Cable early fault positioning method based on frequency spectrum adaptive search

Abstract

The invention belongs to the field of cable fault positioning, and particularly discloses a cable early fault positioning method based on spectrum adaptive search, which comprises the following steps: constructing a fault accurate positioning model for variable time window multivariate parameter identification; determining an optimal data window length and a corresponding effective transient main frequency band lower limit; using a singular value detection algorithm to determine the arrival time of the traveling wave signal, and determining the starting time of a data window by estimating the ratio of the distance to the wave velocity corresponding to the complete attenuation of the traveling wave; determining the ending moment of the data window at the moment when the signal enters a steady state; starting from a first signal data window, firstly determining a transient main frequency in the data window by using fast Fourier transform (FFT), calculating to obtain a fault distance, then translating the data window backwards for a sampling interval, and repeating the steps until the transient main frequency reaches a lower limit of an effective frequency band. According to the method, the accuracy of fault positioning can be effectively improved, different cable fault conditions are adapted, and the robustness of a positioning algorithm is remarkably improved.

Description

Technical Field

[0001] The invention belongs to the field of cable fault location, and in particular relates to a cable early fault location method based on spectrum adaptive search. Background Art

[0002] The actual distribution line operation scenarios are complex, and their operating status is easily affected by multiple factors such as the working environment and system parameters, resulting in multi-source and mixed characteristics of their operating data, which brings great challenges to the accurate identification and location of distribution network faults. With the access of different types of high-proportion new energy and multi-loads to the distribution network, the distribution network structure has changed from a radial type to a multi-source and multi-terminal network architecture. The fault current has significant uncertainty and volatility. Traditional fault feature analysis methods and fault location principles have been difficult to adapt to the complex and diverse operating scenarios of the new distribution network. At the same time, when complex faults such as single-phase non-metallic grounding occur in the distribution network, the fault signal is complex and changeable, which puts higher requirements on fault signal sensing and measurement.

[0003] The early fault characteristics of cables are weak, complex and changeable. The electrical parameters of cables have significant frequency-dependent and time-varying characteristics. The impedance method is easily affected by factors such as transition resistance changes and line parameter fluctuations. Among the existing methods, the traveling wave method is subject to many limitations when applied, such as complex refraction and reflection laws, difficulty in wave head identification, and difficulty in quantifying wave speed. The impedance method is susceptible to factors such as early intermittent arcs and system parameter fluctuations in principle. In addition, the existing methods only consider the impact of a single factor on the ranging algorithm, which is difficult to adapt to the variable characteristics of actual distribution network faults and the complex conditions of the on-site environment. Summary of the invention

[0004] In order to address the shortcomings of the prior art, the present invention aims to provide a cable early fault location method based on spectrum adaptive search. The method starts with multivariate fault transient time domain signals, constructs a fault accurate location model with variable time window multivariate parameter identification, analyzes the mathematical relationship between transient observable electrical quantities and fault distance, and based on the time window adaptive moving algorithm under spectrum autonomy, cyclically searches for fault points along the line, significantly improving the robustness of the location algorithm.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] A cable early fault location method based on spectrum adaptive search is implemented based on the following steps:

[0007] S1. Construct a fault accurate location model with time-varying window multivariate parameter identification;

[0008] S2. Determine the optimal data window length and the corresponding lower limit of the effective transient main frequency band;

[0009] S3. Use the singular value detection algorithm to determine the arrival time of the traveling wave signal, and determine the starting time of the data window by estimating the ratio of the distance corresponding to the complete attenuation of the traveling wave to the wave speed;

[0010] S4. Determine the end time of the data window when the signal enters a steady state;

[0011] S5. Starting from the first signal data window, use the fast Fourier transform FFT to first determine the transient main frequency in the data window, calculate a fault distance, and then shift the data window backward by one sampling interval, repeating the above steps until the transient main frequency reaches the lower limit of the effective frequency band; the fault distance calculated in the last data window is the final located fault distance.

