combining impedance and traveling wave according to claim 1, characterized in that Calculating the wavefront time difference between the incident and reflected waves includes the following steps: Within a defined adaptive time window, cross-correlation analysis is performed on the filtered forward and backward traveling wave signals to calculate the cross-correlation function, and the effective traveling wave signals are selected to shield noise and clutter interference outside the adaptive time window. By analyzing the cross-correlation function, the corresponding time lag parameter is determined. Combined with the sampling data of the traveling wave signal, the time parameter corresponding to the maximum value of the cross-correlation function is selected. A correlation peak sharpness correction factor model is constructed, and the correlation peak sharpness parameter is calculated through the correlation peak sharpness correction factor model to quantify the effective characteristics of the traveling wave signal and distinguish between the effective wavefront and the interference signal. If a wavefront is determined to be valid, the corresponding time lag parameter is the time difference between the incident wave and the reflected wave; if a wavefront is determined to be invalid, the adaptive time window is adjusted, and the wavefront time difference is calculated again.

5. The hybrid fault location method for dual-circuit transmission lines combining impedance and traveling wave as described in claim 1, characterized in that... The calculation of the estimated fault distance includes the following steps: The fault distance per unit value was calculated using the Takagi method, and the first fault distance per unit value was obtained. The fault distance per unit value is calculated using the zero-sequence improved Takagi method to obtain the second fault distance per unit value; The Erikson method was used to establish the equation for calculating the per-unit value of fault distance, and the third per-unit value of fault distance was obtained. The first fault distance per unit value, the second fault distance per unit value, and the third fault distance per unit value are used to obtain the comprehensive fault distance prediction value through adaptive weighted fusion.

6. The dual circuit transmission line hybrid fault location method combining impedance and traveling wave of claim 4, wherein The relevant peak sharpness correction factor model is as follows: ； ； ； In the formula: For the in-window cross-correlation function; Indicates time lag; N represents the number of samples within the adaptive time window; , These are the sampled values ​​of the filtered forward and backward traveling wave signals at the corresponding times; , These represent the mean values ​​of the forward and backward traveling wave signals within the window after filtering, respectively; Indicates the sampling interval; δ is the correlation peak sharpness correction factor; is the time difference between the arrival of the incident wave and the reflected wave; m represents the number of points for hysteresis sampling.

7. The dual-circuit transmission line hybrid fault location method combining impedance and traveling wave according to claim 1, characterized in that The dynamic correction model for the mutual coupling influence coefficient is as follows: ； ； ； ； ； In the formula, W is the adaptive window width; Based on the basic window width, L is the total length of the two-circuit transmission line; η is the traveling wave propagation speed; η is the dynamic mutual coupling influence coefficient; For double-loop mutual impedance; This is the positive sequence impedance of the line; λ is the mutual coupling correction coefficient, and α is the weighting coefficient; This is the load factor correction factor, where S is the current actual load on the line. This is the line's rated load; This is a fault type correction factor; This is the wavefront ambiguity correction factor; This represents the maximum amplitude of the voltage signal change when the fault occurs. This is the rated voltage of the transmission line.

8. The dual-circuit transmission line hybrid fault location method combining impedance and traveling wave of claim 5, wherein Adjusting the adaptive time window includes the following steps: The traveling wave propagation speed is calculated by using a mode-domain wave velocity method that combines mutual inductance and self-inductance parameters; Based on the estimated fault distance, determine the theoretical arrival time of the reflected traveling wave at the fault point, set the center time of the time window, and generate an adaptive time window; A dynamic correction model for the mutual coupling influence coefficient is constructed, the dynamic mutual coupling influence coefficient is calculated, and the window width of the adaptive time window is adjusted according to the dynamic mutual coupling influence coefficient.

9. The dual-circuit transmission line hybrid fault location method combining impedance and traveling wave of claim 1, wherein The extraction of the fault-triggered traveling wave signal includes the following steps: Traveling wave signal mode decoupling: Through Parker mode transformation, the three-phase voltage signals before and after the fault are separated into independent zero-axis mode components, direct-axis mode components and quadrature-axis mode components; Based on the zero-axis mode component, the direct-axis mode component, and the quadrature-axis mode component, calculate the forward incident traveling wave component and the backward reflected traveling wave component; Traveling wave signal filtering and purification: The forward incident traveling wave component and the backward reflected traveling wave component are filtered using a Butterworth high-pass filter.

10. The dual circuit transmission line hybrid fault location method combining impedance and traveling wave of claim 4, wherein It also includes the final fault distance calculation step, which specifically includes the following steps: The estimated fault distance is compensated to offset the positioning deviation caused by mutual coupling interference, and the positioning distance after mutual coupling compensation is obtained. Based on the aforementioned correlation peak sharpness correction factor, the confidence level of the correlation peak is calculated; Based on the initial fused positioning distance, the time window range is narrowed to filter out effective wavefront signals, and iterative optimization is performed until the confidence level of the relevant peak reaches a preset threshold, at which point the iteration stops; the initial fused positioning distance is calculated based on the fault distance estimate. Confidence-weighted fusion: Combining the estimated positioning distance and fault distance after mutual coupling compensation, weights are dynamically assigned based on the confidence of the relevant peaks to obtain a preliminary fusion result; Fusion result verification: Calculate the fusion deviation between the preliminary fusion result and the estimated fault distance. If the fusion deviation exceeds a preset threshold, return to the iterative optimization step until the confidence of the relevant peak reaches the preset threshold and re-optimize until the fusion deviation is less than the preset threshold, and output the final fault distance.

Images