Power transmission line fault monitoring method, system, medium, equipment and product

Abstract

The invention discloses a power transmission line fault monitoring method and system, a medium, equipment and a product in the technical field of power transmission line online detection, and aims to solve the optimization problem of monitoring and identifying power transmission line faults. The method comprises the following steps: carrying out data preprocessing on an acquired transient voltage signal and a current signal of a power transmission line; in a high-frequency signal duration window in a single transient process, if the preprocessed current signal is greater than a preset amplitude for the first time, performing spectrum distribution and energy spectrum distribution analysis on the preprocessed voltage signal to obtain a frequency band energy ratio; and if the frequency band energy proportion exceeds a preset characteristic threshold value, considering that the power transmission line has a fault, and inputting the preprocessed current signal into the trained clustering model for fault identification to obtain a power transmission line fault type. According to the invention, real-time monitoring and rapid identification of the power transmission line fault are realized, emergency measures can be taken at the initial stage of the fault, and large-scale power failure accidents are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of on-line detection of transmission lines, and particularly relates to a method, system, medium, device and product for monitoring transmission line faults. Background Art

[0002] As an important part of the power system, the safety and stability of the operation of transmission lines directly affect the reliability of the power system. With the continuous expansion of the power grid scale and the improvement of complexity, transmission lines are faced with various potential fault threats. Therefore, timely and accurately monitoring and identifying transmission line faults has become an important research direction for ensuring the safe operation of the power system.

[0003] With the advancement of the construction of the new power system, the fault monitoring technology of transmission lines has gradually developed from traditional manual inspection to automation and intelligence. Obtaining a large amount of voltage and current data of the power system safely, accurately and without interference can be used to ensure the safety protection and stable operation of the system. However, due to the existing monitoring and identification technologies still having problems such as low fault identification accuracy, low efficiency and inability to detect in real time, it is difficult to achieve efficient, accurate and real-time monitoring of transmission lines. Summary of the Invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a method, system, medium, device and product for monitoring transmission line faults, which realizes real-time monitoring and rapid identification of transmission line faults, is convenient for engineering applications, and has practical significance and good application prospects.

[0005] To solve the above technical problems, the present invention is implemented by adopting the following technical solutions:

[0006] The present invention provides a method for monitoring transmission line faults, including:

[0007] Performing data preprocessing on the collected transient voltage signal and transient current signal of the transmission line to obtain the preprocessed voltage signal and current signal;

[0008] Within the high-frequency signal duration window of a single transient process, if the preprocessed current signal is greater than or equal to a preset amplitude for the first time, perform the following steps:

[0009] Performing spectrum distribution and energy spectrum distribution analysis on the preprocessed voltage signal to obtain the frequency band energy ratio of the preprocessed voltage signal;

[0010] If the frequency band energy ratio of the preprocessed voltage signal does not exceed a preset characteristic threshold, it is considered that the transmission line has not failed, and continue to observe the transient voltage signal and current signal of the transmission line;

[0011] If the frequency band energy ratio of the preprocessed voltage signal exceeds a preset characteristic threshold, it is considered that a transmission line fault occurs, and the preprocessed current signal is input into a trained clustering model for fault identification to obtain the fault type of the transmission line.

[0012] Optionally, the data preprocessing includes filtering, amplification and inversion calculation.

[0013] Optionally, using a Fourier transform algorithm to perform spectrum distribution and energy spectrum distribution analysis on the preprocessed voltage signal to obtain a frequency band energy ratio of the preprocessed voltage signal includes:

[0014] Extract the maximum peak point of the frequency spectrum in the preprocessed voltage signal And the maximum peak point of the spectrum in the preprocessed voltage signal The frequency point corresponding to the A times of the corresponding amplitude ;

[0015] The frequency point As the upper limit of the characteristic frequency band, according to the upper limit of the characteristic frequency band and the lower limit of the preset characteristic frequency band , get the characteristic frequency band ;

[0016] Calculate the frequency band energy of the preset frequency bands respectively With the characteristic frequency band Band energy , according to the frequency band energy of the preset frequency band With the characteristic frequency band Band energy , calculate the preset frequency band and characteristic frequency band The frequency band energy ratio ;

[0017] The frequency band energy of the preset frequency band The calculation formula is as follows:

[0018]

[0019] in, Indicates the frequency amplitude of the sampling point in the preset frequency band, Indicates the frequency difference between adjacent sampling points in the preset frequency band. Indicates the frequency difference between the upper and lower limits of the spectrum distribution in the preset frequency band. Indicates the starting frequency of the preset frequency band. Indicates the end frequency of the preset frequency band;

[0020] The characteristic frequency band Band energy The calculation formula is as follows:

[0021]

[0022] Wherein, represents the frequency amplitude of the sampling point in the characteristic frequency band in the characteristic frequency band represents the frequency difference between adjacent sampling points in the characteristic frequency band in the characteristic frequency band represents the frequency difference between the upper and lower limits of the spectrum distribution in the characteristic frequency band;

[0023] The frequency band energy ratio The calculation formula is as follows:

[0024] .

