A power transmission line fault intelligent online positioning method and positioning device
By using differential current and voltage traveling wave steepness ratio and short-time energy accumulation processing, the problem of fault location under complex line structures and interference conditions is solved, and high-precision fault point location is achieved.
Patent Information
- Application Number
- CN202611209980.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-11
- Publication Date
- 2026-11-03
AI Technical Summary
Existing two-end traveling wave ranging methods are difficult to accurately capture fault sections and calibrate the arrival time of traveling waves under complex line structures and interference conditions, resulting in low positioning accuracy.
By combining the differential current and voltage traveling wave steepness ratio with short-time energy accumulation, the rising steepness ratio of the current and voltage line-mode components is extracted. The differential operation determines the initial fault interval, and short-time energy accumulation is performed within the interval to calibrate the arrival time of the traveling wave.
It improves the accuracy of fault location and anti-interference capability, enabling precise location of fault points under complex line structures and interference conditions, and reducing wavefront timing calibration errors.
Smart Images
Figure CN123029310A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power transmission line fault location technology, specifically to an intelligent online fault location method and device for power transmission lines. Background Technology
[0002] After a transmission line fault occurs, quickly and accurately locating the fault point is an urgent need to shorten power outage time and improve power supply reliability. Traveling wave-based transmission line fault location technology is widely used in power systems because it is unaffected by system operating modes and transition resistance. Traditional single-end traveling wave ranging relies on identifying reflected waves from the fault point or the opposite bus. When there are numerous impedance discontinuities on the transmission line, the source of reflected waves is complex, wavefront identification is difficult, and the location reliability is low. Double-end traveling wave ranging calculates the fault location using the time difference between the initial traveling wave arriving at the two measurement points. Its accuracy depends on the precise calibration of the wavefront times on both sides.
[0003] In existing technologies, dual-end traveling wave localization commonly employs methods such as amplitude thresholding, zero-crossing detection, or wavelet transform modulus maxima to determine the arrival time of the traveling wave. The amplitude thresholding method is significantly affected by fault type, fault initial angle, and noise interference, making threshold setting difficult and prone to causing wavefront timing deviations. While wavelet transform methods offer multi-scale analysis capabilities, they lack unified standards for wavelet basis selection and decomposition scales, involve large computational loads, have poor real-time performance, and are prone to failure in capturing abrupt changes in wavelet coefficients under weak traveling wave signal conditions such as high-impedance grounding. Regarding the determination of fault intervals using traveling wave characteristics, the distribution pattern of traveling wave amplitude or energy along the line is commonly used. However, when there is transposition, branching, or strong electromagnetic interference on the line, the distribution of traveling wave amplitude and energy along the line no longer changes monotonically, making it difficult to reliably delineate fault intervals and thus affecting the final localization accuracy.
[0004] To address the above issues, it is necessary to solve how to reliably capture the short segment interval where the fault is located under conditions of complex line structure and significant interference, and how to accurately extract the arrival time of the traveling wave after determining the interval, especially when the traveling wave waveform is severely distorted, so as to reduce the impact of wavefront time calibration error on the positioning results. Summary of the Invention
[0005] This invention provides an intelligent online fault location method and device for transmission lines, aiming to overcome the shortcomings of existing methods in terms of ambiguous fault interval discrimination and susceptibility to interference in traveling wave arrival time calibration. By integrating the processing methods of current and voltage traveling wave steepness ratio difference and short-time energy accumulation, the accuracy and engineering adaptability of transmission line fault location are improved.
[0006] To achieve the above objectives, the present invention provides the following technical solution: an intelligent online fault location method for transmission lines, comprising: acquiring three-phase current traveling wave signals and three-phase voltage traveling wave signals at each monitoring node along the transmission line; performing phase-mode transformation on the three-phase current traveling wave signals to obtain current line-mode components, and performing phase-mode transformation on the three-phase voltage traveling wave signals to obtain voltage line-mode components; extracting the rise steepness of the first wavefront leading edge in the current line-mode component, and extracting the rise steepness of the first wavefront leading edge in the voltage line-mode component; calculating the rise steepness of each... The system monitors the traveling wave steepness ratio at each monitoring node; it performs differential calculations on the traveling wave steepness ratio at each monitoring node along the line direction to obtain a differential steepness ratio sequence; based on the position of the zero-crossing point in the differential steepness ratio sequence, it determines the initial interval where the fault point is located; within the initial interval where the fault point is located, it performs short-time energy accumulation calculations on the current line modulus component to obtain a short-time energy accumulation sequence at each sampling time; it extracts the sampling time corresponding to the maximum energy value from the short-time energy accumulation sequence as the arrival time of the fault traveling wave; based on the arrival time of the fault traveling wave and the traveling wave propagation speed at the monitoring nodes at both ends of the initial interval where the fault point is located, it calculates the distance between the fault point and one end of the interval and outputs the fault location result. By using differential calculations on the traveling wave steepness ratio and zero-crossing point identification to determine the initial fault interval, it avoids the precise dependence on the initial traveling wave front time and improves the anti-interference capability of fault interval determination; by using short-time energy accumulation to determine the arrival time of the traveling wave, it effectively suppresses the influence of noise and waveform distortion, obtains a more accurate arrival time, and then, in conjunction with dual-end ranging, achieves high-precision fault location.
[0007] As a technical solution of the present invention, the step of performing phase-mode transformation on the three-phase current traveling wave signal to obtain the current line-mode component specifically includes: linearly combining the three-phase current traveling wave signal according to a preset transformation matrix, wherein the elements in the first row of the transformation matrix are all equal constants, the elements in the second row are, in order, a first constant, a negative first constant, and zero, and the elements in the third row are, in order, a first constant, zero, and a negative first constant; multiplying the transformation matrix with the column vector composed of the three-phase current traveling wave signal to obtain a transformation result containing a zero-mode component and two line-mode components; selecting the component corresponding to the second row element in the transformation result as the current line-mode component, thereby eliminating the influence of the unstable wave velocity and large attenuation of the zero-mode component, making the subsequent steepness feature extraction more stable.
