A method and device for locating a fault in an underground cable based on vibration attenuation

Abstract

The application relates to the technical field of cable fault diagnosis, and discloses an underground cable fault positioning method and device based on vibration attenuation. The method uses a plurality of sensors to collect vibration data of the cable to form a first time sequence; an empirical mode decomposition method is used to decompose the first time sequence into a plurality of intrinsic mode function components; according to the intrinsic mode function components, a plurality of characteristic values of the first time sequence are calculated; a preset vibration positioning model is used to determine the fault position according to the position data of each sensor and the characteristic values; and the characteristic value corresponding to the fault position is matched in a preset database to obtain the fault type corresponding to the fault position. The application improves the efficiency and accuracy of underground cable fault detection, guarantees the operation stability of the power system, and significantly reduces the economic loss and safety risk caused by the delay of the fault.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cable fault diagnosis, in particular to a method and device for locating underground cable fault based on vibration attenuation. BACKGROUND

[0002] With the rapid development of China's economy and the promotion of urbanization construction, overhead cable lines are gradually replaced by underground power cables, which brings significant benefits to the beauty and safety of the city. However, the challenges faced by underground cables have also increased, especially due to the acceleration of urban construction and reconstruction, frequent earthwork construction significantly increases the risk of cable damage. Statistical data shows that one of the main reasons for cable accidents is external damage, among which physical damage caused by road excavation engineering is the most serious.

[0003] Currently, underground cables are mostly laid in a direct-buried manner. Although this method is simple and low in cost, it is extremely susceptible to external construction. At present, the maintenance and detection of cable systems often use the method of manual inspection, which is time-consuming and labor-intensive, and it is difficult to achieve real-time monitoring. The method of using robots for inspection has obvious limitations for underground direct-buried cables, tunnel laying or cable trench laying, and cannot be effectively detected. The current monitoring scheme for cables mostly uses image processing technology to detect the distance of hidden objects by laying monitoring equipment, but in actual application, it is easily affected by environmental light and may have monitoring blind spots, which is difficult to be widely promoted. The method of using optical cable to detect the vibration information along the monitoring area has an initial installation cost higher than other types of monitoring systems, and is sensitive to temperature and humidity, which affects its performance and accuracy. SUMMARY

[0004] The present application provides a method and device for locating underground cable fault based on vibration attenuation, which improves the efficiency and accuracy of underground cable fault detection, ensures the operation stability of the power system, and significantly reduces the economic loss and safety risk caused by delay due to fault.

[0005] In order to solve the above technical problems, the present application provides a method for locating underground cable fault based on vibration attenuation, comprising:

[0006] Collecting vibration data of the cable by using a plurality of sensors to form a first time series; wherein the interval distance of each sensor is equal;

[0007] Using empirical mode decomposition method to decompose the first time series into a plurality of intrinsic mode function components;

[0008] According to each intrinsic mode function component, a plurality of characteristic values of the first time series are calculated; wherein the characteristic values and the sensors are one-to-one corresponding;

[0009] determine the fault position according to the position data and the characteristic value of each sensor by using a preset vibration positioning model;

[0010] match the characteristic value corresponding to the fault position in a preset database to obtain a fault type corresponding to the fault position.

[0011] Further, the first time sequence is decomposed into a plurality of intrinsic mode function components by using the empirical mode decomposition method, specifically:

[0012] determine the first time sequence as an original signal;

[0013] decompose the original signal into an approximate trend item and a residual component in a loop until the residual component is a monotonic function;

[0014] determine the approximate trend item obtained each time as an intrinsic mode function component.

[0015] Further, the original signal is decomposed into an approximate trend item and a residual component, specifically:

[0016] obtain a plurality of maximum points and a plurality of minimum points in the original signal;

[0017] interpolate the plurality of maximum points to obtain an upper envelope, and interpolate the plurality of minimum points to obtain a lower envelope;

[0018] calculate the mean value of the upper envelope and the lower envelope to obtain an approximate trend item:

[0019] subtract the approximate trend item from the original signal to obtain a residual component;

[0020] determine whether the residual component is a monotonic function;

[0021] if the residual component is not a monotonic function, determine the residual component as an original signal.