[0025] Optionally, the clustering model adopts the K-means algorithm, and the training process of the clustering model includes:

[0026] Obtain an abnormal data set containing n line fault category samples;

[0027] According to the abnormal data set, randomly select a point as the clustering center point in each line fault category to obtain n center points;

[0028] Calculate the distance d from each line fault category sample to the n center points respectively;

[0029] According to the distance d from each line fault category sample to the n center points, assign each line fault category sample to the cluster to which the nearest center point belongs to obtain n clusters;

[0030] Calculate the average value of the distances from all line fault category samples in each cluster to the center point of the cluster, and update the center point of each cluster according to the average value;

[0031] Repeat the steps of calculating the distance d from each line fault category sample to the n center points respectively to updating the center point of each cluster according to the average value until the within-cluster sum of squared errors reaches the minimum value to obtain the trained clustering model;

[0032] The within-cluster sum of squared errors is calculated by the following formula:

[0033]

[0034] Wherein, represents the serial number of the line fault category sample, represents the cluster serial number, represents the number of line fault category samples, Indicates the number of clusters, Indicates the th sample of the line fault category, Indicates the center point of the cluster, Indicates the line fault category sample and the center point The square of the Euclidean distance d therebetween.

[0035] Optionally, the fault identification is implemented by the following formula:

[0036]

[0037] wherein, Indicates the threshold value, 1 indicates belonging to this type of fault, and 0 indicates not belonging to this type of fault, Indicates the preprocessed current signal, Indicates the center point of the cluster, Indicates the preprocessed current signal and the center point The square of the Euclidean distance d therebetween.

[0038] In a second aspect, the present invention provides a transmission line fault monitoring system, including:

[0039] A signal acquisition module (14) for: acquiring the transient voltage signal and transient current signal of the transmission line;

[0040] A signal processing and transmission module (15) for: performing data preprocessing on the acquired transient voltage signal and transient current signal of the transmission line to obtain the preprocessed voltage signal and current signal;

[0041] An analysis and judgment module (16) for: performing fault judgment on the transmission line, including:

[0042] Within the high-frequency signal duration window of a single transient process, if the preprocessed current signal is greater than or equal to the preset amplitude for the first time, perform the following steps:

[0043] Performing spectrum distribution and energy spectrum distribution analysis on the preprocessed voltage signal to obtain the band energy ratio of the preprocessed voltage signal;

[0044] If the band energy ratio of the preprocessed voltage signal does not exceed the preset characteristic threshold, it is considered that the transmission line has not failed, and continue to observe the transient voltage signal and current signal of the transmission line;

[0045] If the band energy ratio of the preprocessed voltage signal exceeds a preset feature threshold, it is considered that a fault has occurred in the transmission line, and the preprocessed current signal is input into the trained clustering model for fault identification to obtain the fault type of the transmission line.

[0046] Optionally, the signal processing and transmission module (15) includes a GPS time synchronization unit (8), a signal reading unit (9), and a signal conversion unit (10); the GPS time synchronization unit (8) is used to control the signal reading unit (9) to read the filtered and amplified signal at a fixed frequency and transmit it to the signal conversion unit (10) for inversion calculation to obtain the preprocessed voltage signal and current signal.

[0047] Optionally, the Fourier transform algorithm is used to analyze the spectrum distribution and energy spectrum distribution of the preprocessed voltage signal to obtain the band energy ratio of the preprocessed voltage signal, including:

[0048] Extract the maximum peak point of the spectrum in the preprocessed voltage signal and the maximum peak point of the spectrum in the preprocessed voltage signal The frequency point corresponding to A times the amplitude corresponding to ;

[0049] Use the frequency point as the upper limit of the characteristic frequency band, and according to the upper limit of the characteristic frequency band and the preset lower limit of the characteristic frequency band , obtain the characteristic frequency band ;

[0050] Calculate the band energy of the preset frequency band and the band energy of the characteristic frequency band respectively. According to the band energy of the preset frequency band and the band energy of the characteristic frequency band , calculate the band energy ratio of the preset frequency band and the characteristic frequency band ; The calculation formula of the band energy of the preset frequency band is as follows:

[0051] where represents the frequency amplitude of the sampling point in the preset frequency band,

[0052]

[0053] represents the frequency difference between adjacent sampling points in the preset frequency band, represents the frequency difference between the upper and lower limits of the spectrum distribution in the preset frequency band, represents the frequency difference between the upper and lower limits of the spectrum distribution in the preset frequency band represents the starting frequency of the preset frequency band, represents the ending frequency of the preset frequency band;

[0054] The characteristic frequency band band energy The calculation formula is as follows:

[0055]

[0056] where, represents the frequency amplitude of the sampling point in the characteristic frequency band in the characteristic frequency band, represents the characteristic frequency band the frequency difference between adjacent sampling points in the characteristic frequency band, represents the frequency difference between the upper and lower limits of the spectrum distribution in the characteristic frequency band;

[0057] The band energy ratio The calculation formula is as follows:

[0058] .

[0059] Optionally, the clustering model adopts the K-means algorithm, and the training process of the clustering model includes:

[0060] Obtain an abnormal data set containing n samples of line fault categories;

[0061] According to the abnormal data set, randomly select a point as the clustering center point in each line fault category to obtain n center points;

[0062] Calculate the distance d from each line fault category sample to the n center points respectively;

[0063] According to the distance d from each line fault category sample to the n center points, assign each line fault category sample to the cluster to which the nearest center point belongs to obtain n clusters;

[0064] Calculate the average value of the distances from all line fault category samples in each cluster to the center point of the cluster, and update the center point of each cluster according to the average value;

[0065] Repeat the above steps of calculating the distance d from each line fault category sample to the n center points respectively to updating the center point of each cluster according to the average value until the sum of squared errors within the cluster reaches the minimum value to obtain the trained clustering model;

[0066] The sum of squared errors within the cluster is calculated by the following formula:

[0067]

[0068] Among them, represents the serial number of the line fault category sample, represents the cluster serial number, represents the number of line fault category samples, represents the number of clusters, represents the th line fault category sample, represents the center point of the th cluster, represents the square of the Euclidean distance d between the line fault category sample and the center point

[0069] Optionally, the fault identification is implemented through the following formula:

[0070]

[0071] Among them, represents the threshold value, 1 indicates belonging to this type of fault, and 0 indicates not belonging to this type of fault, represents the preprocessed current signal, represents the center point of the th cluster, represents the square of the Euclidean distance d between the preprocessed current signal and the center point

[0072] In a third aspect, the present invention provides a computer-readable storage medium, on which computer instructions are stored. When the computer instructions are executed by a processor, the steps of any one of the transmission line fault monitoring methods in the first aspect are implemented.

