An AI-based intelligent management system and method for remote monitoring of power grids

Abstract

This invention discloses an artificial intelligence-based intelligent management system and method for remote monitoring of power grids, relating to the field of transmission line fault location technology. The method includes: acquiring fault traveling wave waveform data; extracting the first rise time, second rise time, and time difference between the traveling wave and the measurement points at both ends of the line; establishing a first relationship based on the principle of double-ended traveling wave ranging, describing the relationship between the time difference and the fault distance and the equivalent propagation velocity; establishing a second and third relationship based on the line's frequency variation characteristics, describing the relationship between the first rise time, second rise time, and the fault distance, the equivalent propagation velocity, and the line characteristic constant; and solving the three relationships simultaneously to obtain the equivalent propagation velocity, the fault distance, and the line characteristic constant corresponding to the current fault. This invention, by simultaneously solving the double-ended time difference and the rise time at both ends, achieves the synchronous completion of fault ranging and online calibration of the line's frequency variation characteristics, improving ranging accuracy.

Description

Technical Field

[0001] This invention relates to the field of power transmission line fault location technology, specifically an artificial intelligence-based intelligent management system and method for remote monitoring of power grids. Background Technology

[0002] Fault location in transmission lines is one of the key technologies for ensuring the safe operation of the power grid. Fault location methods based on the traveling wave principle calculate the fault distance by measuring the time difference between the arrival of the transient traveling wave generated by the fault at both ends of the line and combining this with the traveling wave propagation speed. This method has advantages such as high location accuracy and is unaffected by system operating conditions, and is widely used in high-voltage transmission lines.

[0003] However, a technical problem remains unresolved in practical applications: the propagation speed of traveling waves is not constant. Due to the frequency-varying characteristics (i.e., dispersion effect) of transmission line parameters, different frequency components in the traveling wave signal propagate at different speeds, with high-frequency components traveling faster and low-frequency components traveling slower. The traveling wave generated by a fault is a broadband signal containing a continuous spectrum. When it reaches the measurement point, the waveform is distorted, and the wavefront is broadened, forming a smoothly rising leading edge.

[0004] Existing traveling wave ranging devices typically use a fixed wave velocity, failing to consider the impact of frequency variations on the wave velocity. This simplification leads to ranging errors that vary with line length and fault spectrum: for the same fault, if the high-frequency components of the traveling wave dominate, the actual wave velocity is faster, the fixed wave velocity value is lower, and the calculated result is higher; conversely, it is lower.

[0005] To address the aforementioned issues, existing research has attempted to extract the dominant frequency of traveling waves using wavelet transform or to establish a mapping relationship between waveforms and distance using artificial intelligence methods. However, these methods are either limited by the inherent trade-off between time resolution and frequency resolution in signal processing, making it impossible to simultaneously and accurately acquire wavefront timing and dominant frequency information; or they rely on a large amount of historical fault data for training, resulting in insufficient reliability under unseen fault types, and are complex to implement, computationally intensive, and inconvenient for field deployment.

[0006] Furthermore, existing line characteristic parameters (such as frequency-varying characteristic constants) typically need to be obtained in advance through offline testing or theoretical calculations, and cannot be calibrated online in real time. As the line operating environment changes and equipment ages, these pre-obtained parameters will deviate from the actual state of the line, further affecting the accuracy of fault location.

[0007] Therefore, how to achieve adaptive correction of wave velocity and online real-time calibration of line characteristic parameters while considering frequency variation characteristics is a technical problem that urgently needs to be solved in this field.

[0008] Therefore, in order to solve the above problems, this invention proposes an artificial intelligence-based intelligent management system and method for remote monitoring of power grids. Summary of the Invention

[0009] The purpose of this invention is to provide an artificial intelligence-based intelligent management system and method for remote monitoring of power grids, in order to solve the problems raised in the prior art.