[0022] Further, the plurality of characteristic values of the first time sequence are calculated according to each intrinsic mode function component, further comprising:

[0023] a plurality of energy entropies of the first time sequence are calculated according to each intrinsic mode function component, specifically:

[0024]

[0025] wherein, H i is the energy entropy of the i th sensor; p ijis the ratio of the energy contained in the j-th intrinsic mode function component of the i-th sensor to the total energy; m is the number of intrinsic mode function components;

[0026] Alternatively, based on each of the intrinsic mode function components, several waveform factors of the first time series are calculated, specifically as follows:

[0027]

[0028] In the formula, WF i Let a be the waveform factor of the i-th sensor; i (t) represents the vibration signal of the i-th sensor; N represents the signal a. i The total number of sampling points for (t);

[0029] Alternatively, based on each of the intrinsic mode function components, several kurtosis values ​​of the first time series can be calculated, specifically as follows:

[0030]

[0031] In the formula, K i Let a be the kurtosis of the i-th sensor; i (t) represents the vibration signal of the i-th sensor; N represents the signal a. i The total number of sampling points for (t); is the mean of the vibration signal; s is the standard deviation of the vibration signal;

[0032] Alternatively, based on the intrinsic mode function components, several centroid frequencies of the first time series can be calculated, specifically as follows:

[0033]

[0034] In the formula, CF i Let f be the centroid frequency of the i-th sensor; K be the total number of frequency components; f k E represents the k-th frequency component. ik Let be the energy of the frequency component of the i-th sensor in the vibration signal.

[0035] Furthermore, the step of using a preset vibration positioning model to determine the fault location based on the position data and feature values ​​of each sensor specifically involves:

[0036] Using the position data and feature values ​​of each sensor, a preset vibration positioning model is fitted to obtain the relationship function between the feature values ​​and the position.

[0037] Based on the relationship function, the theoretical characteristic value corresponding to each sensor is calculated according to the position data of each sensor;

[0038] Obtain candidate position data corresponding to the two sensors with the largest theoretical eigenvalues;

[0039] The intermediate value of the two candidate location data is determined as the fault location.

[0040] Furthermore, the vibration positioning model is specifically as follows:

[0041] f(x) = a·e -bx +c

[0042] In the formula, f(x) is the characteristic value of the sensor; x is the position data of the sensor; a, b and c are the parameters to be fitted.

[0043] Further, the step of matching the feature values ​​corresponding to the fault location in a preset database to obtain the fault type corresponding to the fault location specifically involves:

[0044] In the first time series, the feature value corresponding to the fault location is obtained;

[0045] The feature value corresponding to the fault location is compared with several data thresholds in a preset database to obtain several similarity scores.

[0046] Obtain the first fault type corresponding to the threshold of data with the highest similarity;

[0047] The first fault type is determined as the fault type corresponding to the fault location.

[0048] Furthermore, the step of using several sensors to collect vibration data of the cable and forming a first time series specifically involves:

[0049] Obtain the starting point position and preset interval distance of the initial sensor installation;

[0050] Calculate the position data of several sensors based on the installation starting point position of the initial sensor and the preset interval distance;

[0051] Based on the position data of each sensor, each sensor is positioned along the longitudinal direction of the cable;

[0052] Vibration data of the cable are collected using the sensors described above to form a first time series.

[0053] Furthermore, the step of using each of the sensors to collect vibration data of the cable and forming a first time series specifically involves:

[0054] After collecting vibration data of the cable using the sensors, the vibration data is denoised using an adaptive filtering algorithm.

[0055] The noise-reduced vibration data is then standardized to form the first time series.

[0056] This invention provides a method for locating underground cable faults based on vibration attenuation. It utilizes several sensors to collect cable vibration data, forming a first time series. An empirical mode decomposition (EMD) method is employed to decompose the first time series into several intrinsic mode function (EMF) components. Based on each EMF component, several characteristic values ​​of the first time series are calculated. Using a preset vibration location model, the fault location is determined based on the sensor position data and characteristic values. The characteristic values ​​corresponding to the fault location are matched against a preset database to derive the fault type. This invention can effectively issue early warnings when buried underground cables are subjected to external impacts, ensuring the stable operation of the power system and the normal functioning of the social economy. It improves the efficiency and accuracy of underground cable fault detection and significantly reduces economic losses and safety risks caused by fault delays, demonstrating high practical value and promising prospects for widespread application.

