A Cable Defect Location Method and Device Based on Data Dimensionality Reduction

Abstract

The present application provides a cable defect location method and device based on data dimensionality reduction. The method obtains incident signals and reflected signals from a signal receiver. Based on a preset number of samples, the number of discrete times and the discrete time intervals of the incident signals and the reflected signals are determined. Based on the number of discrete times and the discrete time intervals, the variable time interval of the time integration variable of the time-frequency domain distribution is determined, so as to determine the time integration variables of each discrete time corresponding to the number of samples through the variable time interval, and a first dimensionality reduction matrix is generated. Based on the correspondence between the time integration variable and the frequency and the first dimensionality reduction matrix, the dimensionality-reduced time-frequency domain distributions of the incident signals and the reflected signals are determined. The cross-correlation function is processed for dimensionality reduction to obtain a dimensionality-reduced cross-correlation function, and a defect location curve corresponding to the dimensionality-reduced cross-correlation function is generated. According to the defect location curve, the defect position of the cable to be measured is determined.

Description

Technical Field

[0001] This application relates to the technical field of power system detection, and in particular, to a cable defect location method and device based on data dimensionality reduction. Background Art

[0002] Cables play an extremely important role in the power system. After the cross-linked polyethylene material technology matured, the electrical strength and mechanical hardness of cables insulated with this material have been greatly improved. In addition, compared with the overhead line transmission method, cables occupy less land area, which is of great significance in urban power transmission and distribution. However, long-term use of cables is prone to adverse factors such as insulation aging and performance deterioration, which can cause local defects in the cables, resulting in phenomena such as local overheating, local damage, and partial discharge of the cables. If not discovered and eliminated in time, it may develop into a permanent fault, posing a threat to the stability of the power system.

[0003] To prevent serious accidents from occurring, each cable line needs to be regularly inspected. Since cables can be several kilometers long, the traditional manual inspection method is extremely time-consuming and laborious, and a new cable defect location technology based on the reflection method has been developed. Currently, cable defect location mainly relies on adding an incident signal at one end of the cable with the circuit disconnected and locating the cable defect based on the collected reflection signal. However, when obtaining the defect location from the collected reflection signal, the amount of calculation for data processing is huge, and it takes a long time of data operation to obtain the accurate defect location, resulting in a long defect location time. In the wild scenario, the maintenance personnel for cable defect location need to wait for a long time to know whether the tested cable has a defect, which affects the user experience. Summary of the Invention

[0004] Embodiments of this application provide a cable defect location method and device based on data dimensionality reduction, which are used to reduce the location time of cable defect location and improve the user experience of cable defect location.

[0005] On the one hand, this application provides a cable defect location method based on data dimensionality reduction, and the method includes:

[0006] Obtain the incident signal and the reflected signal from the signal receiver. Among them, the signal receiver is used to receive the incident signal generated by the signal transmitter at one end of the T-type interface of the cable under test, and to receive the reflected signal after the incident signal passes through the cable under test. Based on the preset number of samples, determine the number of discrete time and the discrete time interval of the incident signal and the reflected signal. Based on the number of discrete time and the discrete time interval, determine the variable time interval of the time integration variable in the time-frequency domain distribution, so as to determine the time integration variables of each discrete time corresponding to the number of samples through the variable time interval, and generate the first dimensionality reduction matrix. Based on the correspondence between the time integration variable and the frequency and the first dimensionality reduction matrix, determine the reduced time-frequency domain distribution of the incident signal and the reflected signal. According to the reduced time-frequency domain distribution and the preset rules, generate the second dimensionality reduction matrix, so as to perform dimensionality reduction processing on the cross-correlation function according to the second dimensionality reduction matrix, obtain the reduced cross-correlation function, and generate a defect location curve corresponding to the reduced cross-correlation function. According to the defect location curve, determine the defect location of the cable under test.

[0007] In an implementation manner of the present application, according to the discrete time interval and the coefficient of the time integration variable in the time-frequency domain distribution, determine the variable time interval of the time integration variable. According to the discrete time interval, determine the number of discrete times of the number of samples, and determine the time integration variables corresponding to each discrete time as the selected variables. According to the ratio of each selected variable to the variable time interval, determine the value shift bits of the time integration variables corresponding to each discrete time. Among them, the value shift bits are used to determine the corresponding incident signal frequency or reflected signal frequency under the time integration variable corresponding to the discrete time. According to the value shift bits, determine the first dimensionality reduction matrix. Among them, the columns of the first dimensionality reduction matrix are discrete times, and the rows of the first dimensionality reduction matrix are time integration variables.

[0008] In an implementation manner of the present application, according to the product of the preset constant, the number of discrete time, and the discrete time interval, determine the maximum value of the time integration variable. According to the maximum value of the time integration variable, determine the value range of the time integration variable. Determine whether the values of the time integration variables corresponding to each discrete time match the value range. In the case where it is determined that the time integration variables corresponding to each discrete time do not match the value range, generate constant adjustment information and send the constant adjustment information to the user terminal. Based on the user's operation on the user terminal, determine the feedback information of the constant adjustment information, so as to adjust the preset constant according to the feedback information.