[0073] In a fourth aspect, the present invention provides a computer device, including:

[0074] A memory for storing computer instructions;

[0075] A processor for executing the computer instructions to implement the steps of any one of the transmission line fault monitoring methods in the first aspect.

[0076] In a fifth aspect, the present invention provides a computer program product, including computer instructions, characterized in that when the computer instructions are executed by a processor, the steps of any one of the transmission line fault monitoring methods in the first aspect are implemented.

[0077] Compared with the prior art, the beneficial effects achieved by the present invention:

[0078] 1. The power transmission line fault monitoring method provided by the present invention uses the spectral power of the voltage signal as the starting criterion for fault judgment. After the starting criterion is met, the clustering method is used to cluster the current signal to achieve real-time monitoring and rapid identification of power transmission line faults, enabling emergency measures to be taken in the initial stage of a fault, preventing the expansion of the fault, and reducing the occurrence of large-scale power outages;

[0079] 2. The power transmission line fault monitoring method provided by the present invention can group similar fault patterns according to the monitored multi-dimensional data by using K-means clustering, thereby helping to quickly identify which parts are abnormal, narrowing the scope of fault investigation, and improving the accuracy of fault location; and the clustering monitoring method can automatically classify the data, reduce manual intervention, and improve the monitoring efficiency; identifying different types of fault patterns and further analyzing them helps to formulate more targeted maintenance and repair strategies.

[0080] 3. The power transmission line fault monitoring system provided by the present invention can extract valuable information from a large amount of monitoring data by setting a signal acquisition module, a signal processing and transmission module, and an analysis and judgment module, realizing real-time monitoring and rapid identification of power transmission line faults; with the accumulation of data, the system can continuously optimize the clustering model, thereby providing accurate fault prediction and location capabilities for decision-making, promoting the intelligent development of fault warning and optimized operation and maintenance, and being convenient for engineering applications, having practical significance and good application prospects;

[0081] 4. The computer-readable storage medium, computer device, and computer program product provided by the present invention can execute the steps of the power transmission line fault monitoring method provided by the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 is a flowchart of the power transmission line fault monitoring method according to an embodiment of the present invention;

[0083] Figure 2 is a schematic diagram of spectrum monitoring according to an embodiment of the present invention;

[0084] Figure 3 is a structural diagram of the power transmission line fault monitoring system according to an embodiment of the present invention;

[0085] Figure 4 is a schematic diagram of the layout of the power transmission line detection device according to an embodiment of the present invention;

[0086] In the figure: 1 - Transmission line to be monitored; 2 - Transient monitoring device; 3 - In-station monitoring point of power system; 4 - Voltage sensor; 5 - Current sensor; 6 - Filtering unit; 7 - Amplifying unit; 8 - GPS time synchronization unit; 9 - Signal reading unit; 10 - Signal conversion unit; 11 - Data transmission unit; 12 - Upper industrial control computer; 13 - Software monitoring unit; 14 - Signal acquisition module; 15 - Signal processing and transmission module; 16 - Analysis and judgment module. Specific embodiments

[0087] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific features in the embodiments of the present application are detailed descriptions of the technical solution of the present application, rather than limitations on the technical solution of the present application. Without conflict, the technical features in the embodiments of the present application and the embodiments can be combined with each other.

[0088] It should be noted that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after.

[0089] Embodiment 1:

[0090] An embodiment of the present invention discloses a method for monitoring faults in a transmission line. Referring to Figure 1 as shown, the specific steps are as follows:

[0091] The present invention provides a method for monitoring faults in a transmission line, including:

[0092] S1. Perform data preprocessing on the collected transient voltage signal and transient current signal of the transmission line to obtain the preprocessed voltage signal and current signal;

[0093] S2. In the high-frequency signal duration window of a single transient process, if the preprocessed current signal is greater than or equal to the preset amplitude for the first time, perform the following steps:

[0094] S3. Analyze the frequency spectrum distribution and energy spectrum distribution of the preprocessed voltage signal to obtain the frequency band energy ratio of the preprocessed voltage signal;

[0095] S4. If the frequency band energy ratio of the preprocessed voltage signal does not exceed the preset characteristic threshold, it is considered that no fault has occurred in the transmission line, and continue to observe the transient voltage signal and current signal of the transmission line;

[0096] S5. If the band energy ratio of the preprocessed voltage signal exceeds the preset feature threshold, it is considered that a fault has occurred in the transmission line. The preprocessed current signal is input into the trained clustering model for fault identification to obtain the fault type of the transmission line.

[0097] Specifically,

[0098] In step S1, the data preprocessing includes filtering, amplification, and inversion calculation; the transient voltage signal and transient current signal of the transmission line collected through filtering and amplification operations are denoised and enhanced, making subsequent fault detection more accurate; what is measured by the sensor is only the differential value of the transient voltage signal and transient current signal, so inversion calculation is also required to restore it to the real voltage and current.