[0008] Preferably, the step of extracting the rising steepness of the first wavefront leading edge in the current line mode component specifically involves: starting from the initial sampling point of the current line mode component, traversing forward along the time axis to find the first sampling point whose amplitude exceeds a preset trigger threshold, and recording this sampling point as the wavefront start point; continuing to traverse forward along the time axis from the wavefront start point to find the first sampling point whose amplitude reaches the wavefront peak point, and recording this sampling point as the wavefront peak point; using the difference between the sampling time of the wavefront peak point and the sampling time of the wavefront start point as the time difference, and using the difference between the amplitude of the wavefront peak point and the amplitude of the wavefront start point as the amplitude difference, dividing the amplitude difference by the time difference as the rising steepness of the first wavefront leading edge in the current line mode component. This method can sensitively and quantitatively characterize the abrupt change features of the wavefront leading edge, providing reliable parameters for subsequent steepness ratio calculations.
[0009] Preferably, the step of calculating the traveling wave steepness ratio at each monitoring node specifically involves: for the same monitoring node, reading the rising steepness of the first wavefront leading edge of the current line-mode component and the rising steepness of the first wavefront leading edge of the voltage line-mode component at that node; dividing the rising steepness of the first wavefront leading edge of the current line-mode component at that node by the rising steepness of the first wavefront leading edge of the voltage line-mode component at that node, and using the quotient as the traveling wave steepness ratio at that monitoring node; arranging the traveling wave steepness ratios of each monitoring node sequentially according to the arrangement order of the monitoring nodes along the line to form a traveling wave steepness ratio sequence. This ratio integrates the wavefront abrupt change information of the current and voltage traveling waves, which helps to enhance the differentiation of fault characteristics.
[0010] Furthermore, the step of performing a differential calculation on the traveling wave steepness ratio at each monitoring node along the line direction to obtain a differential steepness ratio sequence specifically involves: performing a difference calculation on the traveling wave steepness ratios of two adjacent monitoring nodes in the traveling wave steepness ratio sequence, subtracting the preceding term from the subsequent term according to the line direction, to obtain the differential steepness ratio value between each pair of adjacent monitoring nodes; arranging the differential steepness ratio values sequentially along the line direction to form the differential steepness ratio sequence. The differential calculation can highlight the abrupt changes in steepness ratio along the line, facilitating the rapid capture of fault-related feature intervals.
[0011] Furthermore, the step of determining the initial interval of the fault point based on the position of the zero-crossing point in the differential steepness ratio sequence specifically involves: traversing each differential steepness ratio value in the differential steepness ratio sequence, and finding adjacent monitoring node pairs corresponding to zero-crossing points where the value changes from positive to negative or from negative to positive; if there is one zero-crossing point in the differential steepness ratio sequence, the line interval between the adjacent monitoring node pairs corresponding to that zero-crossing point is determined as the initial interval of the fault point; if there are multiple zero-crossing points in the differential steepness ratio sequence, the length of the line interval between the adjacent monitoring node pairs corresponding to each zero-crossing point is calculated, and the shortest interval length is determined as the initial interval of the fault point. This processing logic can take into account the correct selection of intervals under both single fault and complex reflection conditions, improving the robustness of interval determination.
[0012] In a preferred embodiment of the present invention, the step of performing short-time energy accumulation calculation on the current line-mode components within the initial interval where the fault point is located to obtain the short-time energy accumulation sequence at each sampling time specifically includes: obtaining the current line-mode components collected by the monitoring nodes at both ends of the initial interval where the fault point is located, respectively denoted as the first-end current line-mode component and the second-end current line-mode component; for each sampling time of the first-end current line-mode component and the second-end current line-mode component, squaring the amplitude of the sampling time and multiplying it by a preset time window length to obtain the instantaneous energy value of the sampling time; and performing sliding window superposition on the instantaneous energy values of each sampling time of the first-end current line-mode component and the second-end current line-mode component to obtain the short-time energy accumulation value corresponding to each sampling time, forming the first-end short-time energy accumulation sequence and the second-end short-time energy accumulation sequence. Short-time energy accumulation can highlight the concentrated energy segment of the fault traveling wave, weaken the influence of random noise and local oscillations, and make the identification of the arrival time more reliable.
[0013] Preferably, the step of extracting the sampling time corresponding to the maximum energy value from the short-time energy accumulation sequence as the arrival time of the fault traveling wave specifically involves: finding the maximum value in the first short-time energy accumulation sequence and recording the sampling time corresponding to the maximum value as the arrival time of the first fault traveling wave; finding the maximum value in the second short-time energy accumulation sequence and recording the sampling time corresponding to the maximum value as the arrival time of the second fault traveling wave. Determining the arrival time based on the maximum energy value can effectively improve the accuracy of time calibration under low signal-to-noise ratio conditions.
[0014] Preferably, the step of calculating the distance between the fault point and one end of the initial interval based on the arrival time and propagation speed of the fault traveling wave at the monitoring nodes at both ends of the initial interval where the fault point is located specifically involves: calculating the absolute difference between the arrival time of the fault traveling wave at the first end and the arrival time of the fault traveling wave at the second end, denoted as the arrival time difference; multiplying the arrival time difference by the traveling wave propagation speed to obtain the traveling wave propagation distance difference; and adding the traveling wave propagation distance difference to the line length of the initial interval where the fault point is located, and dividing by two to obtain the distance between the fault point and the monitoring node at the first end. This calculation method utilizes the time difference between the two ends and the known interval length to directly provide the accurate distance of the fault point relative to one end of the interval.