[0009] In an implementation manner of the present application, the time-frequency domain distribution values corresponding to the frequencies at each time for the reduced-dimension time-frequency domain distribution are determined. The maximum value among the time-frequency domain distribution values is determined, and in chronological order, the maximum value among the time-frequency domain distribution values is used to generate a second reduced-dimension matrix. The second reduced-dimension matrix includes an incident signal time-frequency reduced-dimension sub-matrix and a reflected signal time-frequency reduced-dimension sub-matrix. According to the central time of the signal duration of the incident signal, the reduced-dimension time-frequency domain distribution is normalized. Based on the normalized reduced-dimension time-frequency domain distribution and the second reduced-dimension matrix, the reduced-dimension cross-correlation function between the incident signal and the reflected signal is determined, so as to obtain the defect location curve corresponding to the cable to be tested according to the reduced-dimension cross-correlation function.

[0010] In an implementation manner of the present application, according to the material of the cable to be tested, the signal propagation speed of the cable to be tested is determined. Based on the signal propagation speed of the cable to be tested and the corresponding relationship between the signal propagation time and the distance from the cable head end, the cross-correlation function is converted into a corresponding distance-related function. The distance-related function is used to represent the relationship between the distance and the signal correlation degree, and the signal correlation degree is used to represent the correlation between the reflected signal and the incident signal. According to the distance-related function, the defect location curve corresponding to the cable to be tested is determined, so as to determine the distance between the defect of the cable to be tested and the cable head end according to the defect location curve.

[0011] In an implementation manner of the present application, the maximum value in the defect location curve is determined. The position of the maximum value at the starting point of the defect location curve is used as the cable head end, and the position of the maximum value at the ending point of the defect location curve is used as the cable tail end. The curve waveforms within a preset interval at each maximum value are matched with the curve waveform of the cable head end and / or the curve waveform of the cable tail end. The position corresponding to the maximum value with a matching degree greater than the preset value in the matching result is determined as the defect position.

[0012] In an implementation manner of the present application, the test information sent by the management and control terminal is obtained. The test information at least includes: cable open-circuit data, cable position data. In the case of determining that the cable to be tested in the test information is open-circuited, incident signal generation information is generated and sent to the signal transmitter. The incident signal generation information includes Gaussian envelope data and linear chirp data.

[0013] In an implementation manner of the present application, based on the function composed of the time integration variable and the frequency in the Wigner distribution, a third reduced-dimension matrix is generated. The number of rows of the third reduced-dimension matrix is the same as the number of columns of the first reduced-dimension matrix. The rows of the third reduced-dimension matrix are the time integration variables, and the columns of the third reduced-dimension matrix are the frequencies. Based on the matrix multiplication processing of the first reduced-dimension matrix and the third reduced-dimension matrix, the reduced-dimension time-frequency domain distribution of the incident signal and the reflected signal is determined.

[0014] On the other hand, an embodiment of the present application provides a cable defect location device based on data dimensionality reduction, and the device includes:

[0015] At least one processor; and a memory communicatively connected to the at least one processor. Wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to:

[0016] Obtain an incident signal and a reflected signal from a signal receiver. The signal receiver is used to receive the incident signal generated by the signal transmitter at one end of the T-shaped interface of the cable to be measured, and to receive the reflected signal after the incident signal passes through the cable to be measured. Based on a preset number of samples, determine the number of discrete time and the discrete time interval of the incident signal and the reflected signal. Based on the number of discrete time and the discrete time interval, determine the variable time interval of the time integration variable of the time-frequency domain distribution, so as to determine the time integration variable of each discrete time corresponding to the number of samples through the variable time interval, and generate a first dimensionality reduction matrix. Based on the correspondence between the time integration variable and the frequency and the first dimensionality reduction matrix, determine the reduced time-frequency domain distribution of the incident signal and the reflected signal. According to the reduced time-frequency domain distribution and a preset rule, generate a second dimensionality reduction matrix, so as to perform dimensionality reduction processing on the cross-correlation function according to the second dimensionality reduction matrix to obtain a reduced cross-correlation function, and generate a defect location curve corresponding to the reduced cross-correlation function. Determine the defect location of the cable to be measured according to the defect location curve.

[0017] In an implementation manner of the present application, the at least one processor is specifically capable of:

[0018] Determine the variable time interval of the time integration variable according to the discrete time interval and the coefficient of the time integration variable in the time-frequency domain distribution. According to the discrete time interval, determine the number of discrete times of the number of samples, and determine the time integration variable corresponding to each discrete time as the selected variable. According to the ratio of each selected variable to the variable time interval, determine the value shift number of the time integration variable corresponding to each discrete time. The value shift number is used to determine the corresponding incident signal frequency or reflected signal frequency under the time integration variable corresponding to the discrete time. Determine the first dimensionality reduction matrix according to the value shift number. The columns of the first dimensionality reduction matrix are discrete times, and the rows of the first dimensionality reduction matrix are time integration variables.