[0057] Accordingly, the present invention provides a fault location device for underground cables based on vibration attenuation, comprising: a data acquisition module, a decomposition module, a calculation module, a location module, and a matching module;

[0058] The acquisition module is used to collect vibration data of the cable using several sensors to form a first time series; wherein the interval between each sensor is equal;

[0059] The decomposition module is used to decompose the first time series into several intrinsic mode function components using the empirical mode decomposition method;

[0060] The calculation module is used to calculate several feature values ​​of the first time series based on each of the intrinsic mode function components; wherein, the feature values ​​correspond one-to-one with the sensors;

[0061] The positioning module is used to determine the fault location based on the position data and feature values ​​of each sensor using a preset vibration positioning model.

[0062] The matching module is used to match the feature values ​​corresponding to the fault location in a preset database to obtain the fault type corresponding to the fault location.

[0063] This invention provides a vibration attenuation-based underground cable fault location device. Based on the organic integration of modules, it utilizes several sensors to collect cable vibration data, forming a first time series. Employing empirical mode decomposition (EMD), the first time series is decomposed into several intrinsic mode function (EMF) components. Based on each EMF component, several characteristic values ​​of the first time series are calculated. Using a preset vibration location model, the fault location is determined based on the position data and characteristic values ​​of each sensor. The characteristic values ​​corresponding to the fault location are matched against a preset database to derive the fault type. This invention can effectively issue early warnings when buried underground cables are subjected to external impacts, ensuring the stable operation of the power system and the normal operation of the social economy. It improves the efficiency and accuracy of underground cable fault detection and significantly reduces economic losses and safety risks caused by fault delays, possessing high practical value and promising prospects for widespread application. Attached Figure Description

[0064] Figure 1 A flowchart illustrating an embodiment of the vibration attenuation-based underground cable fault location method provided by the present invention.

[0065] Figure 2 A schematic diagram of the structure of one embodiment of the vibration data acquisition system provided by the present invention;

[0066] Figure 3 A schematic diagram of a time series provided by the present invention;

[0067] Figure 4 A schematic diagram of an embodiment of the underground cable fault location system based on vibration attenuation provided by the present invention;

[0068] Figure 5 A schematic flowchart illustrating another embodiment of the vibration attenuation-based underground cable fault location method provided by the present invention.

[0069] Figure 6 This is a schematic diagram of one embodiment of the underground cable fault location device based on vibration attenuation provided by the present invention. Detailed Implementation

[0070] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0071] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0072] The following detailed description of some embodiments of the present invention is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0073] Example 1

[0074] See Figure 1 This is a flowchart illustrating an embodiment of the underground cable fault location method based on vibration attenuation provided by the present invention. The method includes steps 101 to 105, each of which is detailed below:

[0075] Step 101: Collect vibration data of the cable using several sensors to form a first time series; wherein the interval between each sensor is equal.

[0076] Furthermore, in the first embodiment of the present invention, vibration data of the cable is collected using several sensors to form a first time series, specifically as follows:

[0077] Obtain the starting point position and preset interval distance of the initial sensor installation;

[0078] Calculate the position data of several sensors based on the installation starting point position of the initial sensor and the preset interval distance;

[0079] Based on the position data of each sensor, each sensor is positioned along the longitudinal direction of the cable;

[0080] Vibration data of the cable are collected using the sensors described above to form a first time series.

[0081] In the first embodiment of the present invention, see Figure 2 This is a structural schematic diagram of an embodiment of the vibration data acquisition system provided by the present invention. By configuring high-sensitivity and high-diamagnetic accelerometer sensors, the acceleration or vibration generated by the cable under external force can be measured. Therefore, a group of multiple accelerometer sensors are evenly arranged on the underground direct-buried cable, with spacing along the longitudinal direction of the cable. Each accelerometer sensor is attached to the cable surface to ensure efficient transmission of vibration signals and minimal signal attenuation. The installation position of each accelerometer sensor can be determined by the following equal-spacing formula:

[0082] L i =L0+(i-1)·d

[0083] In the formula, Li L0 represents the installation position of the i-th accelerometer sensor; L0 represents the starting position of the first sensor installation; and d represents the distance between adjacent sensors.

[0084] In the first embodiment of the present invention, vibration data is collected using sensors arranged longitudinally along the cable, and a time series can be generated. See also Figure 3 This invention provides a schematic diagram of a time series, where the data on the coordinate system is acquired by data acquisition sensors. The horizontal axis represents the distance data along the cable from each data acquisition sensor, and the vertical axis represents the signal amplitude acquired by the corresponding data acquisition sensor, thus constructing a first time series. This invention utilizes sensors to acquire cable vibration data in real time, which helps to identify cable loss or faults in advance, provides a scientific basis for maintenance work, and ensures the stability and continuity of power grid operation.