[0010] To achieve the above objectives, the present invention provides the following technical solution: An artificial intelligence-based method for remote monitoring and intelligent management of power grids includes the following steps: S1. Obtain waveform data of the fault traveling wave, and extract from the waveform data the first rise time of the traveling wave reaching the first end measurement point of the line, the second rise time of the traveling wave reaching the second end measurement point of the line, and the time difference between the two ends of the line. S2. Based on the principle of double-ended traveling wave ranging, establish the first relationship, which describes the correspondence between the time difference of the traveling wave arriving at both ends of the line and the fault distance and the equivalent propagation speed of the traveling wave. S3. Based on the line frequency variation characteristics, establish a second relationship, which describes the correspondence between the first rise time and the fault distance, the equivalent propagation speed of the traveling wave, and the line frequency variation characteristic constant. S4. Based on the line frequency variation characteristics, establish a third relational expression, which describes the correspondence between the second rise time and the fault distance, the equivalent propagation speed of the traveling wave, and the line frequency variation characteristic constant. S5. By combining the first, second, and third relational equations, the equivalent propagation velocity of the traveling wave, the fault distance, and the line frequency-varying characteristic constant corresponding to this fault can be obtained.

[0011] The extraction of the first rise time and the second rise time in S1 specifically includes: The line's operating status is monitored in real time by deploying a first traveling wave acquisition device and a second traveling wave acquisition device at both ends of the transmission line. When a fault traveling wave signal is detected, the waveform acquisition function is triggered to record the fault traveling wave waveform data at the measurement points at both ends of the line. The acquired waveform data is filtered and normalized to obtain the processed first-end waveform data and second-end waveform data. Identify the rising wavefront segment in the first-end waveform data and the second-end waveform data.

[0012] The extraction of the first rise time and the second rise time also includes: Based on the processed waveform data of the first end, determine the peak amplitude of the traveling wave front of the first end; Based on the peak amplitude, a first amplitude threshold and a second amplitude threshold are set at the first end; The first amplitude threshold of the first end is a low amplitude threshold that is lower than the peak amplitude of the first end; The second amplitude threshold of the first end is a high amplitude threshold that is higher than the first amplitude threshold of the first end and lower than the peak amplitude of the first end; The amplitude of the first-end waveform is detected in real time. When the amplitude of the first-end waveform rises from below the first amplitude threshold to reach the first amplitude threshold, the current time is recorded as T1A. When the amplitude of the waveform at the first end continues to rise to reach the second amplitude threshold at the first end, record the current time as T2A, and take the difference between T2A and T1A as the first rise time. Based on the processed waveform data of the second end, determine the peak amplitude of the traveling wave front of the second end; Based on the peak amplitude, a first amplitude threshold and a second amplitude threshold are set at the second end; The first amplitude threshold of the second end is a low amplitude threshold that is lower than the peak amplitude of the second end, and the second amplitude threshold of the second end is a high amplitude threshold that is higher than the first amplitude threshold of the second end and lower than the peak amplitude of the second end. The amplitude of the second-end waveform is detected in real time. When the amplitude of the second-end waveform rises from below the first amplitude threshold to reach the first amplitude threshold, the current time is recorded as T1B. When the waveform amplitude at the second end continues to rise to the second amplitude threshold at the second end, record the current time as T2B, and take the difference between T2B and T1B as the second rise time.

[0013] The extraction of the time difference between the arrival times of the traveling wave at both ends of the line in S1 specifically includes: By deploying a first traveling wave acquisition device and a second traveling wave acquisition device at both ends of the transmission line, the time when the fault traveling wave arrives at the first end measurement point and the time when it arrives at the second end measurement point are obtained, respectively. The two moments are subtracted to obtain the time difference between them. This time difference is used as the time difference between the arrival of the traveling wave at both ends of the line.

[0014] The first relation is as follows: When the traveling wave first reaches the first end of the line, the fault distance L, the total line length Ltotal, the equivalent propagation velocity v, and the time difference ΔT satisfy the following equation: L=(Ltotal-v·ΔT) / 2; When the traveling wave reaches the second end of the line first, the fault distance L, the total line length Ltotal, the equivalent propagation velocity v, and the time difference ΔT satisfy the following equation: L = (Ltotal + v·ΔT) / 2.

[0015] The second relation is as follows: The first rise time Δt1, the fault distance L, the equivalent propagation velocity v, and the line characteristic constant k satisfy the following equation: Δt1=(k·L) / v.

[0016] The third relation is specifically as follows: The second rise time Δt2 and the fault distance L, total line length Ltotal, equivalent propagation speed v, and line characteristic constant k satisfy the following equation: Δt2=k·(Ltotal-L) / v.