[0019] Through the above solution, the present application can simplify the calculation formula of the time-frequency domain distribution through dimensionality reduction, improve the efficiency of defect location, and at the same time simplify the calculation process of determining cable defects by the time-frequency domain reflection method. The calculation of the three-dimensional cross-correlation function with a large amount of calculation is simplified to a two-dimensional cross-correlation function, greatly reducing the storage amount of data and the time consumed by the calculation. Moreover, accurate positioning of cable defects can be achieved based on the two-dimensional cross-correlation function after dimensionality reduction. Through the above solution, the present application can improve the positioning speed of cable defects, improve the calculation efficiency of cable defect location, save the calculation amount, and improve the user experience of cable defect location users. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The schematic embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation to the present application. In the drawings:

[0021] Figure 1 is a schematic flow chart of a method for cable defect location based on data dimensionality reduction in an embodiment of the present application;

[0022] Figure 2 is a schematic diagram of a method for cable defect location based on data dimensionality reduction in an embodiment of the present application;

[0023] Figure 3 is another schematic diagram of a method for cable defect location based on data dimensionality reduction in an embodiment of the present application;

[0024] Figure 4 is still another schematic diagram of a method for cable defect location based on data dimensionality reduction in an embodiment of the present application;

[0025] Figure 5 is yet another schematic diagram of a method for cable defect location based on data dimensionality reduction in an embodiment of the present application;

[0026] Figure 6 is yet another schematic diagram of a method for cable defect location based on data dimensionality reduction in an embodiment of the present application;

[0027] Figure 7 is a schematic structural diagram of a cable defect location device based on data dimensionality reduction in an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments of this application and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without creative efforts belong to the scope of protection of this application.

[0029] To prevent the occurrence of serious accidents, regular maintenance is required for each cable line. Since cables can be several kilometers long, the traditional manual maintenance method is extremely time-consuming and laborious, and a new cable defect location technology based on the reflection method has been developed. Currently, the reflection method mainly includes Time Domain Reflectometry (TDR), Frequency Domain Reflectometry (FDR), and Time-frequency Domain Reflectometry (TFDR). The main differences among the three are the incident signals and the analysis fields: Time Domain Reflectometry emits a pulse signal or a step signal to the target cable and then analyzes the position of the defect on the time-domain waveform; the incident signal of Frequency Domain Reflectometry is a swept-frequency signal, and its direct analysis spectrum is the impedance spectrum in the frequency domain, etc., and then the position of the defect is obtained through integral transformation; Time-frequency Domain Reflectometry emits a linear chirp signal with a Gaussian envelope, converts the obtained time-domain reflected signal to the time-frequency domain, and the position curve of the defect can be obtained through subsequent data processing.

[0030] Time Domain Reflectometry has good positioning ability for serious defects or extreme faults such as open circuits and short circuits, but for weak defects, the defect information cannot be recognized due to the overly weak reflected signal; Frequency Domain Reflectometry can locate weak defects, but for the case where the distance between the end and the defect position is far, it cannot distinguish between the cable end and the defect; although Time-frequency Domain Reflectometry can locate weak defects and effectively distinguish between the cable end and the defect, due to the large amount of calculation involved in data processing, it takes a long time to apply.

[0031] Based on this, the embodiments of this application provide a cable defect location method and device based on data dimensionality reduction to reduce the location time of cable defect positions and improve the user experience of cable defect location.

[0032] The following will describe each embodiment of this application in detail with reference to the drawings.

[0033] The embodiments of this application provide a cable defect location method based on data dimensionality reduction, as Figure 1 shown, this method may include steps S101 - S106:

[0034] S101, the server acquires the incident signal and the reflected signal from the signal receiver.

[0035] Among them, the signal receiver is used to receive the incident signal generated by the signal transmitter at one end of the T-shaped interface of the cable under test, and is also used to receive the reflected signal after the incident signal passes through the cable under test.

[0036] In the application embodiment, the signal receiver can be an oscilloscope, the signal transmitter can be a waveform generator, the server can be a single computer, or a server cluster composed of multiple servers. The connection manners of the signal transmitter, the signal receiver, and the cable under test are as Figure 2 shown. The computer (server) is connected to the waveform generator. The computer can send the designed incident signal generation information to the waveform generator, and the waveform generator generates a chirped signal with a Gaussian envelope as the incident signal. The waveform generator sends the incident signal to the cable under test (RG58 cable) through one end of the T-shaped interface (T-joint). The other end of the T-joint is connected to the oscilloscope (signal receiver), and the oscilloscope sends the collected incident signal and transmitted signal to the computer (server).

[0037] It should be noted that the server, as the execution subject of the cable defect location method based on data dimensionality reduction, is only an exemplary existence. The execution subject is not limited to the server, and the present application does not make specific limitations in this regard.

[0038] In the embodiment of the present application, before the server acquires the incident signal and the reflected signal from the signal receiver, it includes:

[0039] The server acquires the test information sent by the management and control terminal.

[0040] Among them, the test information at least includes: cable open circuit data, cable position data.

[0041] In the actual use process, the management and control terminal can be the terminal of the management personnel responsible for disconnecting the circuit of the cable under test. After the management personnel complete the cable disconnection, they send the cable position data and the cable open circuit data to the server.

[0042] When the server determines that the cable under test in the test information is open circuited, it generates incident signal generation information and sends the incident signal generation information to the signal transmitter.

[0043] Among them, the incident signal generation information includes Gaussian envelope data and linear chirp data.