[0015] An intelligent online fault location device for transmission lines includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of the aforementioned intelligent online fault location method for transmission lines. This method, through online acquisition and processing of traveling wave signals at multiple points along the line, can automatically and quickly determine the fault section and accurately measure the distance to the fault point, exhibiting fast location response and strong anti-interference capability.
[0016] The technical effects and advantages provided by the present invention in the above technical solution are as follows: By performing phase-mode transformation on the three-phase current and voltage traveling wave signals, current line-mode components and voltage line-mode components are obtained. The rising steepness of the first wavefront leading edge of each component is extracted, and the ratio of the rising steepness of the current line-mode component to that of the voltage line-mode component at each monitoring node is calculated to obtain the traveling wave waveform steepness ratio. The steepness ratios of monitoring nodes on both sides of the fault point have opposite deviation directions, while the steepness ratio fluctuations are gentle in non-fault sections. By performing differential operations on the steepness ratio sequence along the line direction, a differential steepness ratio sequence is obtained. The zero-crossing point of the differential steepness ratio sequence corresponds to the position where the polarity of the steepness ratio changes, which can directly indicate the line interval between adjacent monitoring node pairs where the fault point is located. This method does not rely on the absolute magnitude of the traveling wave amplitude, but utilizes the relative relationship between the rate of change of the wavefront leading edge of the current and voltage traveling waves. Because the traveling wave generated at the fault point has different degrees of influence on the rate of change of current and voltage, the steepness ratio shows opposite trends on both sides of the fault point, thus forming a clear zero-crossing point in the differential sequence. Even if the line has transposition or discontinuity points that cause distortion of the traveling wave amplitude, the steepness ratio characteristic can still maintain its ability to point to the fault area, which can effectively narrow the fault search range.
[0017] Within the initial interval where the fault point is located, short-time energy accumulation calculations are performed on the current line-mode components at both ends of the interval to obtain a short-time energy accumulation sequence for each sampling time. The sampling time corresponding to the maximum energy value is extracted from the sequence as the arrival time of the fault traveling wave. When the fault traveling wave arrives, the signal energy rapidly accumulates in a short time. Short-time energy accumulation calculates and superimposes energy values through a sliding window by sampling time, broadening the temporal distribution of local energy at the traveling wave front into a smooth energy sequence. The time corresponding to the maximum energy value is consistent with the actual wave front arrival time, and it is insensitive to local waveform oscillations and continuous interference. Compared to directly using abrupt amplitude changes or wavelet coefficient abrupt changes for wave front calibration, short-time energy accumulation utilizes energy concentration to highlight the energy peak caused by the fault traveling wave from the overall waveform. This avoids misjudgments caused by amplitude fluctuations or noise at a single sampling point, and can still accurately pinpoint the arrival time of the fault traveling wave under waveform distortion and weak traveling wave conditions of high-impedance faults. By combining the accurately calibrated arrival times at both ends of the interval with the traveling wave propagation speed, the distance between the fault point and one end of the interval can be calculated, achieving precise location. Attached Figure Description
[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below.
[0019] Figure 1 This is a flowchart of the intelligent online fault location method for power transmission lines; Figure 2 This is a schematic diagram of phase mode transformation of a three-phase current traveling wave signal; Figure 3 This is a flowchart of a method for determining the arrival time of a fault traveling wave based on short-term energy accumulation. Figure 4 It is the distribution curve of the steepness ratio of the traveling wave waveform at the monitoring node; Figure 5 This is a schematic diagram of the short-time energy accumulation curve and arrival time of the fault traveling wave; Figure 6 This is a schematic diagram of the current line modulus component waveforms at both ends of the fault initial interval and the arrival time of the fault traveling wave. Detailed Implementation
[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0021] See Figure 1This invention provides an intelligent online fault location method for transmission lines, comprising: acquiring three-phase current traveling wave signals and three-phase voltage traveling wave signals at each monitoring node along the transmission line; performing phase-mode transformation on the three-phase current traveling wave signals to obtain current line-mode components, and performing phase-mode transformation on the three-phase voltage traveling wave signals to obtain voltage line-mode components; extracting the rise steepness of the first wavefront leading edge in the current line-mode components, and extracting the rise steepness of the first wavefront leading edge in the voltage line-mode components; calculating the traveling wave curve at each monitoring node based on the rise steepness of the first wavefront leading edge of the current line-mode components and the rise steepness of the first wavefront leading edge of the voltage line-mode components. The differential steepness ratio is calculated by performing a differential steepness ratio operation along the line direction on the traveling wave steepness ratio at each monitoring node. Based on the position of the zero-crossing point in the differential steepness ratio sequence, the initial interval where the fault point is located is determined. Within the initial interval where the fault point is located, the short-time energy accumulation calculation is performed on the current line modulus component to obtain the short-time energy accumulation sequence at each sampling time. The sampling time corresponding to the maximum energy value is extracted from the short-time energy accumulation sequence as the arrival time of the fault traveling wave. Based on the arrival time of the fault traveling wave and the traveling wave propagation speed at the monitoring nodes at both ends of the initial interval where the fault point is located, the distance between the fault point and one end of the interval is calculated, and the fault location result is output.
[0022] Example 1: In specific implementation, please refer to Figure 2 The acquired three-phase current traveling wave signals include phase A, phase B, and phase C current traveling wave signals. Each phase current traveling wave signal corresponds to a set of instantaneous sampled values at each sampling moment. Arranging the sampled values of phase A, phase B, and phase C current traveling wave signals at the same sampling moment according to phase sequence forms a three-phase current column vector. This three-phase current column vector can be understood as a 3x1 matrix.