[0017] The simultaneous solution of S5 includes the following: By combining the first, second, and third equations and eliminating the fault distance L and equivalent propagation speed v, we obtain an equation for the line characteristic constant k. Solving this equation yields the line characteristic constant k corresponding to the current fault. Substituting k into the second or third relation and combining it with the first relation, the equivalent propagation speed v and the fault distance L are calculated.

[0018] An artificial intelligence-based intelligent management system for remote monitoring of power grids includes a waveform acquisition module, a threshold detection module, a time recording module, a rise time calculation module, a time difference measurement module, a storage module, a computing unit, and an output module. The waveform acquisition module is used to monitor the transmission line in real time and record the traveling wave waveform data of faults at both ends of the line. The threshold detection module is used to determine the peak amplitude of the traveling wave front at the corresponding end based on the waveform data at both ends, and to set the first amplitude threshold and the second amplitude threshold at the corresponding end based on the peak amplitude. The time recording module is used to detect the waveform amplitude at both ends in real time. When the amplitude at either end reaches the first amplitude threshold of the corresponding end, the first moment is recorded. When it reaches the second amplitude threshold, the second moment is recorded. The rise time calculation module is used to calculate the first rise time and the second rise time based on the first and second moments recorded at both ends. The time difference measurement module is used to calculate the time difference between the arrival times of the traveling wave at the two ends of the line based on the arrival times of the traveling wave at the measurement points at both ends of the line. The storage module is used to store the total length of the line; The calculation unit is used to establish the first relation, the second relation, and the third relation, and solve them simultaneously to obtain the equivalent propagation speed of the traveling wave, the fault distance, and the line frequency characteristic constant; The output module is used to output the fault distance and line characteristic constants.

[0019] Compared with the prior art, the beneficial effects of the present invention are: 1. Traditional traveling wave ranging methods assume a constant wave velocity and do not consider the differences in propagation speed of different frequency components caused by the frequency variation characteristics of the line. In reality, the traveling wave of a fault is a broadband signal, and its wave velocity varies with the spectrum distribution. Using a fixed wave velocity inevitably introduces systematic errors. This invention extracts the rise time of the wavefronts of the traveling wave at both ends of the line, establishes the relationship between the rise time at both ends and the fault distance, wave velocity, and line characteristic constant, and solves it simultaneously with the relationship between the time difference at both ends. The rise time, which is regarded as waveform distortion and suppressed or ignored in traditional methods, is transformed into an effective signal carrying frequency variation information. This allows the wave velocity to be dynamically adjusted according to the actual spectrum of each fault, thereby eliminating the ranging error caused by the fixed wave velocity assumption and improving the ranging accuracy.

[0020] 2. Traditional traveling wave ranging methods only use time difference as a single piece of information to solve for fault distance, and pre-acquire line characteristics as known constants, failing to perceive changes in line parameters due to environmental factors, aging, etc. This invention solves for fault distance, equivalent propagation velocity, and line characteristic constants simultaneously as three unknowns by combining the three relationships of time difference at both ends and rise time at both ends. In a single fault, the fault location and the current frequency-varying characteristic constant of the line are obtained simultaneously, realizing online real-time calibration of line characteristic parameters and providing a new data source for line health monitoring and aging trend assessment. Attached Figure Description

[0021] Figure 1 This is a flowchart of an artificial intelligence-based intelligent management method for remote monitoring of power grids according to the present invention. Detailed Implementation

[0022] 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, and 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.

[0023] Example: Figure 1 As shown, the present invention provides a technical solution. An artificial intelligence-based method for remote monitoring and intelligent management of power grids includes the following steps: S1. Obtain waveform data of the fault traveling wave, and extract from the waveform data the first rise time of the traveling wave reaching the first end measurement point of the line, the second rise time of the traveling wave reaching the second end measurement point of the line, and the time difference between the two ends of the line. S2. Based on the principle of double-ended traveling wave ranging, establish the first relationship, which describes the correspondence between the time difference of the traveling wave arriving at both ends of the line and the fault distance and the equivalent propagation speed of the traveling wave. S3. Based on the line frequency variation characteristics, establish a second relationship, which describes the correspondence between the first rise time and the fault distance, the equivalent propagation speed of the traveling wave, and the line frequency variation characteristic constant. S4. Based on the line frequency variation characteristics, establish a third relational expression, which describes the correspondence between the second rise time and the fault distance, the equivalent propagation speed of the traveling wave, and the line frequency variation characteristic constant. S5. By combining the first, second, and third relational equations, the equivalent propagation velocity of the traveling wave, the fault distance, and the line frequency-varying characteristic constant corresponding to this fault can be obtained.