[0044] When the server determines that the cable under test has an open circuit, it can determine the incident signal generation information, which includes Gaussian envelope data and linear chirp data. Based on the Gaussian envelope data and linear chirp data, a linearly chirped incident signal with a Gaussian envelope can be generated. The specific formula is as follows:

[0045]

[0046] Among them, s(t) is the function of the incident signal to be generated, t is time, α is a constant inversely related to the signal duration, β is a constant affecting the signal frequency bandwidth, t0 is the time center of the signal, and ω0 is the angular frequency center of the signal.

[0047] In this application, considering the length of the cable and the sampling rate of the signal generator, the signal duration is selected as 50 ns (equivalent to α taking 2.21×10^16), the frequency bandwidth B s is 120 MHz, the time center t0 is 0 s, the angular frequency center ω0 is 2π×150 MHz, and β can be obtained according to the following formula:

[0048]

[0049] In the embodiment of this application, the incident signal and the reflected signal received by the signal receiver are as Figure 3 、 Figure 4 shown.

[0050] S102. The server determines the discrete time quantity and the discrete time interval of the incident signal and the reflected signal based on a preset sampling quantity.

[0051] In the embodiment of this application, the preset sampling quantity can be set by the user. For example, the sampling quantity is set to 10,000, and these 10,000 sampling points at least include all the signal components of the incident signal and the reflected signal. As Figure 4 shown, the 10,000 sampling points can start from -2×10 -8 seconds and end at a time point before 2×10 -7 seconds. The discrete time quantity corresponds to the preset sampling quantity, and the discrete time interval is the difference between adjacent sampling points. For example, the time of the first sampling point is T1, and the time of the adjacent second sampling point is T2, then the discrete time interval is T2 - T1.

[0052] S103. The server determines the variable time interval of the time integration variable of the time-frequency domain distribution based on the discrete time quantity and the discrete time interval, so as to determine the time integration variable of each discrete time corresponding to the sampling quantity through the variable time interval and generate a first dimensionality reduction matrix.

[0053] In an embodiment of the present application, the server determines a variable time interval of a time integration variable of a time-frequency domain distribution based on the discrete time quantity and the discrete time interval, so as to determine, through the variable time interval, the time integration variables of each discrete time corresponding to the sampling quantity and generate a first dimensionality reduction matrix, which specifically includes:

[0054] First, the server determines the variable time interval of the time integration variable according to the discrete time interval and the coefficient of the time integration variable in the time-frequency domain distribution. The time-frequency domain distribution is a Wigner distribution.

[0055] Secondly, the server determines each discrete time of the sampling quantity and determines the time integration variable corresponding to each discrete time as a selected variable.

[0056] The server can obtain the variable time interval of the time integration variable according to the discrete time interval and the coefficient of the time integration variable in the time-frequency domain distribution, where the specific formula of the time-frequency domain distribution is as follows:

[0057]

[0058] where W(t, ω) is the time-frequency domain distribution function, τ is the time integration variable, s * is the conjugate signal, s is the incident signal or the reflected signal, and ω is the angular frequency. The coefficient of the time integration variable in the time-frequency domain distribution is The server will determine that the variable time interval Δτ = 2Δt, where Δt is the discrete time interval.

[0059] Then, the server determines the value shift number of the time integration variable corresponding to each discrete time according to the ratio of each selected variable and the variable time interval.

[0060] where the value shift number is used to determine the corresponding frequency under the time integration variable corresponding to the discrete time.

[0061] In order to prevent the situation that the variable cannot be equivalent to the existing data during the process of transforming the independent variable t of s(t) into t and τ, the present application will construct a corresponding autocorrelation function based on the position information of the independent variable, that is, for each time point, there is a corresponding number of moving positions of the jth τ, and its moving number is defined as follows:

[0062]

[0063] where the subscript z in the above formula represents taking the integer part, and d τj is the value shift number of the jth τ.

[0064] Before the present application determines the value shift number of the time integration variable corresponding to each discrete time according to the ratio of each selected variable and the variable time interval, the method further includes:

[0065] The server determines the maximum value of the time integral variable according to the product of a preset constant, the number of discrete time intervals, and the discrete time interval.

[0066] An embodiment of the present application can determine the maximum value of the time integral variable through the following formula:

[0067] τ max = 10N·Δt

[0068] where τ max is the maximum value of the time integral variable, 10 is the preset constant, and N is the number of discrete time intervals.

[0069] The server determines the value range of the time integral variable according to the maximum value of the time integral variable.

[0070] In the embodiment of the present application, the integration range of τ is symmetric. Therefore, the absolute value of the lower bound of its negative part is also the same as that of τ max , so the value range is [-τ max , τ max

[0071] The server determines whether the values of the time integral variables corresponding to each discrete time match the value range.

[0072] In the embodiment of the present application, the server can determine whether the time integral variable corresponding to the discrete time is within the value range, so as to determine whether it matches.

[0073] When the server determines that the time integral variables corresponding to each discrete time do not match the value range, it generates constant adjustment information and sends the constant adjustment information to the user terminal.

[0074] That is, when the time integral variable corresponding to the discrete time is not within the value range, it indicates that there is an error in the value of the time integral variable. Therefore, constant adjustment information can be generated to prompt the user.

[0075] Based on the user's operation on the user terminal, the feedback information of the constant adjustment information is determined, and the preset constant is adjusted according to the feedback information.