[0023] The preset transformation matrix is a 3x3 matrix. The first row of the transformation matrix consists of all equal constants; in one implementation, the equal constant is 1. The second row consists of the first constant, negative first constant, and zero; in one implementation, the first constant is 1, therefore the specific values of the second row are 1, -1, and 0. The third row consists of the first constant, zero, and negative first constant, using the same value assignment method; the specific values of the third row are 1, 0, and -1. The expression for the transformation matrix is as follows: In this transformation matrix, the first row has three elements all being 1, which is used to extract the zero-mode component; the second row has elements 1, -1, and 0, and the third row has elements 1, 0, and -1, which are used to extract two different line-mode components.
[0024] The transformation matrix is multiplied by the three-phase current column vector, following the rules of matrix multiplication. The multiplication process is as follows: each element of the first row of the transformation matrix is multiplied by the corresponding element of the three-phase current column vector, and then the results are summed to obtain the first element in the transformation result, which is the zero-mode component. The elements 1, -1, and 0 of the second row of the transformation matrix are multiplied by the sampled values of the traveling wave signals of phase A, phase B, and phase C currents in the three-phase current column vector, respectively, to calculate... The result is used as the second element in the transformation result, which is a line mode component, denoted as the first line mode component. The elements 1, 0, and -1 in the third row of the transformation matrix are multiplied by the sampled values of the A-phase current traveling wave signal, B-phase current traveling wave signal, and C-phase current traveling wave signal in the three-phase current column vector, respectively, to calculate... The result is taken as the third element in the transformation result, which is another linear mode component, denoted as the second linear mode component. The transformation result is also a 3-row, 1-column column vector, containing the zero-mode component, the first linear mode component, and the second linear mode component.
[0025] In the above transformation results, the component corresponding to the second row element of the transformation matrix is selected as the current line-mode component, that is, the first line-mode component is selected as the current line-mode component. The current line-mode component reflects the main high-frequency transient characteristics of the transmission line fault traveling wave signal, eliminating the unfavorable factor that the zero-mode component is greatly affected by frequency. After the three-phase current traveling wave signal is transformed by the above phase mode, each monitoring node can obtain the corresponding current line-mode component. The processing method for obtaining the voltage line-mode component by performing phase mode transformation on the three-phase voltage traveling wave signal is the same as the above process, only the input signal is replaced with the three-phase voltage traveling wave signal, and the transformation matrix structure is the same, so it will not be described again.
[0026] Example 2: In practice, the process of extracting the rising steepness of the first wavefront of the current line mode component is performed independently for each monitoring node. The current line mode component corresponds to a series of discrete sampling points, each of which includes the sampling time and the amplitude at that sampling time. The sampling times are arranged from smallest to largest along the time axis, with the starting sampling point being the earliest sampling point in the current line mode component data sequence.
[0027] The preset start-up threshold is determined based on the background noise level of the current line-mode component of the transmission line when no fault occurs. During the period after the transmission line is put into operation and no fault occurs, the current line-mode component is continuously collected, and the maximum amplitude during this period is recorded. The maximum amplitude is multiplied by a sensitivity coefficient to obtain the preset start-up threshold. The sensitivity coefficient is set to 2 to ensure that the preset start-up threshold is higher than the maximum noise amplitude during normal operation, avoiding false start-ups caused by noise, while also ensuring sensitive response to fault traveling wave fronts.
[0028] Starting from the initial sampling point of the current line mode component, the process iterates along the positive time axis, comparing the amplitude of each sampling point with a preset threshold. When the first sampling point whose amplitude exceeds the preset threshold is found, the iteration stops, and this sampling point is recorded as the starting point of the current line mode component wavefront. The sampling time of the starting point of the current line mode component wavefront is recorded as the starting time of the current line mode component wavefront, and the amplitude of the starting point of the current line mode component wavefront is recorded as the starting amplitude of the current line mode component wavefront.
[0029] Starting from the initial point of the current line-mode component wavefront, sampling points are continuously traversed along the positive time axis, monitoring the amplitude change trend of subsequent sampling points. During the traversal, if it is found that the amplitude of the next adjacent sampling point is less than the amplitude of the current sampling point, then the current sampling point is determined to be the peak point of the current line-mode component wavefront. The peak point of the current line-mode component wavefront is the top of the rising edge of the wavefront, and subsequent sampling points enter the falling edge of the wavefront. The sampling time of the peak point of the current line-mode component wavefront is recorded as the peak time of the current line-mode component wavefront, and the amplitude of the peak point of the current line-mode component wavefront is recorded as the peak amplitude of the current line-mode component wavefront.
[0030] Calculate the difference between the peak time of the current line mode component wavefront and the start time of the current line mode component wavefront to obtain the time difference of the leading edge of the current line mode component wavefront, denoted by the symbol. The difference between the peak amplitude of the current line mode component wavefront and the initial amplitude of the current line mode component wavefront is calculated to obtain the amplitude difference at the leading edge of the current line mode component wavefront, denoted by the symbol. The rise steepness of the first wavefront in the current line mode component is represented by dividing the difference in amplitude of the leading edge of the current line mode component by the time difference of the leading edge of the current line mode component. The calculation formula is as follows: ,in The unit is ampere. The unit is seconds. The unit is amperes per second.