[0099] In step S2, the duration of the high-frequency signal in the entire transient process is generally 3 - 4 ms. Therefore, in this embodiment, the transient monitoring clock duration is taken as 3 ms, and the current signal is used as the trigger mechanism. Within the duration window of the high-frequency signal in a single transient process, the moment when the instantaneous value of the current is first greater than or equal to 1.5 times the normal current amplitude is taken as the fault moment.

[0100] In step S3, the Fourier transform algorithm is used to analyze the spectral distribution and energy spectral distribution of the preprocessed voltage signal; in this embodiment, the transient voltage signal within 3 ms after the fault moment is found as the object of spectral distribution and energy spectral distribution analysis; Refer to Figure 2 As shown in the spectral monitoring schematic diagram, the spectral distribution and energy spectral distribution of the preprocessed voltage signal are analyzed to obtain the band energy ratio of the preprocessed voltage signal, including:

[0101] Extract the maximum peak point of the spectrum in the preprocessed voltage signal And the maximum peak point of the spectrum in the preprocessed voltage signal The frequency point corresponding to A times the amplitude corresponding to it ;

[0102] Take the frequency point As the upper limit of the characteristic frequency band, according to the upper limit of the characteristic frequency band And the preset lower limit of the characteristic frequency band , obtain the characteristic frequency band ; In this embodiment Take 50 Hz;

[0103] Calculate the band energy of the preset frequency band And the band energy of the characteristic frequency band respectively, according to the band energy of the preset frequency band And the band energy of the characteristic frequency band And the characteristic frequency band Band energy , the band energy ratios of the preset frequency band and the characteristic frequency band are calculated ; ;

[0104] The calculation formula for the band energy of the preset frequency band is as follows: ;

[0105]

[0106] wherein, represents the frequency amplitude of the sampling points in the preset frequency band, represents the frequency difference between adjacent sampling points in the preset frequency band, represents the frequency difference between the upper and lower limits of the frequency spectrum distribution in the preset frequency band, represents the starting frequency of the preset frequency band, represents the ending frequency of the preset frequency band;

[0107] The calculation formula for the band energy of the characteristic frequency band is as follows: ;

[0108]

[0109] wherein, represents the characteristic frequency band the frequency amplitude of the sampling points in, represents the characteristic frequency band the frequency difference between adjacent sampling points in; represents the frequency difference between the upper and lower limits of the frequency spectrum distribution in the characteristic frequency band;

[0110] The calculation formula for the band energy ratio is as follows:

[0111] .

[0112] In step S4, when the band energy ratio of the preset frequency band and the characteristic frequency band in the preprocessed voltage signal is less than the preset ratio , it is determined that a fault has occurred in the transmission line; In this embodiment, according to previous measurement results, an optional implementation manner is provided, and the preset frequency band range is set to 50 - 1 kHz, and the preset ratio is 0.8.

[0113] In step S5, the clustering model uses the K-means algorithm to divide the collected m abnormal data sets into n categories, where n is the number of line fault types; according to the m abnormal data sets, a point is selected as the initial clustering center point in each of the n line fault categories to obtain n center points; the distance d from each line fault category sample in the abnormal data set to the n center points is calculated respectively, and it is assigned to the cluster with the smallest distance, and then the average value of the distances from all line fault category samples in each cluster to the center point of the cluster is calculated, and each cluster center is updated according to the average value until the sum of squared errors within the cluster reaches the minimum value; this algorithm determines the optimal centroids of the n clusters by minimizing the sum of squared errors within the cluster formula. The specific training process of the clustering model includes:

[0114] Obtain an abnormal data set containing n line fault category samples;

[0115] According to the abnormal data set, a point is randomly selected as the clustering center point in each line fault category to obtain n center points;

[0116] Calculate the distance d from each line fault category sample to the n center points respectively;

[0117] According to the distance d from each line fault category sample to the n center points, each line fault category sample is assigned to the cluster to which the nearest center point belongs to obtain n clusters;

[0118] Calculate the average value of the distances from all line fault category samples in each cluster to the center point of the cluster, and update the center point of each cluster according to the average value;

[0119] Repeat the steps of calculating the distance d from each line fault category sample to the n center points to updating the center point of each cluster according to the average value until the sum of squared errors within the cluster Reaches the minimum value to obtain a trained clustering model;

[0120] The sum of squared errors within the cluster Is calculated by the following formula:

[0121]

[0122] Where, Represents the serial number of the line fault category sample, Represents the cluster serial number, Represents the number of line fault category samples, Represents the number of clusters, Represents the th line fault category sample, Represents the center point of the th cluster, Represents the line fault category sample The square of the Euclidean distance d from the center point between them.

[0123] For the transmission line fault monitoring method provided in this embodiment, during real-time monitoring, first calculate the spectral power of the preprocessed voltage signal, and use it as the starting criterion. If the spectral energy distribution of the preprocessed voltage signal exceeds the threshold , it is considered that a line fault has occurred; subsequently, calculate the distances between the preprocessed current signal and the best centroids of each fault . If the minimum distance is less than the corresponding threshold, the fault type can be determined; the fault identification is realized through the following formula:

[0124]

[0125] wherein, represents the threshold, 1 indicates belonging to this type of fault, 0 indicates not belonging to this type of fault, represents the preprocessed current signal, represents the center point of the cluster, represents the preprocessed current signal and the center point between them. The square of the Euclidean distance d.