[0076] The user can adjust the preset constant to a larger value. For example, if the original preset constant is 10 and there is a mismatch, then the preset constant can be set to 20.

[0077] Finally, the server determines the first dimensionality reduction matrix according to the number of bits of value movement.

[0078] Among them, the columns of the first dimensionality reduction matrix are discrete times, and the rows of the first dimensionality reduction matrix are time integral variables.

[0079] ​In an embodiment of the present application, after the server determines the value shift number, it can determine the corresponding values of the time integral variables at each time point, and the value is the amplitude. The first dimensionality reduction matrix generated in the above embodiment corresponds to a part of the matrix.

[0080] S104. The server determines the dimensionality-reduced time-frequency domain distribution of the incident signal and the reflected signal based on the correspondence between the time integral variable and the frequency and the first dimensionality reduction matrix.

[0081] The dimensionality-reduced time-frequency domain distribution includes time and frequency.

[0082] In an embodiment of the present application, the server determines the dimensionality-reduced time-frequency domain distribution of the incident signal and the reflected signal based on the first dimensionality reduction matrix and the correspondence between the time integral variable and the frequency, specifically including:

[0083] First, the server generates a third dimensionality reduction matrix based on the function composed of the time integral variable and the frequency in the Wigner distribution.

[0084] Among them, the number of rows of the third dimensionality reduction matrix is the same as the number of columns of the first dimensionality reduction matrix. The rows of the third dimensionality reduction matrix are time integral variables, and the columns of the third dimensionality reduction matrix are frequencies.

[0085] The third dimensionality reduction matrix corresponds to the e -jωτ part, and represents e -jωτ as a two-dimensional matrix with rows as frequencies and columns as time integral variables.

[0086] Then, the server determines the dimensionality-reduced time-frequency domain distribution of the incident signal and the reflected signal based on matrix multiplication of the first dimensionality reduction matrix and the third dimensionality reduction matrix.

[0087] In an embodiment of the present application, as Figure 5 shown, before the above dimensionality reduction operation, the matrix corresponding to the time-frequency domain distribution e -jωτ part is shown as S1 and is a three-dimensional matrix. After the dimensionality reduction processing in the above embodiment, e -jωτ can be converted into the matrix multiplication form of the S2 part, thereby reducing the three-dimensional data to two-dimensional data.

[0088] Through the above solution, the time-frequency domain distribution can be dimensionally reduced and simplified, reducing the computational amount of data and the data operation time.

[0089] S105. The server generates a second dimensionality reduction matrix according to the dimensionality-reduced time-frequency domain distribution and a preset rule, and performs dimensionality reduction processing on the cross-correlation function according to the second dimensionality reduction matrix to obtain a dimensionality-reduced cross-correlation function, and generates a defect location curve corresponding to the dimensionality-reduced cross-correlation function.

[0090] In the embodiment of the present application, the server generates a second dimensionality reduction matrix according to the dimensionality reduction time-frequency domain distribution and a preset rule, and performs dimensionality reduction processing on the cross-correlation function according to the second dimensionality reduction matrix to obtain a dimensionality reduction cross-correlation function, and generates a defect location curve corresponding to the dimensionality reduction cross-correlation function, specifically including:

[0091] First, the server determines the time-frequency domain distribution values of the frequencies corresponding to each time of the dimensionality reduction time-frequency domain distribution.

[0092] Next, the server determines the maximum value among the time-frequency domain distribution values, and generates a second dimensionality reduction matrix by arranging the maximum values in the time-frequency domain distribution values in chronological order.

[0093] Among them, the second dimensionality reduction matrix includes an incident signal time-frequency dimensionality reduction sub-matrix and a reflected signal time-frequency dimensionality reduction sub-matrix.

[0094] Specifically, the server can determine the time-frequency domain distribution value according to the following formula:

[0095] w r (t) = Max[W r (t, ω1), W r (t, ω2), W r (t, ω3),...]

[0096] w s (t) = Max[W s (t, ω1), W s (t, ω2), W s (t, ω3),...]

[0097] Among them, w r (t) is the maximum value in the time-frequency domain distribution value of the reflected signal, w s (t) is the maximum value in the time-frequency domain distribution value of the incident signal, ω1, ω2... represent the mth frequency corresponding to the time, where m is a natural number. According to the maximum values corresponding to each time, an incident signal time-frequency dimensionality reduction sub-matrix and a reflected signal time-frequency dimensionality reduction sub-matrix can be obtained.

[0098] Then, the server normalizes the dimensionality reduction time-frequency domain distribution according to the central time of the signal duration of the incident signal.

[0099] The normalization process is as follows:

[0100]

[0101] Among them, T s is the central time of the signal duration of the incident signal.

[0102] Finally, the server determines the reduced-dimension cross-correlation function of the incident signal and the reflected signal based on the normalized reduced-dimension time-frequency domain distribution and the second reduced-dimension matrix, so as to obtain the defect location curve corresponding to the cable to be measured according to the reduced-dimension cross-correlation function.

[0103] In the embodiment of the present application, the reduced-dimension cross-correlation function is expressed as:

[0104]

[0105] where C′(t) is the reduced-dimension cross-correlation function.