[0031] In the same manner, all the above operations are performed on the voltage line mode components collected from the same monitoring node to obtain the rising steepness of the first wavefront leading edge in the voltage line mode component. Specifically, this includes: obtaining the background noise level of the voltage line mode component; setting the preset start threshold of the voltage line mode component to twice the maximum noise amplitude during normal operation; traversing the sampling points of the voltage line mode component to determine the wavefront start point and peak point; calculating the wavefront leading edge time difference and wavefront leading edge amplitude difference; dividing the wavefront leading edge amplitude difference by the wavefront leading edge time difference to obtain the rising steepness of the first wavefront leading edge in the voltage line mode component. The rising steepness of the voltage line mode component is represented by the symbol... This means that the calculation formula follows the same principle as described above. The unit is volts per second.
[0032] For the same monitoring node, read the steepness of the rise of the current line-mode component that has been calculated at that monitoring node. and the steepness of the rise of voltage line modulus components The quotient obtained by dividing the rise steepness of the current line-mode component by the rise steepness of the voltage line-mode component is taken as the traveling wave steepness ratio of the monitoring node. For a set of monitoring nodes distributed along the transmission line, let the total number of monitoring nodes be... The, the The ratio of the traveling wave steepness of each monitoring node is denoted as . , The value ranges from 1 to The integers are sequentially increased according to the arrangement order of the monitoring nodes along the line.
[0033] See Figure 4 In the figure, the horizontal axis represents the monitoring node numbers arranged along the transmission line, ranging from 1 to 600, and the vertical axis represents the traveling wave steepness ratio calculated at the corresponding monitoring node. The curves, marked with purple solid lines and dots, indicate the trend of the traveling wave steepness ratio of each monitoring node as a function of the node number.
[0034] As shown in the figure, within the monitoring node numbering range of approximately 1 to 240, the traveling wave waveform steepness ratio fluctuates steadily between approximately 0.48 and 0.52, exhibiting a relatively stable low-amplitude state. This reflects that the ratio of the rising steepness of the first wavefront of the current line-mode component and the voltage line-mode component at this monitoring node is basically constant. Within the range of approximately 240 to 360, the traveling wave waveform steepness ratio increases significantly, forming a peak region. The waveform steepness ratio reaches its maximum value of approximately 0.82, with the peak position roughly concentrated around node number 300. This indicates that the fault traveling wave characteristics are obvious at the monitoring node in this range, and the rising steepness of the current line-mode component relative to the voltage line-mode component varies considerably. After node number 360, the traveling wave waveform steepness ratio quickly drops back to near its initial stable level, maintaining a range of approximately 0.48 to 0.52, indicating that the fault effect has disappeared and the waveform has returned to normal. The significant peak portion of the steepness ratio curve corresponds to the interval between monitoring node numbers 240 and 360. This indicates that, according to Example 2, the steepness ratio sequence formed by calculating the steepness ratio of each monitoring node's traveling wave waveform and arranging it along the line direction contains significantly abnormal steepness ratio changes, suggesting that the fault point may be located within this interval.
[0035] Example 3: In practical implementation, the traveling wave steepness ratio sequence is composed of the traveling wave steepness ratios of each monitoring node arranged sequentially along the transmission line direction. Assume that a total of [number missing] monitoring nodes are deployed along the transmission line. The monitoring node, the first The steepness of the traveling wave waveform at each monitoring node is compared to... express, Take 1 to Integers, the steepness of the traveling wave waveform is greater than It is a dimensionless value. The steepness ratio sequence of the traveling wave is denoted as... .
[0036] When performing a difference operation on the steepness ratio sequence of a traveling wave, the operation of subtracting the preceding term from the following term is performed on adjacent terms in the sequence according to the line direction. The adjacent monitoring nodes are paired by the first The monitoring node and the first It consists of several monitoring nodes. Take 1 to Integers. The difference steepness ratio is used... The calculation method is as follows: in, Representing the The ratio of the steepness of the traveling wave waveform at each monitoring node. Representing the The ratio of the steepness of the traveling wave waveform at each monitoring node. This represents the differential steepness ratio corresponding to that pair of adjacent monitoring nodes. For all The above calculations are performed sequentially to obtain the values, resulting in a total of There are several differential steepness ratios. These differential steepness ratios are arranged in order of the line direction to form a differential steepness ratio sequence, denoted as [Sequence Number]. .
[0037] After obtaining the difference steepness ratio sequence, iterate through each difference steepness ratio value in the sequence and examine the sign change between two adjacent difference steepness ratio values. When the... Individual steepness ratio With the Individual steepness ratio When one value is positive and the other is negative, a zero-crossing point is determined to exist between the corresponding adjacent monitoring node pairs. The adjacent monitoring node pair corresponding to the later differential steepness ratio of the two adjacent differential steepness ratios whose signs change is determined as the adjacent monitoring node pair corresponding to the zero-crossing point. Specifically, if and The pair of adjacent monitoring nodes corresponding to the zero-crossing point is the first... A pair of adjacent monitoring nodes; if and The adjacent monitoring node pairs corresponding to the zero-crossing point are also the first. A pair of adjacent monitoring nodes.
[0038] The total number of zero-crossing points found in the differential steepness ratio sequence is counted. When the total number of zero-crossing points equals 1, the line interval between the adjacent monitoring node pair corresponding to that unique zero-crossing point is directly determined as the initial interval where the fault point is located. When the total number of zero-crossing points is greater than 1, for each zero-crossing point, its corresponding adjacent monitoring node pair is determined, and the length of the line interval spanned by that adjacent monitoring node pair is read. The line interval length is the actual physical distance between two adjacent monitoring nodes along the transmission line, and this physical distance data is pre-stored in the system's line parameter database. The line interval lengths corresponding to all zero-crossing points are compared, and the line interval with the smallest length value is selected as the initial interval where the fault point is located.