[0126] Embodiment 2:

[0127] This embodiment of the present invention discloses a transmission line fault monitoring system. Referring to Figure 3 as shown, it includes:

[0128] A signal acquisition module (14) for: acquiring the transient voltage signal and transient current signal of the transmission line;

[0129] A signal processing and transmission module (15) for: performing data preprocessing on the acquired transient voltage signal and transient current signal of the transmission line to obtain the preprocessed voltage signal and current signal;

[0130] An analysis and judgment module (16) for: performing fault judgment on the transmission line, including:

[0131] Within the high-frequency signal duration window of a single transient process, if the preprocessed current signal is greater than or equal to the preset amplitude for the first time, perform the following steps:

[0132] Perform spectral distribution and energy spectrum distribution analysis on the preprocessed voltage signal to obtain the frequency band energy ratio of the preprocessed voltage signal;

[0133] If the band energy ratio of the preprocessed voltage signal does not exceed the preset characteristic threshold, it is considered that the transmission line has not failed, and the transient voltage signal and current signal of the transmission line are continuously observed;

[0134] If the band energy ratio of the preprocessed voltage signal exceeds the preset characteristic threshold, it is considered that the transmission line has failed, and the preprocessed current signal is input into the trained clustering model for fault identification to obtain the fault type of the transmission line.

[0135] Specifically,

[0136] Refer to Figure 3 As shown, the signal acquisition module 14 includes a voltage sensor 4 and a current sensor 5, which are used to acquire the transient voltage signal and current signal of the transmission line; the bandwidth of the voltage sensor 4 and the current sensor 5 is 50 Hz - 1 MHz, and its measurement principle includes but is not limited to optical measurement and electromagnetic induction.

[0137] The data preprocessing includes filtering, amplification, and inversion calculation; the transient voltage signal and transient current signal of the transmission line collected through filtering and amplification operations are denoised and enhanced, making subsequent fault detection more accurate; what is measured by the sensor is only the differential value of the transient voltage signal and transient current signal, so inversion calculation is also required to restore it to the real voltage and current.

[0138] The signal processing and transmission module 15 includes a filtering unit 6, an amplification unit 7, a GPS time synchronization unit 8, a signal reading unit 9, a signal conversion unit 10, and a data transmission unit 11; the filtering unit 6 uses a band-pass filter with a bandwidth of 50 - 1 MHz; the transient voltage signal and current signal of the transmission line are denoised and enhanced through the filtering unit 6 and the amplification unit 7, and then the signal reading unit 9 reads the signal at a fixed frequency and transmits it to the signal conversion unit 10 for inversion calculation to obtain a digital signal; finally, the digital signal is transmitted to the analysis and judgment module 16 through the data transmission unit 11; the GPS time synchronization unit 8 is used to control the frequency of the signal reading unit 9 to read the signal, which can achieve the synchronization of the transient voltage and current measurement of the transmission line, and monitor the magnitudes of the transient voltage signal and current signal of the transmission line to be detected in real time. The error can be controlled at the ns level, which can effectively reduce the error of reading the transient voltage signal and current signal.

[0139] The duration of the high-frequency signal in the entire transient process of the transmission line is generally 3 - 4 ms. Therefore, in this embodiment, the transient monitoring clock duration is set to 3 ms, and the current signal is used as the trigger mechanism. Within the duration window of the high-frequency signal in a single transient process, the moment when the instantaneous value of the current is first greater than or equal to 1.5 times the conventional current amplitude is used as the fault moment.

[0140] The analysis and judgment module 16 includes a software monitoring unit 13, which is used on the one hand to analyze the frequency-domain characteristics of the preprocessed voltage signal, extract characteristic parameters, and compare the extracted characteristic parameters with preset characteristic thresholds respectively to realize the fault monitoring of the transmission line; on the other hand, it clusters the preprocessed current signal by the k-means clustering method to realize the identification of the transmission line fault.

[0141] In this embodiment, an upper industrial control computer 12 can also be added to the analysis and judgment module 16. The data transmission unit 11 sends the processed signal to the substation intranet main station system by using the local virtual private dial-up network, uploads the data to the upper industrial control computer 12, and the station staff obtains the measured signal through the software monitoring unit 13 installed in the upper industrial control computer 12 by using digital-to-analog conversion to realize the human-computer information interaction.

[0142] Specifically, the software monitoring unit 13 analyzes the spectrum of the preprocessed voltage signal through the Fourier transform (FFT) algorithm, extracts the characteristic parameters therein, and compares them with the preset characteristic thresholds. If the characteristic parameters of the signal meet the requirements, for this detection point, it can be preliminarily considered that the transmission line has a fault; in this embodiment, the transient voltage signal within 3 ms after the fault moment is found as the analysis object of the spectrum distribution and energy spectrum distribution; reference Figure 2 As shown in the spectrum monitoring schematic diagram, the spectrum distribution and energy spectrum distribution of the preprocessed voltage signal are analyzed to obtain the frequency band energy ratio of the preprocessed voltage signal, including:

[0143] Extract the maximum peak point of the spectrum in the preprocessed voltage signal and the maximum peak point of the spectrum in the preprocessed voltage signal the frequency point corresponding to A times the amplitude corresponding to ;

[0144] Take the frequency point as the upper limit of the characteristic frequency band, and according to the upper limit of the characteristic frequency band and the preset lower limit of the characteristic frequency band , obtain the characteristic frequency band ; in this embodiment take 50 Hz;

[0145] Calculate the frequency band energy of the preset frequency band and the frequency band energy of the characteristic frequency band respectively, and according to the frequency band energy of the preset frequency band and the frequency band energy of the characteristic frequency band and the frequency band energy of the characteristic frequency band , calculate the frequency band energy ratio of the preset frequency band and the characteristic frequency band , calculate the frequency band energy ratio of the preset frequency band and the characteristic frequency band ​ ;