[0024] The extraction of the first rise time and the second rise time in S1 specifically includes: The line's operating status is monitored in real time by deploying a first traveling wave acquisition device and a second traveling wave acquisition device at both ends of the transmission line. When a fault traveling wave signal is detected, the waveform acquisition function is triggered to record the fault traveling wave waveform data at the measurement points at both ends of the line. The acquired waveform data is filtered and normalized to obtain the processed first-end waveform data and second-end waveform data. Identify the rising wavefront segment in the first-end waveform data and the second-end waveform data.

[0025] The extraction of the first rise time and the second rise time also includes: Based on the processed waveform data of the first end, determine the peak amplitude of the traveling wave front of the first end; Based on the peak amplitude, a first amplitude threshold and a second amplitude threshold are set at the first end; The first amplitude threshold of the first end is a low amplitude threshold that is lower than the peak amplitude of the first end; The second amplitude threshold of the first end is a high amplitude threshold that is higher than the first amplitude threshold of the first end and lower than the peak amplitude of the first end; The amplitude of the first-end waveform is detected in real time. When the amplitude of the first-end waveform rises from below the first amplitude threshold to reach the first amplitude threshold, the current time is recorded as T1A. When the amplitude of the waveform at the first end continues to rise to reach the second amplitude threshold at the first end, record the current time as T2A, and take the difference between T2A and T1A as the first rise time. Based on the processed waveform data of the second end, determine the peak amplitude of the traveling wave front of the second end; Based on the peak amplitude, a first amplitude threshold and a second amplitude threshold are set at the second end; The first amplitude threshold of the second end is a low amplitude threshold that is lower than the peak amplitude of the second end, and the second amplitude threshold of the second end is a high amplitude threshold that is higher than the first amplitude threshold of the second end and lower than the peak amplitude of the second end. The amplitude of the second-end waveform is detected in real time. When the amplitude of the second-end waveform rises from below the first amplitude threshold to reach the first amplitude threshold, the current time is recorded as T1B. When the waveform amplitude at the second end continues to rise to the second amplitude threshold at the second end, record the current time as T2B, and take the difference between T2B and T1B as the second rise time.

[0026] The extraction of the time difference between the arrival times of the traveling wave at both ends of the line in S1 specifically includes: By deploying a first traveling wave acquisition device and a second traveling wave acquisition device at both ends of the transmission line, the time when the fault traveling wave arrives at the first end measurement point and the time when it arrives at the second end measurement point are obtained, respectively. The two moments are subtracted to obtain the time difference between them. This time difference is used as the time difference between the arrival of the traveling wave at both ends of the line.

[0027] The first relation is as follows: When the traveling wave first reaches the first end of the line, the fault distance L, the total line length Ltotal, the equivalent propagation velocity v, and the time difference ΔT satisfy the following equation: L=(Ltotal-v·ΔT) / 2; When the traveling wave reaches the second end of the line first, the fault distance L, the total line length Ltotal, the equivalent propagation velocity v, and the time difference ΔT satisfy the following equation: L = (Ltotal + v·ΔT) / 2.

[0028] The second relation is as follows: The first rise time Δt1, the fault distance L, the equivalent propagation velocity v, and the line characteristic constant k satisfy the following equation: Δt1=(k·L) / v.

[0029] The third relation is specifically as follows: The second rise time Δt2 and the fault distance L, total line length Ltotal, equivalent propagation speed v, and line characteristic constant k satisfy the following equation: Δt2=k·(Ltotal-L) / v.

[0030] The simultaneous solution of S5 includes the following: By combining the first, second, and third equations and eliminating the fault distance L and equivalent propagation speed v, we obtain an equation for the line characteristic constant k. Solving this equation yields the line characteristic constant k corresponding to the current fault. Substitute k into the second or third relational expression, and combine it with the first relational expression to calculate the equivalent propagation speed v and the fault distance L.

[0031] Specifically, the process of simultaneously solving the first, second, and third relational expressions in S5 is as follows: First, divide the second relational expression Δt1 = (k·L) / v by the third relational expression Δt2 = k·(Ltotal - L) / v to eliminate the unknowns k and v, and obtain: Δt1 / Δt2 = L / (Ltotal L); From this, the fault distance L is solved: L = [Δt1 / (Δt1 + Δt2)]×Ltotal; This formula indicates that the fault distance is uniquely determined only by the ratio of the rising times at both ends, without the need to know the wave speed and the line characteristic constant in advance.