[0039] Example 4: In specific implementation, please refer to Figure 3 The initial interval where the fault point is located has been determined by the zero-crossing analysis of the differential steepness ratio sequence. This initial interval is defined by the first and second monitoring nodes adjacent to each other along the transmission line. The current line-mode components collected by the first monitoring node during the fault period are recorded as the first-end current line-mode components; the current line-mode components collected by the second monitoring node during the fault period are recorded as the second-end current line-mode components. The first-end current line-mode components consist of a series of discrete amplitudes arranged in chronological order of sampling time, and the second-end current line-mode components also consist of a series of discrete amplitudes arranged in chronological order of sampling time, with each sampling time corresponding to one amplitude.
[0040] Before performing short-time energy accumulation calculations, the length of the preset time window needs to be determined. The preset time window length is indicated by the symbol... It indicates that the unit is seconds. The value of is determined jointly based on the typical duration of the traveling wave front (TWF) of a transmission line fault and the system sampling frequency. The system sampling frequency is set to 1 MHz, corresponding to a sampling interval of 1 microsecond. The time span from the initial rising edge to the peak value and then back to the background noise level of the fault TWF is typically between 50 and 200 microseconds. To fully cover the main energy portion of the TWF and suppress interference introduced by noise before and after the TWF, the preset time window length is set... The value is set to 100 microseconds, which is the length of 100 sampling intervals. The reason for choosing 100 microseconds is that a 100-microsecond window contains 100 sampling points in a 1-microsecond sampling interval, which can encompass most of the transient energy of the wavefront, while not mixing in too much wavefront oscillation and background noise, thus helping to improve the signal-to-noise ratio of short-term energy accumulation.
[0041] For each sampling moment of the first-terminal current line-mode component, read the amplitude corresponding to that sampling moment and use the sign... This indicates the first current line-mode component at the first terminal. The amplitude at each sampling time, This is the sampling sequence number. The amplitude... Perform the squaring operation to obtain the square value. Then multiply the squared value by the preset time window length. The resulting product is taken as the instantaneous energy value at that sampling moment, denoted by the symbol. The expression for its calculation is: in, The first terminal current line mode component is in the... The instantaneous energy value at each sampling moment, in ampere-second squares; The first terminal current line-mode component The amplitude at each sampling moment, in amperes; The preset time window length is in seconds. Using the same method, the amplitude is read at each sampling moment of the second-terminal current line-mode component. , The sampling sequence number of the second-terminal current line-mode component is used to calculate... The second terminal current line-mode component is obtained. Instantaneous energy value at each sampling time .
[0042] After calculating the instantaneous energy value, a sliding window is used to stack the instantaneous energy values to obtain the short-term cumulative energy value at each sampling time. The length of the sliding window is measured in terms of the number of sampling points and is denoted by a symbol. express, The value is determined by the preset time window length. Divide by the sampling interval to obtain, i.e. The sliding window moves along the positive time axis in steps of 1 sampling point. For the first terminal current line-mode component, when the center of the sliding window is aligned with the first... At each sampling time, the range of sampling sequence numbers covered by the window is: ,in The instantaneous energy values corresponding to all sampling times within the window. Perform accumulation, and the accumulated result is used as the first... The short-time energy accumulation value at each sampling time is denoted as... For the sequence preceding At each sampling time, since the sliding window has not yet fully entered the data sequence, the short-time energy accumulation value is set to 0 or symmetrical extension processing is used to ensure that the length of the final first-end short-time energy accumulation sequence is consistent with the total number of sampling points of the first-end current line mode component. The same sliding window superposition operation is performed on the second-end current line mode component to obtain the second-end short-time energy accumulation sequence, where the values are denoted as... , This is the sampling time sequence number.
[0043] After obtaining the first short-time energy accumulation sequence, traverse all sequences in the sequence. The values are compared one by one to find the maximum value in the sequence. The sampling time corresponding to the maximum value is recorded, and this sampling time is taken as the arrival time of the first fault traveling wave, denoted by the symbol. This indicates that, similarly, all short-time energy accumulation sequences at the second end are traversed. Find the maximum value in the sequence, record the sampling time corresponding to the maximum value, and take this sampling time as the arrival time of the second-terminal fault traveling wave, using the symbol... Indicates the arrival time of the traveling wave at the first fault end. Arrival time of the traveling wave at the second end fault These reflect the time information of the first arrival of the fault traveling wave at the first and second end monitoring nodes, respectively. Their accuracy depends on the sampling interval and the focusing effect of short-term energy accumulation.
[0044] See Figure 5 In the figure, the horizontal axis represents the sampling time in microseconds (μs), ranging from 0 to 500 μs; the vertical axis represents the short-time energy accumulation value in ampere-seconds (A·s). The curve is the short-time energy accumulation sequence obtained by calculating the current line-mode component of the first-end monitoring node through short-time energy accumulation. The curve as a whole exhibits three distinct stages of change: Within the sampling time interval of approximately 0 to 150 μs, the short-time energy accumulation value remains at an extremely low level close to zero, indicating that the amplitude of the current line modulus component is small during this period, and no fault traveling wave signal appears. From approximately 150 μs, the short-time energy accumulation value rises rapidly, with the curve steeply upward, continuously increasing until it reaches a peak at approximately 250 μs. The peak short-time energy accumulation value is approximately 4.8 A·s, marking the concentrated release stage of the fault traveling wave front energy. The peak value is the arrival time of the fault traveling wave (marked by a red dot in the figure with a red dashed line passing vertically through it), corresponding to a sampling time of approximately 250 μs. Afterward, the short-time energy accumulation value decreases rapidly, exhibiting an exponential decay trend, and approaches zero after 350 μs, indicating that the fault traveling wave front energy has been basically dissipated, entering the wave tail and background noise stage.