[0146] The band energy of the preset frequency band is calculated as follows:

[0147]

[0148] wherein, represents the frequency amplitude of the sampling points in the preset frequency band, represents the frequency difference between adjacent sampling points in the preset frequency band, represents the frequency difference between the upper and lower limits of the spectrum distribution in the preset frequency band, represents the starting frequency of the preset frequency band, represents the ending frequency of the preset frequency band;

[0149] The band energy of the characteristic frequency band is calculated as follows: is calculated as follows:

[0150]

[0151] wherein, represents the characteristic frequency band the frequency amplitude of the sampling points in represents the characteristic frequency band the frequency difference between adjacent sampling points in represents the frequency difference between the upper and lower limits of the spectrum distribution in the characteristic frequency band;

[0152] The band energy ratio is calculated as follows:

[0153] .

[0154] Specifically, the software monitoring unit 13 analyzes the spectrum distribution and energy spectrum distribution of the preprocessed voltage signal through the Fourier transform (FFT) algorithm, extracts the maximum peak point of the spectrum in the preprocessed voltage signal and the maximum peak point of the spectrum in the preprocessed voltage signal the frequency point corresponding to 0.0025 (A = 0.005) times the amplitude corresponding to it, and determines the upper limit of the characteristic frequency band The lower limit of the characteristic frequency band is a preset parameter, where . After obtaining the signal power spectrum of the characteristic frequency band 50Hz - 10kHz, calculate the band energy of the preset frequency band 50 - 10kHz and the band energy of the characteristic frequency band ; The band energy of the preset frequency band and the characteristic frequency band ​ Band energy , the band energy of the preset frequency band and the characteristic frequency band is calculated Ratio of band energy .

[0155] Among them, the formula for calculating the band energy of the preset frequency band of 50 - 10 kHz is as follows: The formula is as follows:

[0156]

[0157] Among them, represents the frequency amplitude of the sampling points in the preset frequency band, represents the frequency difference between adjacent sampling points in the preset frequency band, represents the frequency difference between the upper and lower limits of the frequency spectrum distribution in the preset frequency band.

[0158] The formula for calculating the ratio of band energy is as follows:

[0159]

[0160] Assume that Figure 2 the maximum peak points extracted from the two voltage signals and the characteristic frequency band are outside 50 - 300 kHz, and the proportion of the band energy of 50 - 50 kHz in the total energy is less than 0.8, then it is determined that the circuit breaker arc fault occurs, otherwise there is no arc fault.

[0161] The clustering model uses the K - means algorithm to divide the m abnormal data sets collected into n categories, where n is the number of line fault types; according to the m abnormal data sets, one point is selected as the initial clustering center point in each of the n line fault categories to obtain n center points; the distance d from each line fault category sample in the abnormal data set to the n center points is calculated respectively, and it is assigned to the cluster with the smallest distance, then the average value of the distances from all line fault category samples in each cluster to the center point of the cluster is calculated, and each cluster center is updated according to the average value until the sum of squared errors within the cluster reaches the minimum value; this algorithm determines the optimal centroids of the n clusters by minimizing the within - cluster sum - of - squares error formula. The specific training process of the clustering model includes:

[0162] Obtain an abnormal data set containing n line fault category samples;

[0163] According to the abnormal data set, randomly select one point as the clustering center point in each line fault category to obtain n center points;

[0164] Calculate the distance d from each line fault category sample to the n center points respectively;

[0165] According to the distances d from each line fault category sample to the n center points, each line fault category sample is assigned to the cluster to which the nearest center point belongs, obtaining n clusters;

[0166] Calculate the average value of the distances from all line fault category samples in each cluster to the center point of that cluster, and update the center point of each cluster according to the average value;

[0167] Repeat the steps of calculating the distances d from each line fault category sample to the n center points respectively to updating the center point of each cluster according to the average value until the within-cluster sum of squared errors reaches the minimum value, obtaining the trained clustering model;

[0168] The within-cluster sum of squared errors is calculated by the following formula:

[0169]

[0170] where represents the serial number of the line fault category sample, represents the cluster serial number, represents the number of line fault category samples, represents the number of clusters, represents the th line fault category sample, represents the center point of the th cluster, represents the line fault category sample and the center point the square of the Euclidean distance d between them.

[0171] During real-time monitoring, first calculate the spectral power of the preprocessed voltage signal and use it as the startup criterion. If the spectral energy distribution of the preprocessed voltage signal exceeds the threshold , it is considered that a line fault has occurred; then calculate the distances between the preprocessed current signal and the respective fault optimal centroids . If the minimum distance is less than the corresponding threshold, the fault type can be determined; the fault identification is achieved through the following formula:

[0172]

[0173] where represents the threshold, 1 represents belonging to this type of fault, 0 represents not belonging to this type of fault, represents the preprocessed current signal, represents the th cluster center point, represents the preprocessed current signal and the center point The square of the Euclidean distance d between them.

[0174] Reference Figure 4 As shown, the signal acquisition module and the signal processing and transmission module proposed in this embodiment can be encapsulated in the transient monitoring device 2, fixed on the circuit under test 1 by means of a buckle, and wirelessly transmitted to the monitoring point 3 in the power system power station.