[0032] Secondly, judge the order of arrival of the traveling wave at both ends of the line according to the size relationship between L and Ltotal / 2: If L ≤ Ltotal / 2, the fault point is close to the first end, and the traveling wave arrives at the first end first. At this time, the first relational expression takes the minus sign form: L = (Ltotal - v·ΔT) / 2; If L > Ltotal / 2, the fault point is close to the second end, and the traveling wave arrives at the second end first. At this time, the first relational expression takes the plus sign form: L = (Ltotal + v·ΔT) / 2.

[0033] Substitute the obtained L and the measured time difference ΔT into the corresponding first relational expression to solve the equivalent propagation speed v of the traveling wave: v = [Ltotal 2L] / ΔT (when L < Ltotal / 2) or v = [2L Ltotal] / ΔT (when L > Ltotal / 2); Finally, substitute the obtained L and v into the second relational expression Δt1 = (k·L) / v to obtain the line frequency-variable characteristic constant k corresponding to this fault: k = (Δt1·v) / L; Up to this point, through the measured data Δt1, Δt2, ΔT of a single fault and the known total line length Ltotal, the fault distance L, the equivalent propagation speed v, and the line frequency-variable characteristic constant k are simultaneously solved, realizing the synchronous completion of fault location and on-line calibration of line characteristics.

[0034] In the present invention, every time a fault occurs, the system automatically collects traveling wave data and simultaneously solves the k value at the current moment by联立三个关系式; The k value reflects the actual frequency-dependent characteristics of the line under current environmental conditions and aging state. As the number of faults accumulates, the system obtains a series of k-value sequences that change over time, thereby enabling continuous tracking of changes in the frequency characteristics of the line, which is the essence of online calibration.

[0035] An artificial intelligence-based intelligent management system for remote monitoring of power grids includes a waveform acquisition module, a threshold detection module, a time recording module, a rise time calculation module, a time difference measurement module, a storage module, a computing unit, and an output module. The waveform acquisition module is used to monitor the transmission line in real time and record the traveling wave waveform data of faults at both ends of the line. The threshold detection module is used to determine the peak amplitude of the traveling wave front at the corresponding end based on the waveform data at both ends, and to set the first amplitude threshold and the second amplitude threshold at the corresponding end based on the peak amplitude. The time recording module is used to detect the waveform amplitude at both ends in real time. When the amplitude at either end reaches the first amplitude threshold of the corresponding end, the first moment is recorded. When it reaches the second amplitude threshold, the second moment is recorded. The rise time calculation module is used to calculate the first rise time and the second rise time based on the first and second moments recorded at both ends. The time difference measurement module is used to calculate the time difference between the arrival times of the traveling wave at the two ends of the line based on the arrival times of the traveling wave at the measurement points at both ends of the line. The storage module is used to store the total length of the line; The calculation unit is used to establish the first relation, the second relation, and the third relation, and solve them simultaneously to obtain the equivalent propagation speed of the traveling wave, the fault distance, and the line frequency characteristic constant; The output module is used to output the fault distance and line characteristic constants.

[0036] Example 1 The technical solution of the present invention will be described in detail below with reference to specific examples.

[0037] This embodiment takes a 500kV AC transmission line as an example. The line is 100km long and has substations A and B at both ends. Traveling wave acquisition devices are deployed in the substations to monitor the line's operating status in real time and record fault traveling wave waveform data.

[0038] Suppose that a single-phase ground fault occurs on the line on a certain day, and the fault point is located 30km away from terminal A, that is, the fault distance L=30km.

[0039] Since the fault point is close to end A, the traveling wave will reach the measurement point at end A first, and then reach the measurement point at end B. According to the frequency-dependent characteristics of the line, the propagation speeds of different frequency components in the traveling wave differ. Let the equivalent propagation speed of the traveling wave corresponding to this fault be v = 0.295 km / μs (this value is within the typical range of 0.29~0.30 km / μs for overhead line traveling waves), and the line frequency-dependent characteristic constant k = 0.012 (determined by the line's geometry and physical parameters).