[0045] Example 5: In practice, the arrival time of the traveling wave at the first fault is the time of its first arrival recorded by the first monitoring node, and the arrival time of the traveling wave at the second fault is the time of its first arrival recorded by the second monitoring node. The first and second monitoring nodes are the two endpoints of the initial interval where the fault point is located. The first monitoring node is located on one side of the initial interval along the line direction, and the second monitoring node is located on the other side. The arrival time of the traveling wave at the first fault is denoted as... The arrival time of the traveling wave at the second fault end is denoted as... Both are time values expressed in seconds. Calculation and The absolute value of the difference is denoted as the time difference of arrival, and the result is denoted by the sign. express, The unit is seconds.
[0046] The speed of traveling waves is represented by symbols. The value is expressed in meters per second. In transmission lines, the propagation speed of traveling waves is determined by the inductance and capacitance per unit length of the conductors, and its value is close to the speed of electromagnetic waves in a vacuum. For overhead transmission lines, the traveling wave propagation speed is... Values Meters per second; for cable lines, the traveling wave propagation speed varies depending on the dielectric constant of the cable insulation material. Values meters per second to A constant value within the range of meters per second. In practical implementation, the corresponding traveling wave propagation speed value is pre-stored according to the actual type of transmission line.
[0047] Time difference of arrival Multiply by the speed of the traveling wave The resulting product is denoted as the difference in traveling wave propagation distance, and is represented by the symbol... It indicates that the unit is meters. The traveling wave propagation distance difference represents the distance the traveling wave travels along the line within the time interval of the arrival time difference. The line length of the initial section where the fault point is located is represented by the symbol... The physical quantity represented is in meters. The value is obtained by pre-measuring the tower coordinates and line direction of the transmission line, and its value is the actual length of the conductor along the transmission line between the installation positions of the first and second end monitoring nodes.
[0048] Calculate the distance from the fault point to the first monitoring node, using the symbol... It indicates that the unit is meters. The calculation formula is: in, It is the length of the line in the initial section where the fault point is located. It is the difference in the propagation distance of the traveling waves. This is the distance between the fault point and the first monitoring node. The calculated distance... Values and line length When comparing the values, Set it to 0 when it is less than 0, and when Greater than When to set it as This limits the location of the fault to within the initial range.
[0049] The calculated distance from the fault point to the first monitoring node Combined with the line tower number or geographical coordinates of the first monitoring node, the information is converted into absolute location information on the transmission line, which is then output as the fault location result. The fault location result includes the tower section where the fault point is located, the distance to the nearest tower, and the corresponding line kilometer marker.
[0050] See Figure 6 In the figure, the horizontal axis represents the sampling time in microseconds (μs), ranging from 0 to 600 μs; the vertical axis represents the current amplitude in amperes (A), ranging from approximately -70A to 70A. The figure contains two curves, representing the current line modulus component at the first end (solid line) and the current line modulus component at the second end (dashed line), corresponding to the current line modulus component data collected by the two endpoint monitoring nodes in the initial interval where the fault point is located.
[0051] As shown in the figure, the waveform of the current line-mode component at the first end exhibits a distinct wavefront at approximately 100 μs. The amplitude rapidly rises to its peak value and then decays exponentially, falling back to near the background noise level around 200 μs, reflecting the typical transient characteristics of a fault traveling wave signal. The waveform of the current line-mode component at the second end also exhibits a wavefront at approximately 180 μs. Subsequently, the amplitude also rises rapidly and gradually decays to the background noise level. The waveform shape is similar to the first end, but the wavefront appears significantly later.
[0052] The figure shows two vertical dashed lines indicating the sampling times of the initial wavefronts of the current line mode components at the first and second ends, respectively. The initial wavefront at the first end is approximately 100 μs, and at the second end it is approximately 180 μs. This time difference reflects the transmission characteristic of the fault traveling wave arriving at the first monitoring node first and then at the second monitoring node.
[0053] According to the technical solution of Embodiment 5, the arrival time of the first-end fault traveling wave... The arrival time of the second-terminal fault traveling wave is approximately 100 μs. The time is approximately 180 μs. Calculate the absolute value of the difference between the two. Approximately 80 μs. Combined with the known traveling wave propagation speed... The difference in traveling wave propagation distance can be calculated. And then through the formula Determine the exact location of the fault point from the first monitoring node.
[0054] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.
Claims
1. A method and device for intelligent online fault location of transmission lines, characterized in that, include: Collect three-phase current traveling wave signals and three-phase voltage traveling wave signals at each monitoring node along the transmission line; The three-phase current traveling wave signal is subjected to phase mode transformation to obtain the current line mode component, and the three-phase voltage traveling wave signal is subjected to phase mode transformation to obtain the voltage line mode component. Extract the steepness of the first wavefront in the current line mode component and extract the steepness of the first wavefront in the voltage line mode component. The traveling wave steepness ratio at each monitoring node is calculated based on the steepness of the first wavefront of the current line mode component and the steepness of the first wavefront of the voltage line mode component. The traveling wave steepness ratio at each monitoring node is differentially calculated along the line direction to obtain a differential steepness ratio sequence; The initial interval where the fault point is located is determined based on the position of the zero-crossing point in the differential steepness ratio sequence; Within the initial interval where the fault point is located, short-time energy accumulation calculation is performed on the current line mode component to obtain the short-time energy accumulation sequence at each sampling time. The sampling time corresponding to the maximum energy value is extracted from the short-time energy accumulation sequence as the arrival time of the fault traveling wave; Based on the arrival time and propagation speed of the fault traveling wave at the monitoring nodes at both ends of the initial interval where the fault point is located, the distance between the fault point and one end of the interval is calculated, and the fault location result is output.