[0085] Furthermore, in the first embodiment of the present invention, vibration data of the cable is collected using each of the sensors to form a first time series, specifically as follows:

[0086] After collecting vibration data of the cable using the sensors, the vibration data is denoised using an adaptive filtering algorithm.

[0087] The noise-reduced vibration data is then standardized to form the first time series.

[0088] In the first embodiment of the present invention, after acquiring vibration data at various locations on the cable using sensors, an adaptive filtering algorithm is used to denoise the acquired data, removing noise interference from the vibration signal. The data is then standardized to ensure consistency and comparability during the analysis. The processed vibration data is recorded as a time series a. i (t) is used to accurately extract vibration characteristics, providing an accurate and stable data foundation for subsequent fault analysis.

[0089] Step 102: Using the empirical mode decomposition method, the first time series is decomposed into several intrinsic mode function components.

[0090] Furthermore, in the first embodiment of the present invention, the empirical mode decomposition method is used to decompose the first time series into several intrinsic mode function components, specifically as follows:

[0091] The first time series is determined as the original signal;

[0092] The original signal is decomposed into an approximate trend term and a residual component in a loop until the residual component is a monotonic function.

[0093] The approximate trend term obtained from each decomposition is determined as the intrinsic mode function component.

[0094] Furthermore, in the first embodiment of the present invention, the original signal is decomposed into an approximate trend term and a residual component, specifically as follows:

[0095] Obtain several maxima and several minima from the original signal;

[0096] Interpolation is performed on several of the maxima to obtain the upper envelope, and interpolation is performed on several of the minima to obtain the lower envelope;

[0097] Calculate the mean of the upper envelope and the lower envelope to obtain an approximate trend term:

[0098] Subtract the approximate trend term from the original signal to obtain the residual component;

[0099] Determine whether the residual component is a monotonic function;

[0100] If the residual component is not a monotonic function, then the residual component is determined as the original signal.

[0101] In the first embodiment of the present invention, the empirical mode decomposition method is used to decompose the time series into multiple intrinsic mode function components (IMFs), denoted as e ij (t), where j is the corresponding IMF index. The time series decomposition process is as follows: first determine signal a i The upper envelope e is obtained by interpolating the maximum and minimum points in (t) using the maximum points. max (t), the lower envelope e is obtained by interpolation at the minimum point. min (t). After obtaining the upper and lower envelopes, the mean m(t) of the upper and lower envelopes is calculated and used as an approximate trend term of the signal: By acquiring signal a i Details are extracted by finding the difference between (t) and the approximate trend term: d(t) = a(t) - m(t), where d(t) is the residual component. After each decomposition, it is determined whether the residual component contains no intrinsic mode function components or whether it is a monotonic function r(t). If the residual component is not a monotonic function and can still be decomposed, the iterative decomposition continues in the above manner. When the residual component is a monotonic function, the time series a can be reorganized. i The mathematical form of (t): In the formula, m i r(t) represents the intrinsic mode function component after each decomposition of the i-th sensor, and r(t) represents the residual component. Based on the above decomposition method, the time series can be converted into intrinsic mode function components. Each intrinsic mode function component corresponds to a specific frequency component in the vibration signal, thereby revealing the intrinsic structure and properties of the signal in detail, reflecting the vibration characteristics of the cable, and providing key information for identifying changes in the cable's condition.

[0102] Step 103: Calculate several feature values ​​of the first time series based on each of the intrinsic mode function components; wherein, the feature values ​​correspond one-to-one with the sensors.

[0103] Furthermore, in the first embodiment of the present invention, the step of calculating several feature values ​​of the first time series based on each of the intrinsic mode function components further includes:

[0104] Based on the intrinsic mode function components, several energy entropies of the first time series are calculated, specifically as follows:

[0105]

[0106] In the formula, H i p is the energy entropy of the i-th sensor; ij is the ratio of the energy contained in the j-th intrinsic mode function component of the i-th sensor to the total energy; m is the number of intrinsic mode function components;

[0107] Alternatively, based on each of the intrinsic mode function components, several waveform factors of the first time series are calculated, specifically as follows:

[0108]

[0109] In the formula, WF i Let a be the waveform factor of the i-th sensor; i (t) represents the vibration signal of the i-th sensor; N represents the signal a. i The total number of sampling points for (t);