[0040] The theoretical measurement value can be calculated from the laws of traveling wave propagation: The time difference ΔT between the two ends of the traveling wave is obtained from the principle of double-end ranging. Since the traveling wave arrives at end A first, we have ΔT=(Ltotal-2L) / v=(100-60) / 0.295=40 / 0.295≈135.593μs.

[0041] The rise time Δt1 recorded at measurement point A is formed by the arrival time difference between high-frequency and low-frequency components, satisfying Δt1=(k·L) / v=(0.012×30) / 0.295=0.36 / 0.295≈1.2203μs.

[0042] The rise time Δt2 recorded at the measurement point B satisfies Δt2=k·(Ltotal-L) / v=0.012×70 / 0.295=0.84 / 0.295≈2.8475μs.

[0043] The aforementioned Δt1, Δt2, and ΔT are the raw data acquired by the acquisition device during actual faults. The data will then be processed according to the method of this invention.

[0044] First, the first rise time Δt1, the second rise time Δt2, and the time difference ΔT are extracted from the fault traveling wave waveform data.

[0045] The specific extraction method is as follows: the waveforms acquired at both ends are filtered and amplitude normalized preprocessed to identify the rising segment of the wavefront; the peak amplitude of the wavefronts at both ends is determined respectively, and the first amplitude threshold is set to 10% peak value and the second amplitude threshold is set to 90% peak value; amplitude changes are detected in real time, and the time T1A when the amplitude at end A first reaches the 10% peak value and the time T2A when it first reaches the 90% peak value are recorded, and Δt1=T2A-T1A is calculated; similarly, T1B and T2B at end B are recorded, and Δt2=T2B-T1B is calculated; at the same time, the time when the initial wavefront of the traveling wave is received at end A and end B is recorded, and the absolute value of the difference is ΔT.

[0046] After obtaining the measured data, three relationships are established. The first relationship is based on the principle of double-ended traveling wave ranging, describing the relationship between the time difference ΔT and the fault distance L and the equivalent propagation velocity v: when the traveling wave arrives at the first end (end A) first, L=(Ltotal-v·ΔT) / 2; when the traveling wave arrives at the second end (end B) first, L=(Ltotal+v·ΔT) / 2.

[0047] The second relation is based on the line frequency variation characteristics and describes the relationship between the first rise time Δt1 and L, v, and the line characteristic constant k: Δt1=(k·L) / v.

[0048] The third relation is also based on frequency-varying characteristics, describing the relationship between the second rise time Δt2 and L, v, and k: Δt2 = k·(Ltotal-L) / v.

[0049] Dividing the second and third relations and eliminating the unknowns k and v, we get: Δt1 / Δt2=L / (Ltotal L); Substituting the measured values ​​Δt1 = 1.2203 μs and Δt2 = 2.8475 μs, calculate the ratio: 1.2203 / 2.8475≈0.4286; Therefore, the fault distance L can be calculated: L=[Δt1 / (Δt1+Δt2)]×Ltotal≈30.00km; The result is completely consistent with the preset fault distance, and there is no need to know the wave velocity and line characteristic constants in advance.

[0050] Since L = 30km is less than Ltotal / 2 = 50km, it is determined that the traveling wave arrives at the first end first. Therefore, the minus sign form of the first relation is used: L=(Ltotal-v·ΔT) / 2; Substituting the known values ​​of L = 30 km and ΔT = 135.593 μs, we can solve for the equivalent propagation speed v: v≈0.295km / μs; Finally, substituting the obtained L and v into the second relation, we obtain the line frequency characteristic constant k corresponding to this fault: k=(Δt1·v) / L=0.012; Thus, by using the measured data Δt1, Δt2, and ΔT from a single fault, the fault distance L, equivalent propagation velocity v, and line characteristic constant k were simultaneously solved, achieving the simultaneous completion of fault location and online calibration of line frequency characteristics.

[0051] To compare the advantages of this invention, the traditional fixed wave velocity method is used for synchronization calculation. Traditional methods typically use a wave velocity of 98%–99% of the speed of light. For ease of distinction, the fixed wave velocity used in the traditional method will be denoted as vfixed. Here, vfixed = 0.3 km / μs (corresponding to 300 m / μs). Substituting this into the two-end ranging formula (traveling wave arrives at end A first): Lfixed=[(Ltotal vfixed)·ΔT] / 2=29.661km; Compared to the actual fault distance of 30km, the traditional method has an error of 339 meters, while the method of this invention has zero error (within the calculation accuracy). If the actual location of the fault is confirmed to be 30km by the line inspection, the accuracy of the method of the present invention is much higher than that of the traditional method, and at the same time the frequency-varying characteristic constant k of the current line is obtained, providing real-time parameters for subsequent fault location and line status monitoring.