[0012] Furthermore, in step S1, the fault accurate location model for multivariate parameter identification with a time-varying window is constructed specifically including:

[0013] Based on the difference in time-frequency distribution characteristics of transient signals before and after the fault point, the line parameters are taken into account, and the difference in the modulus voltage and current components at the virtual fault point is calculated using the voltage and current at the head and end to construct a fault accurate distance measurement function; among them, the line parameters and signal data window are the main factors affecting the distance measurement results. To solve the fault distance, it is necessary to know the cable parameters and determine the signal data window.

[0014] A fault location cyclic search model including joint identification of line parameters and fault distance is established under transient time-varying windows:

[0015]

[0016] In the formula, x F To calculate the fault location, N is the number of data windows, D U (t,ω) is the integral of the time domain signal difference of the p-mode voltage before and after the virtual fault point in the data window, D I (t,ω) is the integral of the time domain signal difference of the p-mode current before and after the virtual fault point in the data window, u p_L (x,ω,t) and u p_R (x, ω, t) are the p-mode voltage time domain signals before and after the virtual fault point, i p_L (x,ω,t) and i p_R (x, ω, t) are the time domain signals of the p-mode current before and after the virtual fault point, t represents the starting time of the data window, and t 0 is the starting time of the first data window, δ is the length of each data window, t N Indicates the starting time of the last data window, ω is the transient main frequency of the signal in the data window, ω min and ω max They represent the minimum and maximum transient main frequencies of the signal in the data window, Z is the line impedance parameter, and Z min and Zmax are the minimum and maximum values ​​of line impedance parameters, Y is the line admittance, and Y min and Y max are the minimum and maximum values ​​of the line admittance parameters respectively; L represents the line length; x represents the distance.

[0017] Furthermore, determining the optimal data window length in step S2 specifically includes:

[0018] For different cable lines, the optimal data window length may be different; it is necessary to calculate the optimal data window length before the fault occurs; set several data window lengths T k for:

[0019] T k = kΔt (2);

[0020] Where Δt is the unit sampling time, k is a positive integer; T k ≤T min , T min is the minimum value of the difference between the start and end time of the time window;

[0021] The positioning error is obtained by calculating the positioning of each data window length at different virtual fault points:

[0022] E ik =|x ture -x F |; i = 1, ..., n (3);

[0023] Among them, E ik is the location error of the kth time window of the ith virtual fault point, n is the number of virtual fault points, x ture is the actual fault location;

[0024] The average error E for each data window length k for:

[0025]

[0026] The data window length with the smallest average error value is selected as the optimal data window length T, that is,

[0027] T=min(E k ) (5).

[0028] Furthermore, in step S2, the lower limit of the effective transient main frequency is determined as follows:

[0029] In order to quickly determine whether the data window position meets the judgment requirements, the transient main frequency is selected as the criterion; for each virtual fault point, the signal obtained through the optimal data window is used to extract the transient main frequency f using the fast Fourier transform FFT. main,i, take the maximum value as the lower limit of the effective transient main frequency, that is,

[0030] f main,l =max(f main,i ) (6).

[0031] Furthermore, determining the start time of the data window in step S3 specifically includes:

[0032] The starting time of the effective signal time window needs to avoid the traveling wave process, and the starting time of the data window is set as the time when the traveling wave is completely attenuated; first determine the arrival time of the traveling wave signal, obtain the traveling wave signal of the fault current or voltage through the sampling device, use the singular value decomposition method to analyze the change of the main component of the signal matrix, and determine the arrival time T of the traveling wave signal by identifying the significant change point of the singular value. 0 ;

[0033] Since the speed of the traveling wave changes during propagation, the complete decay time T of the traveling wave is determined according to the traveling wave decay curve. d , then the data window start time T start for:

[0034] T start =T 0 +T d (7).

[0035] Furthermore, determining the end time of the data window in step S4 specifically includes:

[0036] Since only the transient signal after the high-frequency component of the traveling wave is attenuated is used for fault location, the end time of the data window needs to be set at the time when the signal enters the steady state;

[0037] Extract the main frequency band f of the signal through fast Fourier transform main , if the main frequency of the signal no longer changes significantly, that is

[0038]

[0039] Among them, ε is a very small threshold, the selection of which should be determined based on the characteristics of the signal itself and the accuracy requirements to ensure the stability of the signal main frequency.

[0040] The signal is judged to enter a steady state, and this moment is defined as the end time of the data window T end .