[0175] In summary, for the transmission line fault monitoring system proposed in this embodiment, when a fault occurs in the transmission line, such as single-phase grounding, interphase short circuit or line break, etc., the electromagnetic wave propagation characteristics in the system will change significantly, thus generating high-frequency components in the voltage and current waveforms; installing the system on the transmission line with a high fault incidence rate, using the current signal as the trigger mechanism, analyzing the spectral characteristics of the transient voltage signal of the transmission line, extracting characteristic parameters, obtaining the transmission line fault monitoring result, and identifying the line fault by means of clustering.

[0176] Embodiment 3:

[0177] This embodiment provides a computer-readable storage medium, on which computer instructions are stored. When the computer instructions are executed by a processor, the steps of the transmission line fault monitoring method described in any one of Embodiment 1 are implemented.

[0178] Embodiment 4:

[0179] This embodiment provides a computer device, including:

[0180] A memory for storing computer instructions;

[0181] A processor for executing the computer instructions to implement the steps of the transmission line fault monitoring method described in any one of the first aspects.

[0182] Embodiment 5:

[0183] This embodiment provides a computer program product, including computer instructions, characterized in that when the computer instructions are executed by a processor, the steps of the transmission line fault monitoring method described in any one of Embodiment 1 are implemented.

[0184] In summary of the above embodiments, the transmission line fault monitoring method, system, medium, device and product provided by the present invention realize real-time monitoring and rapid identification of transmission line faults by specifically researching transmission line fault monitoring and identification technologies, can take emergency measures at the initial stage of the fault, prevent the expansion of the fault, and reduce the occurrence of large-scale power outage accidents.

[0185] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0186] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0187] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implement the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0188] These computer program instructions can also be loaded onto a computer or other programmable data processing devices, so that a series of operation steps are executed on the computer or other programmable devices to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable devices provide steps for implementing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0189] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the spirit and scope of the present invention as protected by the claims. All of these are within the protection scope of the present invention.

Claims

1. A transmission line fault monitoring method, characterized in that, Including: Performing data preprocessing on the collected transient voltage signal and transient current signal of the transmission line to obtain the preprocessed voltage signal and current signal; Within the high-frequency signal duration window of a single transient process, if the preprocessed current signal is greater than or equal to the preset amplitude for the first time, perform the following steps: Analyzing the spectral distribution and energy spectral distribution of the preprocessed voltage signal to obtain the band energy ratio of the preprocessed voltage signal; If the band energy ratio of the preprocessed voltage signal does not exceed the preset characteristic threshold, it is considered that the transmission line has not failed, and continue to observe the transient voltage signal and current signal of the transmission line; If the band energy ratio of the preprocessed voltage signal exceeds the preset characteristic threshold, it is considered that the transmission line has failed, and input the preprocessed current signal into the trained clustering model for fault identification to obtain the fault type of the transmission line.

2. The power transmission line fault monitoring method according to claim 1, characterized in that The data preprocessing includes filtering, amplification, and inversion calculation.

3. The transmission line fault monitoring method according to claim 1, wherein, Using the Fourier transform algorithm to analyze the spectral distribution and energy spectral distribution of the preprocessed voltage signal to obtain the band energy ratio of the preprocessed voltage signal, including: Extract the maximum peak point of the spectrum in the preprocessed voltage signal and the maximum peak point of the spectrum in the preprocessed voltage signal the frequency point corresponding to A times the amplitude corresponding to it ; Take the said frequency point as the upper limit of the characteristic frequency band, and according to the upper limit of the characteristic frequency band and the preset lower limit of the characteristic frequency band , obtain the characteristic frequency band ; Calculate the band energy of the preset frequency band respectively and the characteristic frequency band of the band energy , according to the band energy of the preset frequency band and the characteristic frequency band of the band energy , calculate the band energy ratio of the preset frequency band and the characteristic frequency band ; ; The band energy of the preset frequency band The calculation formula is as follows: Among them, represents the frequency amplitude of the sampling points in the preset frequency band, represents the frequency difference between adjacent sampling points in the preset frequency band, represents the frequency difference between the upper and lower limits of the spectrum distribution in the preset frequency band, represents the starting frequency of the preset frequency band, represents the termination frequency of the preset frequency band; The characteristic frequency band of the band energy is calculated as follows: Among them, represents the frequency amplitude of the sampling points in the characteristic frequency band, represents the characteristic frequency band the frequency difference between adjacent sampling points in, represents the frequency difference between the upper and lower limits of the spectrum distribution in the characteristic frequency band; The energy ratio of the frequency band is calculated as follows: 。 4. The transmission line fault monitoring method according to claim 1, wherein The clustering model adopts the K-means algorithm, and the training process of the clustering model includes: Obtaining an abnormal data set containing n line fault category samples; According to the abnormal data set, randomly select a point as the clustering center point in each line fault category to obtain n center points; Calculating the distance d from each line fault category sample to the n center points respectively; According to the distance d from each line fault category sample to the n center points, assign each line fault category sample to the cluster to which the nearest center point belongs to obtain n clusters; Calculating the average value of the distances from all line fault category samples in each cluster to the center point of the cluster, and updating the center point of each cluster according to the average value; Repeat the steps of calculating the distances \(d\) from each sample of each line fault category to the \(n\) center points respectively to updating the center points of each cluster according to the average value until the sum of squared errors within the cluster reaches the minimum value, and a trained clustering model is obtained; The sum of squared errors within the cluster is calculated by the following formula: Among them, represents the serial number of the line fault category sample, represents the cluster serial number, represents the number of line fault category samples, represents the number of clusters, represents the th line fault category sample, represents the center point of the th cluster, represents the square of the Euclidean distance d between the line fault category sample and the center point .