[0052] Example 2 To further verify the theoretical correctness of the method of the present invention and its innovativeness that distinguishes it from the prior art, a reverse verification embodiment was designed.

[0053] This embodiment is based on the same line parameters: Ltotal = 100km, assuming the line frequency-varying characteristic constant k = 0.012, and the fault occurring 30km from point A. Assume the actual equivalent propagation speed changes to v = 0.295km / μs due to different spectral distributions. From the traveling wave propagation law, the measured data are: ΔT = 135.593μs, Δt1 = 1.2203μs, Δt2 = 2.8475μs.

[0054] Using the method of this invention, L = 30 km is first obtained from the ratio of Δt1 and Δt2, then v = 0.295 km / μs is obtained from L and ΔT, and finally k = 0.012 is obtained from L and v, thus completely restoring the preset value.

[0055] Now, assuming the traditional fixed wave velocity method uses vfixed = 0.3 km / μs, substituting it into the measured ΔT, we calculate Lfixed = 29.661 km, with an error of 339 meters.

[0056] The method of this invention is accurate and error-free. Furthermore, if the line characteristic constant k changes to 0.013 due to environmental variations, while the traditional method still uses the offline calibration value of 0.012, the ranging error will further increase. This invention, by calculating k in real time for each fault, can adaptively track changes in line parameters, fundamentally overcoming the systematic errors caused by inaccurate offline parameters.

[0057] The above embodiments fully demonstrate that by combining the three relationships of time difference between the two ends and rise time at both ends, the present invention solves the fault distance, equivalent propagation speed and line characteristic constant as three unknowns simultaneously. This not only achieves high-precision fault location but also realizes online real-time calibration of line frequency-varying characteristics, providing a brand-new technical solution for remote monitoring of power grids.

[0058] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. An artificial intelligence-based intelligent management method for remote monitoring of power grids, characterized in that: Includes the following steps: S1. Obtain waveform data of the fault traveling wave, and extract from the waveform data the first rise time of the traveling wave reaching the first end measurement point of the line, the second rise time of the traveling wave reaching the second end measurement point of the line, and the time difference between the two ends of the line. S2. Based on the principle of double-ended traveling wave ranging, establish the first relationship, which describes the correspondence between the time difference of the traveling wave arriving at both ends of the line and the fault distance and the equivalent propagation speed of the traveling wave. S3. Based on the line frequency variation characteristics, establish a second relationship, which describes the correspondence between the first rise time and the fault distance, the equivalent propagation speed of the traveling wave, and the line frequency variation characteristic constant. S4. Based on the line frequency variation characteristics, establish a third relational expression, which describes the correspondence between the second rise time and the fault distance, the equivalent propagation speed of the traveling wave, and the line frequency variation characteristic constant. S5. By combining the first, second, and third relational equations, the equivalent propagation velocity of the traveling wave, the fault distance, and the line frequency-varying characteristic constant corresponding to this fault can be obtained.

2. The intelligent management method for remote monitoring of power grids based on artificial intelligence according to claim 1, characterized in that: The extraction of the first rise time and the second rise time in S1 specifically includes: The line's operating status is monitored in real time by deploying a first traveling wave acquisition device and a second traveling wave acquisition device at both ends of the transmission line. When a fault traveling wave signal is detected, the waveform acquisition function is triggered to record the fault traveling wave waveform data at the measurement points at both ends of the line. The acquired waveform data is filtered and normalized to obtain the processed first-end waveform data and second-end waveform data. Identify the rising wavefront segment in the first-end waveform data and the second-end waveform data.