[0041] Furthermore, the cyclic search for the fault location in step S5 is specifically as follows:

[0042] By starting the data window from the start time of the signal data window, the transient main frequency in the data window is calculated using the fast Fourier transform. If the transient main frequency is greater than the lower limit of the effective transient main frequency, the search is completed, and the fault location is calculated using the precise positioning model; if the transient main frequency is less than the lower limit of the effective transient main frequency, the data window is shifted by one sampling interval; the above steps are repeated until the transient main frequency reaches the lower limit of the effective frequency band.

[0043] The present invention has the following beneficial effects:

[0044] The present invention proposes a cable early fault location method based on spectrum adaptive search. Starting from the multivariate fault transient time domain signal, a fault accurate location model with variable time window multivariate parameter identification is constructed, the mathematical relationship between transient observable electrical quantities and fault distance is analyzed, and based on the time window adaptive moving algorithm under spectrum autonomy, the fault point is cyclically searched along the line, which significantly improves the robustness of the location algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a simulation topology diagram of the distribution network of the present invention. DETAILED DESCRIPTION

[0046] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention and are not intended to limit the scope of use of the present invention.

[0047] like Figure 1 As shown, the present invention proposes a cable early fault location method based on spectrum adaptive search, which specifically includes the following steps:

[0048] S1. Construct a fault accurate location model based on time-varying window multivariate parameter identification; specifically including:

[0049] Based on the difference in time-frequency distribution characteristics of transient signals before and after the fault point, the line parameters are considered, and the difference in the modulus voltage and current components at the virtual fault point is calculated using the voltage and current at the head and end. At the same time, the line parameters and signal data window are the main factors affecting the distance measurement results. Solving the fault distance requires knowing the cable parameters and determining the signal data window. A fault location cyclic search model that includes joint identification of line parameters and fault distance is established under the transient variable time window:

[0050]

[0051] In the formula, x F To calculate the fault location, N is the number of data windows, D U (t,ω) is the integral of the time domain signal difference of the p-mode voltage before and after the virtual fault point in the data window, D I (t,ω) is the integral of the time domain signal difference of the p-mode current before and after the virtual fault point in the data window, u p_L(x,ω,t) and u p_R (x, ω, t) are the p-mode voltage time domain signals before and after the virtual fault point, i p_L (x,ω,t) and i p_R (x, ω, t) are the time domain signals of the p-mode current before and after the virtual fault point, t represents the starting time of the data window, and t 0 is the starting time of the first data window, δ is the length of each data window, t N Indicates the starting time of the last data window, ω is the transient main frequency of the signal in the data window, ω min and ω max They represent the minimum and maximum transient main frequencies of the signal in the data window, Z is the line impedance parameter, and Z min and Z max are the minimum and maximum values ​​of line impedance parameters, Y is the line admittance, and Y min and Y max are the minimum and maximum values ​​of the line admittance parameters respectively; L represents the line length; x represents the distance.

[0052] S2. Determine the optimal data window length and the corresponding lower limit of the effective transient main frequency band.

[0053] Wherein, determining the optimal data window length includes:

[0054] For different cable lines, the optimal data window length may be different, so before a fault occurs, it is necessary to calculate the optimal data window length. k for:

[0055] T k = kΔt (2);

[0056] Where Δt is the unit sampling time, k is a positive integer; T k ≤T min , T min is the minimum value of the difference between the start and end time of the time window;

[0057] The positioning error is obtained by calculating the positioning of each data window length at different virtual fault points:

[0058] E ik =|x ture -x F |; i = 1, ..., n (3);

[0059] Among them, E ik is the location error of the kth time window of the ith virtual fault point, n is the number of virtual fault points, x ture is the actual fault location.

[0060] The average error E for each data window lengthk for:

[0061]

[0062] The data window length with the smallest average error value is selected as the optimal data window length T, that is,

[0063] T=min(E k ) (5).

[0064] The specific determination of the lower limit of the effective transient main frequency is:

[0065] In order to quickly determine whether the data window position meets the judgment requirements, the transient main frequency is selected as the criterion; for each virtual fault point, the signal obtained through the optimal data window is used to extract the transient main frequency f using the fast Fourier transform FFT. main,i , take the maximum value as the lower limit of the effective transient main frequency, that is,

[0066] f main,l =max(f main,i ) (6).