5. The transmission line fault monitoring method according to claim 1, characterized in that The fault identification is realized through the following formula: Among them, represents the threshold value, 1 represents belonging to this type of fault, and 0 represents not belonging to this type of fault. represents the current signal after preprocessing. represents the center point of the cluster. represents the square of the Euclidean distance d between the current signal after preprocessing and the center point.

6. A transmission line fault monitoring system, characterized in that, Including: A signal acquisition module (14) for collecting the transient voltage signal and transient current signal of the transmission line; A signal processing and transmission module (15) for performing data preprocessing on the collected transient voltage signal and transient current signal of the transmission line to obtain the preprocessed voltage signal and current signal; An analysis and judgment module (16) for performing fault judgment on the transmission line, including: Within the high-frequency signal duration window of a single transient process, if the preprocessed current signal is greater than or equal to the preset amplitude for the first time, perform the following steps: Analyzing the spectral distribution and energy spectral distribution of the preprocessed voltage signal to obtain the band energy ratio of the preprocessed voltage signal; If the band energy ratio of the preprocessed voltage signal does not exceed the preset characteristic threshold, it is considered that the transmission line has not failed, and continue to observe the transient voltage signal and current signal of the transmission line; If the band energy ratio of the preprocessed voltage signal exceeds the preset characteristic threshold, it is considered that the transmission line has failed, and input the preprocessed current signal into the trained clustering model for fault identification to obtain the fault type of the transmission line.

7. The transmission line fault monitoring system according to claim 6, characterized in that The signal processing and transmission module (15) includes a GPS time synchronization unit (8), a signal reading unit (9), and a signal conversion unit (10); the GPS time synchronization unit (8) is configured to control the signal reading unit (9) to read the filtered and amplified signal at a fixed frequency and transmit it to the signal conversion unit (10) for inversion calculation to obtain the preprocessed voltage signal and current signal.

8. The transmission line fault monitoring system according to claim 6, wherein The Fourier transform algorithm is used to analyze the spectrum distribution and energy spectrum distribution of the preprocessed voltage signal to obtain the frequency band energy ratio of the preprocessed voltage signal, including: Extract the maximum peak point of the spectrum in the preprocessed voltage signal and the maximum peak point of the spectrum in the preprocessed voltage signal The frequency point corresponding to A times the amplitude corresponding to it ; Take the frequency point as the upper limit of the characteristic frequency band. According to the upper limit of the characteristic frequency band and the preset lower limit of the characteristic frequency band , obtain the characteristic frequency band ; Calculate the band energy of the preset frequency band respectively and the characteristic frequency band of the band energy , according to the band energy of the preset frequency band and the characteristic frequency band of the band energy , calculate the band energy ratio of the preset frequency band and the characteristic frequency band ; ; The band energy of the preset frequency band The calculation formula is as follows: Among them, represents the frequency amplitude of the sampling points in the preset frequency band, represents the frequency difference between adjacent sampling points in the preset frequency band, represents the frequency difference between the upper and lower limits of the spectrum distribution in the preset frequency band, represents the starting frequency of the preset frequency band, represents the ending frequency of the preset frequency band; The characteristic frequency band The energy of the frequency band The calculation formula is as follows: Among them, represents the frequency amplitude of the sampling point in the characteristic frequency band, represents the characteristic frequency band the frequency difference between adjacent sampling points in, represents the frequency difference between the upper and lower limits of the spectrum distribution in the characteristic frequency band; The proportion of band energy The calculation formula is as follows: 。 9. The transmission line fault monitoring system according to claim 6, wherein, The clustering model adopts the K-means algorithm, and the training process of the clustering model includes: Obtaining an abnormal data set containing n line fault category samples; According to the abnormal data set, randomly selecting a point as the clustering center point in each line fault category to obtain n center points; Calculating the distance d from each line fault category sample to the n center points respectively; According to the distance d from each line fault category sample to the n center points, allocating each line fault category sample to the cluster to which the nearest center point belongs to obtain n clusters; Calculating the average value of the distances from all line fault category samples in each cluster to the center point of the cluster, and updating the center point of each cluster according to the average value; Repeat the steps of calculating the distances \(d\) from each sample of each line fault category to the \(n\) center points respectively to updating the center points of each cluster according to the average value until the sum of squared errors within the cluster reaches the minimum value, and a trained clustering model is obtained; The sum of squared errors within the cluster is calculated by the following formula: Among them, represents the serial number of the line fault category sample, represents the cluster serial number, represents the number of line fault category samples, represents the number of clusters, represents the th line fault category sample, represents the center point of the th cluster, represents the square of the Euclidean distance d between the line fault category sample and the center point .

10. The transmission line fault monitoring system according to claim 6, wherein The fault identification is achieved through the following formula: Among them, represents the threshold value, 1 represents belonging to this type of fault, and 0 represents not belonging to this type of fault. represents the current signal after preprocessing. represents the center point of the cluster. represents the square of the Euclidean distance d between the current signal after preprocessing and the center point.

11. A computer-readable storage medium having computer instructions stored thereon, characterized in that, When the computer instruction is executed by the processor, the steps of the transmission line fault monitoring method described in any one of claims 1-5 are implemented.

12. A computer device, characterized in that, Including: A memory for storing computer instructions; A processor for executing the computer instructions to implement the steps of the transmission line fault monitoring method described in any one of claims 1-5.

13. A computer program product comprising computer instructions, characterized in that, When the computer instruction is executed by the processor, the steps of the transmission line fault monitoring method described in any one of claims 1-5 are implemented.

Images