3. The intelligent management method for remote monitoring of power grids based on artificial intelligence according to claim 2, characterized in that: The extraction of the first rise time and the second rise time also includes: For each end of the first-end waveform data and the second-end waveform data, perform the following operations: Determine the peak amplitude of the traveling wave front at this end based on the waveform data at this end; Based on the peak amplitude, a first amplitude threshold and a second amplitude threshold are set at this end; The first amplitude threshold is a low amplitude threshold that is lower than the peak amplitude, and the second amplitude threshold is a high amplitude threshold that is higher than the first amplitude threshold and lower than the peak amplitude. The waveform amplitude at this end is detected in real time. When the waveform amplitude at this end rises from below the first amplitude threshold to reach the first amplitude threshold, the current moment is recorded as the first time. When the waveform amplitude at this end continues to rise to reach the second amplitude threshold, the current moment is recorded as the second time. The difference between the second time and the first time corresponding to this end is taken as the rise time of this end; The rise time obtained based on the waveform data at the first end is the first rise time, and the rise time obtained based on the waveform data at the second end is the second rise time.

4. The intelligent management method for remote monitoring of power grids based on artificial intelligence according to claim 1, characterized in that: The extraction of the time difference between the arrival times of the traveling wave at both ends of the line in S1 specifically includes: By deploying a first traveling wave acquisition device and a second traveling wave acquisition device at both ends of the transmission line, the time when the fault traveling wave arrives at the first end measurement point and the time when it arrives at the second end measurement point are obtained, respectively. The two moments are subtracted to obtain the time difference between them. This time difference is used as the time difference between the arrival of the traveling wave at both ends of the line.

5. The intelligent management method for remote monitoring of power grids based on artificial intelligence according to claim 1, characterized in that: The first relation is as follows: When the traveling wave first reaches the first end of the line, the fault distance L, the total line length Ltotal, the equivalent propagation velocity v, and the time difference ΔT satisfy the following equation: L=(Ltotal-v·ΔT) / 2; When the traveling wave reaches the second end of the line first, the fault distance L, the total line length Ltotal, the equivalent propagation velocity v, and the time difference ΔT satisfy the following equation: L = (Ltotal + v·ΔT) / 2.

6. The intelligent management method for remote monitoring of power grids based on artificial intelligence according to claim 1, characterized in that: The second relation is as follows: The first rise time Δt1, the fault distance L, the equivalent propagation velocity v, and the line characteristic constant k satisfy the following equation: Δt1=(k·L) / v.

7. The intelligent management method for remote monitoring of power grids based on artificial intelligence according to claim 6, characterized in that: The third relation is specifically as follows: The second rise time Δt2 and the fault distance L, total line length Ltotal, equivalent propagation speed v, and line characteristic constant k satisfy the following equation: Δt2=k·(Ltotal-L) / v.

8. The intelligent management method for remote monitoring of power grids based on artificial intelligence according to claim 6, characterized in that: The simultaneous solution of S5 includes the following: By combining the first, second, and third equations and eliminating the fault distance L and equivalent propagation speed v, we obtain an equation for the line characteristic constant k. Solving this equation yields the line characteristic constant k corresponding to the current fault. Substituting k into the second or third relation and combining it with the first relation, the equivalent propagation speed v and the fault distance L are calculated.

9. An artificial intelligence-based intelligent management system for remote monitoring of power grids, applied to the artificial intelligence-based intelligent management method for remote monitoring of power grids as described in any one of claims 1-8, characterized in that: It includes a waveform acquisition module, a threshold detection module, a time recording module, a rise time calculation module, a time difference measurement module, a storage module, a calculation unit, and an output module; The waveform acquisition module is used to monitor the transmission line in real time and record the traveling wave waveform data of faults at both ends of the line. The threshold detection module is used to determine the peak amplitude of the traveling wave front at the corresponding end based on the waveform data at both ends, and to set the first amplitude threshold and the second amplitude threshold at the corresponding end based on the peak amplitude. The time recording module is used to detect the waveform amplitude at both ends in real time. When the amplitude at either end reaches the first amplitude threshold of the corresponding end, the first moment is recorded. When it reaches the second amplitude threshold, the second moment is recorded. The rise time calculation module is used to calculate the first rise time and the second rise time based on the first and second moments recorded at both ends. The time difference measurement module is used to calculate the time difference between the arrival times of the traveling wave at the two ends of the line based on the arrival times of the traveling wave at the measurement points at both ends of the line. The storage module is used to store the total length of the line; The calculation unit is used to establish the first relation, the second relation, and the third relation, and solve them simultaneously to obtain the equivalent propagation speed of the traveling wave, the fault distance, and the line frequency characteristic constant; The output module is used to output the fault distance and line characteristic constants.

Images