[0106] S106. The server determines the defect location of the cable to be measured according to the defect location curve.

[0107] In the embodiment of the present application, the server determines the defect location of the cable to be measured according to the defect location curve, which specifically includes:

[0108] The server determines the signal propagation speed of the cable to be measured according to the material of the cable to be measured.

[0109] The signal propagation speed of the cable to be measured can be determined from the database, or the server obtains it through the network.

[0110] The server converts the cross-correlation function into a corresponding distance correlation function based on the signal propagation speed of the cable to be measured and the corresponding relationship between the signal propagation time and the distance from the cable head.

[0111] where the distance correlation function is used to represent the relationship between the distance and the signal correlation degree, and the signal correlation degree is used to represent the correlation between the reflected signal and the incident signal.

[0112] The corresponding relationship between the signal transmission speed, the signal propagation time, and the distance from the cable head can be expressed as:

[0113]

[0114] where x is the distance from the cable head, v is the signal propagation speed, and t is the signal propagation time.

[0115] After obtaining the above-mentioned reduced-dimension cross-correlation function, the independent variable time t can be converted into a distance correlation function with the distance from the cable as the independent variable.

[0116] The server determines the defect location curve corresponding to the cable to be measured according to the distance correlation function, so as to determine the distance between the defect of the cable to be measured and the cable head according to the defect location curve.

[0117] In the embodiment of the present application, the image of the distance correlation function is as Figure 6 shown, Figure 6 in which there is a break at 9 m from the head end,Figure 6 It also includes a defect location curve of the cross-correlation function without performing the above-mentioned dimensionality reduction operation.

[0118] In the embodiment of the present application, the server determines the defect location of the cable to be tested according to the defect location curve, specifically including:

[0119] First, the server determines the maximum value in the defect location curve.

[0120] For example Figure 6 , the defect location curve has four maximum values.

[0121] Next, the server takes the position of the maximum value at the starting point in the defect location curve as the head end of the cable, and takes the position of the maximum value at the ending point in the defect location curve as the tail end of the cable.

[0122] That is, the maximum value at 0 meters is the head end of the cable, and the maximum value at 15 meters is the tail end of the cable.

[0123] Then, the server matches the curve waveforms within the preset interval at each maximum value with the curve waveform at the head end of the cable and / or the curve waveform at the tail end of the cable.

[0124] In the embodiment of the present application, there is a maximum value at 5 meters, and this maximum value is within the preset interval. For example, the preset interval is determined with the maximum value as the center, and the left and right neighborhoods at 1 meter are the preset interval, that is, the preset interval for the maximum value at 5 meters is [4, 6]. The server determines the curve waveform within the preset interval and matches this curve waveform with the curve waveforms in the preset intervals at the head end and the tail end of the cable.

[0125] Finally, the server determines the position corresponding to the maximum value with a matching degree greater than the preset value in the matching result as the defect location.

[0126] In the embodiment of the present application, the matching can calculate the similarity of the curve waveforms, such as calculating the Fréchet distance, to obtain the matching degree, or the matching degree can also be calculated by other methods. The present application does not make specific limitations in this regard. Subsequently, the server determines whether the matching degree is greater than the preset value, and takes the position corresponding to the maximum value with a matching degree greater than the preset value as the defect location. For example, if the server calculates that the matching degree of the curve waveform at the preset interval corresponding to 9 meters with the curve waveforms at the head end and the tail end of the cable is greater than the preset value, then 9 meters is taken as the defect location. In the embodiment of the present application, the preset value can be set during actual use, and the present application does not make specific limitations in this regard.

[0127] Through the above solution, the operation process of determining cable defects by the time-frequency domain reflection method can be reduced in dimension and simplified. The calculation of the three-dimensional cross-correlation function with a large amount of calculation is reduced in dimension and simplified to a two-dimensional cross-correlation function, greatly reducing the storage amount of data and the time consumed by the operation. Moreover, based on the two-dimensional cross-correlation function after dimension reduction, the accurate positioning of cable defects can be realized. The present application also simplifies the calculation formula of the time-frequency domain distribution in dimension, thereby further reducing the operation time and the data processing complexity. The present application can improve the positioning speed of cable defects and the computing efficiency of the processor, and can improve the user experience of cable defect locators.

[0128] Figure 7 FIG. is a schematic structural diagram of a cable defect positioning device based on data dimension reduction provided by an embodiment of the present application, as Figure 7 shown. The device includes:

[0129] At least one processor; and a memory communicatively connected to the at least one processor. Wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to:

[0130] Obtain the incident signal and the reflected signal from the signal receiver. The signal receiver is used to receive the incident signal generated by the signal transmitter at one end of the T-shaped interface of the cable to be measured, and to receive the reflected signal after the incident signal passes through the cable to be measured. Based on the preset number of samples, determine the number of discrete time and the discrete time interval of the incident signal and the reflected signal. Based on the number of discrete time and the discrete time interval, determine the variable time interval of the time integration variable of the time-frequency domain distribution, so as to determine the time integration variable of each discrete time corresponding to the number of samples through the variable time interval, and generate a first dimension reduction matrix. Based on the correspondence between the time integration variable and the frequency and the first dimension reduction matrix, determine the dimension-reduced time-frequency domain distribution of the incident signal and the reflected signal. According to the dimension-reduced time-frequency domain distribution and the preset rule, generate a second dimension reduction matrix, so as to perform dimension reduction processing on the cross-correlation function according to the second dimension reduction matrix, obtain the dimension-reduced cross-correlation function, and generate a defect positioning curve corresponding to the dimension-reduced cross-correlation function. According to the defect positioning curve, determine the defect position of the cable to be measured.