[0067] S3. Use singular value detection algorithm to determine the arrival time T of the traveling wave signal 0 At the same time, the starting time of the data window is determined by estimating the ratio of the distance corresponding to the complete attenuation of the traveling wave to the wave speed.

[0068] The starting time of the effective signal time window needs to avoid the traveling wave process, and the starting time of the data window is set as the time when the traveling wave is completely attenuated; first determine the arrival time of the traveling wave signal, obtain the traveling wave signal of the fault current or voltage through the sampling device, use the singular value decomposition method to analyze the change of the main component of the signal matrix, and determine the arrival time T of the traveling wave signal by identifying the significant change point of the singular value. 0 ;

[0069] Since the speed of the traveling wave changes during propagation, the complete decay time T of the traveling wave can be determined according to the traveling wave decay curve. d , then the data window start time T start for:

[0070] T start =T 0 +T d (7).

[0071] S4. Determine the end time of the data window when the signal enters a steady state. Specifically include:

[0072] Since only the transient signal after the high-frequency component of the traveling wave is attenuated is used for fault location, the end time of the data window needs to be set at the moment when the signal enters a steady state.

[0073] Extract the main frequency band f of the signal through fast Fourier transformmain , if the main frequency of the signal no longer changes significantly, that is

[0074]

[0075] Among them, ε is a very small threshold, the selection of which should be determined based on the characteristics of the signal itself and the accuracy requirements to ensure the stability of the signal main frequency.

[0076] The signal is judged to enter a steady state, and this moment is defined as the end time of the data window T end .

[0077] S5. Starting from the first signal data window, use the fast Fourier transform FFT to first determine the transient main frequency in the data window, calculate a fault distance, and then shift the data window backward by one sampling interval, repeat the above steps until the transient main frequency reaches the lower limit of the effective frequency band; the fault distance calculated in the last data window is the final located fault distance. Specifically:

[0078] The cyclic search of the fault location is a crucial part of the entire fault location process. By starting the data window from the start time of the signal data window, the transient main frequency in the data window is calculated using the fast Fourier transform. If the transient main frequency is greater than the lower limit of the effective transient main frequency, the search is completed and the fault location is calculated using the precise positioning model; if the transient main frequency is less than the lower limit of the effective transient main frequency, the data window is shifted by one sampling interval; repeat the above steps until the transient main frequency reaches the lower limit of the effective frequency band.

[0079] Compared with the traditional transient location method, which mainly uses single fault information within a very short time window such as the arrival time, amplitude or transient main frequency of the traveling wave, it is easily affected by the wave head identification accuracy and the signal time window selection. Starting from the multivariate fault transient time domain signal, the present invention constructs a fault accurate location model with variable time window multivariate parameter identification, analyzes the mathematical relationship between transient observable electrical quantities and fault distance, and based on the time window adaptive moving algorithm under spectrum autonomy, cyclically searches for fault points along the line, significantly improving the robustness of the location algorithm; and can confirm the fault location multiple times under different time windows until accurate fault location is achieved. This cyclic search method can effectively improve the accuracy of fault location and adapt to different cable fault conditions.

[0080] Simulation experiment:

[0081] In order to verify the effectiveness of the cable early fault location method based on spectrum adaptive search, the present invention conducts the following simulation case analysis.

[0082] 1. Simulation system construction

[0083] In PSCAD / EMTDC, Figure 1The distribution network simulation model shown in the figure. The cable line adopts the cable model YJV-8.7 / 15kV, and its line parameters are as follows: the line length is l 1 =5km, l 2 =5km, l 3 =15km, l 4 =10km, l 5 =10km, l 6 =10km. Unit length impedance parameter Z 0 =0.15+j0.4Ω / km, unit length admittance parameter Y 0 =j4.5×10 -6 S / km; j is an imaginary unit, which represents the reactive power in the power system. Monitoring devices are installed at the head end of each line to collect voltage and current signals.

[0084] At 0.3s, on line l 2 A single-phase grounding early fault is set at 3 km from the head end. The sampling frequency is set to 20 kHz and the total sampling time is 1 s.