2. The intelligent online fault location method for transmission lines according to claim 1, characterized in that, The step of performing phase-mode transformation on the three-phase current traveling wave signal to obtain the current line-mode component specifically includes: The three-phase current traveling wave signals are linearly combined according to a preset transformation matrix. The elements in the first row of the transformation matrix are all equal constants, the elements in the second row are the first constant, the negative first constant, and zero, and the elements in the third row are the first constant, zero, and the negative first constant, respectively. Multiplying the transformation matrix by a column vector composed of three-phase current traveling wave signals yields a transformation result containing a zero-mode component and two line-mode components. The component corresponding to the element in the second row of the transformation result is selected as the current line mode component.
3. The intelligent online fault location method for transmission lines according to claim 1, characterized in that, The step of extracting the rising steepness of the first wavefront in the current line mode component specifically includes: Starting from the initial sampling point of the current line mode component, traverse along the time axis in the positive direction to find the first sampling point whose amplitude exceeds the preset start threshold, and record this sampling point as the wavefront start point; Starting from the wavefront starting point, continue traversing along the positive time axis to find the first sampling point that satisfies the amplitude reaching the wavefront peak point, and record this sampling point as the wavefront peak point; The time difference is the difference between the sampling time of the peak point of the wavefront and the sampling time of the wavefront start point, and the amplitude difference is the difference between the amplitude of the peak point of the wavefront and the amplitude of the wavefront start point. The value obtained by dividing the amplitude difference by the time difference is used as the rising steepness of the first wavefront leading edge in the current line mode component.
4. The intelligent online fault location method for transmission lines according to claim 1, characterized in that, The step of calculating the steepness ratio of the traveling wave waveform at each monitoring node specifically includes: For the same monitoring node, read the steepness of the rise of the first wavefront of the current line mode component and the steepness of the rise of the first wavefront of the voltage line mode component at that node. The quotient obtained by dividing the steepness of the first wavefront of the current line mode component at the node by the steepness of the first wavefront of the voltage line mode component at the node is used as the traveling wave waveform steepness ratio at the monitoring node. According to the arrangement order of each monitoring node along the line, the traveling wave steepness ratios of each monitoring node are arranged sequentially to form a traveling wave steepness ratio sequence.
5. The intelligent online fault location method for transmission lines according to claim 4, characterized in that, The step of performing differential calculations on the traveling wave steepness ratio at each monitoring node along the line direction to obtain a differential steepness ratio sequence specifically includes: The difference between the travel wave steepness ratios of two adjacent monitoring nodes in the traveling wave steepness ratio sequence is calculated by subtracting the preceding term from the following term according to the line direction, to obtain the differential steepness ratio between each pair of adjacent monitoring nodes. The differential steepness ratios are arranged sequentially according to the line direction to form the differential steepness ratio sequence.
6. The intelligent online fault location method for transmission lines according to claim 1, characterized in that, The step of determining the initial interval where the fault point is located based on the position of the zero-crossing point in the differential steepness ratio sequence specifically includes: Traverse each differential steepness ratio value in the differential steepness ratio sequence and find the adjacent monitoring node pairs corresponding to the zero-crossing points where the values change from positive to negative or from negative to positive. If there is a zero-crossing point in the differential steepness ratio sequence, then the line interval between the adjacent monitoring node pairs corresponding to the zero-crossing point is determined as the initial interval where the fault point is located. If there are multiple zero-crossing points in the differential steepness ratio sequence, the line interval length between adjacent monitoring node pairs corresponding to each zero-crossing point is calculated, and the shortest interval length is determined as the initial interval where the fault point is located.
7. The intelligent online fault location method for transmission lines according to claim 1, characterized in that, The step of performing short-time energy accumulation calculation on the current line mode component within the initial interval where the fault point is located, and obtaining the short-time energy accumulation sequence at each sampling time, specifically includes: The current line mode components collected by the monitoring nodes at both ends of the initial interval where the fault point is located are obtained and denoted as the first end current line mode component and the second end current line mode component, respectively. For each sampling moment of the first end current line mode component and the second end current line mode component, the square of the amplitude of the sampling moment is multiplied by the preset time window length to obtain the instantaneous energy value of the sampling moment; The instantaneous energy values of the first-end current line mode component and the second-end current line mode component at each sampling time are superimposed using a sliding window to obtain the short-time energy accumulation value corresponding to each sampling time, thus forming the first-end short-time energy accumulation sequence and the second-end short-time energy accumulation sequence.
8. The intelligent online fault location method for transmission lines according to claim 7, characterized in that, The step of extracting the sampling time corresponding to the maximum energy value from the short-time energy accumulation sequence as the arrival time of the fault traveling wave specifically includes: Find the maximum value from the first short-time energy accumulation sequence and record the sampling time corresponding to the maximum value as the arrival time of the first fault traveling wave. Find the maximum value from the short-time energy accumulation sequence at the second end, and record the sampling time corresponding to the maximum value as the arrival time of the fault traveling wave at the second end.
9. The intelligent online fault location method for transmission lines according to claim 8, characterized in that, The step of calculating the distance between the fault point and one end of the interval based on the arrival time and propagation speed of the fault traveling wave at the monitoring nodes at both ends of the initial interval where the fault point is located is specifically as follows: Calculate the absolute difference between the arrival time of the first-end fault traveling wave and the arrival time of the second-end fault traveling wave, and denot it as the arrival time difference; Multiply the arrival time difference by the traveling wave propagation speed to obtain the traveling wave propagation distance difference; Add the line length of the initial interval where the fault point is located to the difference in traveling wave propagation distance, and then divide by two to obtain the distance between the fault point and the first end monitoring node.
10. An intelligent online fault location device for transmission lines, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the intelligent online fault location method for transmission lines as described in any one of claims 1 to 9.