[0131] In the embodiment of the present application, the at least one processor can specifically:

[0132] Determine the variable time interval of the time integration variable according to the discrete time interval and the coefficient of the time integration variable in the time-frequency domain distribution. According to the discrete time interval, determine the sampling number of discrete times, and determine the time integration variable corresponding to each discrete time as the selected variable. According to the ratio of each selected variable to the variable time interval, determine the value shift number of the time integration variable corresponding to each discrete time. The value shift number is used to determine the corresponding incident signal frequency or reflected signal frequency under the time integration variable corresponding to the discrete time. According to the value shift number, determine the first dimensionality reduction matrix. The columns of the first dimensionality reduction matrix are discrete times, and the rows of the first dimensionality reduction matrix are time integration variables.

[0133] Each embodiment in this application is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the device embodiments, since they are basically similar to the method embodiments, the description is relatively simple. For the relevant parts, reference can be made to the partial description of the method embodiments.

[0134] The device provided in the embodiments of this application corresponds one-to-one with the method. Therefore, the device also has beneficial technical effects similar to those of its corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the device will not be elaborated here.

[0135] It should also be noted that the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, commodity or device. Without further limitation, the element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, commodity or device including the element.

[0136] The above are only the embodiments of this application and are not used to limit this application. For those skilled in the art, various changes and modifications can be made to this application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall be included within the scope of the claims of this application.

Claims

1. A cable defect location method based on data dimensionality reduction, characterized in that The method includes: Obtaining an incident signal and a reflected signal from a signal receiver; wherein, the signal receiver is used to receive the incident signal generated by a signal transmitter at one end of a T-shaped interface of a cable to be measured, and is also used to receive the reflected signal after the incident signal passes through the cable to be measured; Determining the discrete time quantity and the discrete time interval of the incident signal and the reflected signal based on a preset sampling quantity; Determining the variable time interval of the time integration variable of the time-frequency domain distribution based on the discrete time quantity and the discrete time interval, so as to determine the time integration variable of each discrete time corresponding to the sampling quantity through the variable time interval, and generating a first dimensionality reduction matrix; Determining the dimensionality-reduced time-frequency domain distribution of the incident signal and the reflected signal based on the correspondence between the time integration variable and the frequency and the first dimensionality reduction matrix; Generating a second dimensionality reduction matrix according to the dimensionality-reduced time-frequency domain distribution and a preset rule, so as to perform dimensionality reduction processing on the cross-correlation function according to the second dimensionality reduction matrix, obtaining a dimensionality-reduced cross-correlation function, and generating a defect location curve corresponding to the dimensionality-reduced cross-correlation function; Determining the defect location of the cable to be measured according to the defect location curve; Wherein, generating a second dimensionality reduction matrix according to the dimensionality-reduced time-frequency domain distribution and a preset rule, so as to perform dimensionality reduction processing on the cross-correlation function according to the second dimensionality reduction matrix, obtaining a dimensionality-reduced cross-correlation function, and generating a defect location curve corresponding to the dimensionality-reduced cross-correlation function, specifically including: Determining the time-frequency domain distribution values of the frequencies corresponding to each time of the dimensionality-reduced time-frequency domain distribution; Determining the maximum value among each of the time-frequency domain distribution values, and generating the second dimensionality reduction matrix in chronological order with the maximum values in the time-frequency domain distribution values; wherein, the second dimensionality reduction matrix includes an incident signal time-frequency dimensionality reduction sub-matrix and a reflected signal time-frequency dimensionality reduction sub-matrix; Normalizing the dimensionality-reduced time-frequency domain distribution according to the central time of the signal duration of the incident signal; Determining the dimensionality-reduced cross-correlation function of the incident signal and the reflected signal based on the normalized dimensionality-reduced time-frequency domain distribution and the second dimensionality reduction matrix, so as to obtain the defect location curve corresponding to the cable to be measured according to the dimensionality-reduced cross-correlation function; Wherein, determining the dimensionality-reduced time-frequency domain distribution of the incident signal and the reflected signal based on the correspondence between the time integration variable and the frequency and the first dimensionality reduction matrix, specifically including: Generating a third dimensionality reduction matrix based on a function composed of the time integration variable and the frequency in the Wigner distribution; wherein, the number of rows of the third dimensionality reduction matrix is the same as the number of columns of the first dimensionality reduction matrix; the rows of the third dimensionality reduction matrix are the time integration variables, and the columns of the third dimensionality reduction matrix are the frequencies; Determining the dimensionality-reduced time-frequency domain distribution of the incident signal and the reflected signal based on matrix multiplication processing of the first dimensionality reduction matrix and the third dimensionality reduction matrix.