[0085] 2. Fault location calculation and result analysis

[0086] 2.1 Data window length calculation

[0087] 1) Calculation of optimal data window length

[0088] According to equations (2) to (5), after calculation, the data window length with the smallest average error value is selected as 0.001s.

[0089] 2) Determination of the lower limit of effective transient main frequency

[0090] For each virtual fault point, the signal obtained through the optimal data window is used to extract the transient main frequency using Fast Fourier Transform (FFT). The lower limit of the effective frequency band f is calculated. min =800Hz.

[0091] 3) Determine the start time of the data window

[0092] The traveling wave signal of the fault current or voltage is obtained through the sampling device, and the singular value decomposition method is used to analyze the changes in the principal components of the signal matrix to determine the arrival time of the traveling wave signal. The complete attenuation time of the traveling wave is determined according to the traveling wave attenuation curve, and the data window start time T is determined. start =0.315s.

[0093] 4) Determine the end time of the data window

[0094] The main frequency band of the signal is extracted by fast Fourier transform. When it is found that the main frequency of the signal no longer changes significantly, the signal is judged to have entered a steady state, and this moment is defined as the end time of the data window. The end time of the data window T is calculated endt =0.4s.

[0095] 2.2 Cyclic search for fault location

[0096] From the first signal data window (starting time is T start , the end time is T end ) and use FFT to first determine the transient main frequency f in the data window. 1 =712Hz. Since f 1 >f min , then the fault distance x is calculated using the precise positioning model (Formula (1)) F1 = 2.78km. Then the data window is shifted back by one sampling interval, and the above steps are repeated to obtain the transient main frequency f in the next data window. 2 =734Hz and fault distance x F2 =2.83km, and so on. The obtained cycle search results are shown in Table 1.

[0097] Serial number Transient main frequency Fault distance 1 712Hz 2.78km 2 734Hz 2.83km 3 779Hz 2.92km 4 815Hz 3.02km

[0098] Table 1 Cyclic search simulation data

[0099] 2.3 Positioning Results Analysis

[0100] With the translation and calculation of the data window, new fault distance estimates are continuously obtained. After multiple cycles of searching, when the transient main frequency gradually approaches the lower limit of the effective frequency band f min When , a relatively stable fault distance estimation value is finally obtained. After multiple calculations, the final fault distance is stable at x F = about 3.02 km, which is smaller than the actual fault location of 3 km and meets the actual engineering needs, thereby verifying the effectiveness and accuracy of the cable early fault location method based on spectrum adaptive search of the present invention.

[0101] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A cable early fault location method based on spectrum adaptive search, characterized by: The method is implemented based on the following steps: S1. Construct a fault accurate location model with time-varying window multivariate parameter identification; S2. Determine the optimal data window length and the corresponding lower limit of the effective transient main frequency band; S3. Use the singular value detection algorithm to determine the arrival time of the traveling wave signal, and determine the starting time of the data window by estimating the ratio of the distance corresponding to the complete attenuation of the traveling wave to the wave speed; S4. Determine the end time of the data window when the signal enters a steady state; S5. Starting from the first signal data window, use the fast Fourier transform FFT to first determine the transient main frequency in the data window, calculate a fault distance, and then shift the data window backward by one sampling interval, repeating the above steps until the transient main frequency reaches the lower limit of the effective frequency band; the fault distance calculated in the last data window is the final located fault distance.

2. The cable early fault location method based on spectrum adaptive search according to claim 1 is characterized in that: The construction of the fault accurate location model for time-varying window multivariate parameter identification in step S1 specifically includes: Based on the difference in time-frequency distribution characteristics of transient signals before and after the fault point, the line parameters are taken into account, and the difference in the modulus voltage and current components at the virtual fault point is calculated using the voltage and current at the head and end to construct a fault accurate distance measurement function; among them, the line parameters and signal data window are the main factors affecting the distance measurement results. To solve the fault distance, it is necessary to know the cable parameters and determine the signal data window. A fault location cyclic search model including joint identification of line parameters and fault distance is established under transient time-varying windows: In the formula, x F To calculate the fault location, N is the number of data windows, D U (t,ω) is the integral of the time domain signal difference of the p-mode voltage before and after the virtual fault point in the data window, D I (t,ω) is the integral of the time domain signal difference of the p-mode current before and after the virtual fault point in the data window, u p_L (x,ω,t) and u p_R (x, ω, t) are the p-mode voltage time domain signals before and after the virtual fault point, i p_L (x,ω,t) and i p_R (x, ω, t) are the time domain signals of the p-mode current before and after the virtual fault point, t represents the starting time of the data window, t0 is the starting time of the first data window, δ is the length of each data window, t N Indicates the starting time of the last data window, ω is the transient main frequency of the signal in the data window, ω min and ω max They represent the minimum and maximum transient main frequencies of the signal in the data window, Z is the line impedance parameter, and Z min and Z max are the minimum and maximum values ​​of line impedance parameters, Y is the line admittance, and Y min and Y max are the minimum and maximum values ​​of the line admittance parameters respectively; L represents the line length; x represents the distance.

3. The cable early fault location method based on spectrum adaptive search according to claim 1 is characterized in that: In step S2, the optimal data window length is determined, which specifically includes: For different cable lines, the optimal data window length may be different; it is necessary to calculate the optimal data window length before the fault occurs; set several data window lengths T k for: T k =kΔt (2); Where Δt is the unit sampling time, k is a positive integer; T k ≤T min , T min is the minimum value of the difference between the start and end time of the time window; The positioning error is obtained by calculating the positioning of each data window length at different virtual fault points: E ik =|x ture -x F |;i=1,...,n (3); Among them, E ik is the location error of the kth time window of the ith virtual fault point, n is the number of virtual fault points, x ture is the actual fault location; The average error E for each data window length k for: The data window length with the smallest average error value is selected as the optimal data window length T, that is, T=min(E k ) (5)。 4. The cable early fault location method based on spectrum adaptive search according to claim 3 is characterized in that: In step S2, the lower limit of the effective transient main frequency is determined as follows: In order to quickly determine whether the data window position meets the judgment requirements, the transient main frequency is selected as the criterion; for each virtual fault point, the signal obtained through the optimal data window is used to extract the transient main frequency f using the fast Fourier transform FFT. main,i , take the maximum value as the lower limit of the effective transient main frequency, that is, f main,l =max(f main,i ) (6)。 5. The cable early fault location method based on spectrum adaptive search according to claim 1 is characterized in that: Determining the start time of the data window in step S3 specifically includes: The starting time of the effective signal time window needs to avoid the traveling wave process, and the starting time of the data window is set as the time when the traveling wave is completely attenuated; first determine the arrival time of the traveling wave signal, obtain the traveling wave signal of the fault current or voltage through the sampling device, use the singular value decomposition method to analyze the change of the main components of the signal matrix, and determine the arrival time T0 of the traveling wave signal by identifying the significant change point of the singular value; Since the speed of the traveling wave changes during propagation, the complete decay time T of the traveling wave is determined according to the traveling wave decay curve. d , then the data window start time T start for: T start =T0+T d (7)。 6. The cable early fault location method based on spectrum adaptive search according to claim 1 is characterized in that: Determining the end time of the data window in step S4 specifically includes: Since only the transient signal after the high-frequency component of the traveling wave is attenuated is used for fault location, the end time of the data window needs to be set at the time when the signal enters the steady state; Extract the main frequency band f of the signal through fast Fourier transform main , if the main frequency of the signal no longer changes significantly, that is Among them, ε is a very small threshold, the selection of which should be determined based on the characteristics of the signal itself and the accuracy requirements to ensure the stability of the signal main frequency. The signal is judged to enter a steady state, and this moment is defined as the end time of the data window T end .

7. The cable early fault location method based on spectrum adaptive search according to claim 1 is characterized in that: The cyclic search for the fault location in step S5 is specifically as follows: By starting the data window from the start time of the signal data window, the transient main frequency in the data window is calculated using the fast Fourier transform. If the transient main frequency is greater than the lower limit of the effective transient main frequency, the search is completed, and the fault location is calculated using the precise positioning model; if the transient main frequency is less than the lower limit of the effective transient main frequency, the data window is shifted by one sampling interval; the above steps are repeated until the transient main frequency reaches the lower limit of the effective frequency band.

Images