2. The method according to claim 1, characterized in that, Based on the discrete time quantity and the discrete time interval, determine the variable time interval of the time integration variable of the time-frequency domain distribution, so as to determine the time integration variable of each discrete time corresponding to the sampling quantity through the variable time interval, and generate a first dimensionality reduction matrix, specifically including: Determine the variable time interval of the time integration variable according to the discrete time interval and the coefficient of the time integration variable in the time-frequency domain distribution; Determine the discrete time of the sampling quantity according to the discrete time interval, and determine the time integration variable corresponding to each discrete time as the selected variable; Determine the value shift number of the time integration variable corresponding to each discrete time according to the ratio of each selected variable to the variable time interval; wherein, the value shift number is used to determine the corresponding incident signal frequency or reflection signal frequency under the time integration variable corresponding to the discrete time; Determine the first dimensionality reduction matrix according to the value shift number; wherein, the columns of the first dimensionality reduction matrix are the discrete times, and the rows of the first dimensionality reduction matrix are the time integration variables.

3. The method according to claim 2, wherein Before determining the value shift number of the time integration variable corresponding to each discrete time according to the ratio of each selected variable to the variable time interval, the method further includes: Determine the maximum value of the time integration variable according to the product of a preset constant, the discrete time quantity and the discrete time interval; Determine the value range of the time integration variable according to the maximum value of the time integration variable; Determine whether the value of the time integration variable corresponding to each discrete time matches the value range; If not, generate constant adjustment information and send the constant adjustment information to the user terminal; Based on the operation of the user on the user terminal, determine the feedback information of the constant adjustment information, so as to adjust the preset constant according to the feedback information.

4. The method according to claim 1, wherein Determine the defect position of the cable under test according to the defect location curve, specifically including: Determine the signal propagation speed of the cable under test according to the material of the cable under test; Based on the signal propagation speed of the cable under test and the corresponding relationship between the signal propagation time and the distance from the cable head, convert the cross-correlation function into a corresponding distance correlation function; wherein, the distance correlation function is used to represent the relationship between the distance and the signal correlation degree, and the signal correlation degree is used to represent the correlation between the reflection signal and the incident signal; Determine the defect location curve corresponding to the cable under test according to the distance correlation function, so as to determine the distance between the defect of the cable under test and the cable head according to the defect location curve.

5. The method according to claim 1, wherein Determine the defect position of the cable under test according to the defect location curve, specifically including: Determine the maximum value in the defect location curve; Take the position of the maximum value at the starting point of the defect location curve as the cable head, and take the position of the maximum value at the ending point of the defect location curve as the cable end; Match the curve waveforms within the preset intervals at each of the maxima with the curve waveform at the head end of the cable and / or the curve waveform at the tail end of the cable; Determine the corresponding positions of the maxima with a matching degree greater than the preset value in the matching result as the defect positions.

6. The method according to claim 1, wherein Before obtaining the incident signal and the reflected signal from the signal receiver, the method further includes: Obtain the test information sent by the control terminal; wherein, the test information at least includes: cable open circuit data, cable position data; In the case of determining that the cable to be tested in the test information is open circuited, generate incident signal generation information and send the incident signal generation information to the signal transmitter; wherein, the incident signal generation information includes Gaussian envelope data and linear chirp data.

7. A cable defect location device based on data dimensionality reduction, characterized in that, The device adopts a cable defect location method based on data dimensionality reduction as described in any one of the above claims 1-6, and the device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: Obtain the incident signal and the reflected signal from the signal receiver; wherein, the signal receiver is used to receive the incident signal generated by the signal transmitter at one end of the T-joint of the cable to be tested, and to receive the reflected signal after the incident signal passes through the cable to be tested; Based on the preset number of samples, determine the number of discrete time and the discrete time interval of the incident signal and the reflected signal; Based on the number of discrete time and the discrete time interval, determine the variable time interval of the time integral variable of the time-frequency domain distribution, and through the variable time interval, determine the time integral variable of each discrete time corresponding to the number of samples, and generate a first dimensionality reduction matrix; Based on the correspondence between the time integral variable and the frequency and the first dimensionality reduction matrix, determine the reduced time-frequency domain distribution of the incident signal and the reflected signal; Generate a second dimensionality reduction matrix according to the reduced time-frequency domain distribution and the preset rules, and perform dimensionality reduction processing on the cross-correlation function according to the second dimensionality reduction matrix to obtain a reduced cross-correlation function, and generate a defect location curve corresponding to the reduced cross-correlation function; Determine the defect position of the cable to be tested according to the defect location curve.

8. The device according to claim 7, wherein The at least one processor is specifically capable of: Determine the variable time interval of the time integral variable according to the discrete time interval and the coefficient of the time integral variable in the time-frequency domain distribution; Determine the number of discrete time of the number of samples according to the discrete time interval, and determine the time integral variable corresponding to each discrete time as the selected variable; Determine the value shift bits of the time integral variable corresponding to each discrete time according to the ratio of each selected variable to the variable time interval; wherein, the value shift bits are used to determine the corresponding incident signal frequency or reflected signal frequency at the time integral variable corresponding to the discrete time; Determine the first dimensionality reduction matrix according to the number of shifted bits of the value; wherein, the columns of the first dimensionality reduction matrix are the discrete time, and the rows of the first dimensionality reduction matrix are the time integration variables.

Images