[0110] Alternatively, based on each of the intrinsic mode function components, several kurtosis values ​​of the first time series can be calculated, specifically as follows:

[0111]

[0112] In the formula, K i Let a be the kurtosis of the i-th sensor; i (t) represents the vibration signal of the i-th sensor; N represents the signal a. i The total number of sampling points for (t); is the mean of the vibration signal; s is the standard deviation of the vibration signal;

[0113] Alternatively, based on the intrinsic mode function components, several centroid frequencies of the first time series can be calculated, specifically as follows:

[0114]

[0115] In the formula, CF iLet f be the centroid frequency of the i-th sensor; K be the total number of frequency components; f k E represents the k-th frequency component. ik Let be the energy of the frequency component of the i-th sensor in the vibration signal.

[0116] In the first embodiment of the present invention, for each sensor, the intrinsic mode function components obtained by the empirical mode decomposition method can be used to calculate the characteristic values ​​of its vibration signal, and then the fault location and fault type can be determined based on the characteristic values. The characteristic values ​​of the vibration signal include the energy entropy H. i Waveform factor WF i , kurtosis K i and the center of gravity frequency CF i Energy entropy H i It characterizes the uniformity of signal energy distribution, reflecting the complexity and randomness of the vibration signal; waveform factor WF i It reflects the sharpness of the signal morphology; kurtosis K i The tail thickness of the signal probability distribution was quantified, and the centroid frequency indicated the CF. i The location where signal energy is concentrated in the frequency domain.

[0117] Step 104: Using a preset vibration positioning model, determine the fault location based on the position data and characteristic values ​​of each sensor.

[0118] Furthermore, in the first embodiment of the present invention, a preset vibration positioning model is used to determine the fault location based on the position data and feature values ​​of each sensor, specifically as follows:

[0119] Using the position data and feature values ​​of each sensor, a preset vibration positioning model is fitted to obtain the relationship function between the feature values ​​and the position.

[0120] Based on the relationship function, the theoretical characteristic value corresponding to each sensor is calculated according to the position data of each sensor;

[0121] Obtain candidate position data corresponding to the two sensors with the largest theoretical eigenvalues;

[0122] The intermediate value of the two candidate location data is determined as the fault location.

[0123] Furthermore, in the first embodiment of the present invention, the vibration positioning model is specifically as follows:

[0124] f(x) = a·e -bx +c

[0125] In the formula, f(x) is the characteristic value of the sensor; x is the position data of the sensor; a, b and c are the parameters to be fitted.

[0126] In the first embodiment of the present invention, when performing fault location analysis of underground cables, an empirical mathematical model can be used to quantitatively describe the attenuation law of vibration characteristic quantities with distance. This mathematical model can be expressed as a function f(x) with respect to distance x: f(x) = a·e -bx +c, where a, b, and c are key model parameters that reveal the physical phenomena related to vibration characteristic quantities and the intrinsic connection between them and cable fault location. Parameter a represents the initial intensity of the vibration characteristic quantity at the vibration source (i.e., the possible cable fault point), representing the initial amplitude of the vibration signal and closely related to the cable's response amplitude when subjected to external interference. A high a value usually indicates a large interference source, suggesting more severe cable damage or fault. Parameter b reflects the attenuation rate of the vibration characteristic quantity with increasing distance, reflecting the attenuation characteristics of characteristic quantities such as frequency and amplitude during propagation. A larger b value indicates that the vibration signal attenuates faster; by comparing and analyzing the attenuation rates at different distances, the range of the fault location can be accurately determined. Parameter c represents the level of vibration characteristic quantity that remains unchanged when the vibration propagates over a long distance, i.e., the residual vibration level. This reflects that regardless of the distance, a certain degree of vibration characteristic quantity can always be detected, which is related to the conductivity of the medium or the structure of the cable itself.

[0127] In the first embodiment of this invention, after extracting the feature values ​​corresponding to each sensor, the function f(x) described above is fitted using the position data and feature values ​​of each sensor to obtain the values ​​of parameters a, b, and c. Then, the position data of each sensor is substituted into the function f(x) to calculate the magnitude of the corresponding feature value for each sensor, and the two largest feature values ​​are selected. The midpoint between the positions of the first and second largest feature values ​​is determined as the cable fault location. This invention achieves fault location based on vibration attenuation and feature quantity attenuation laws, and is suitable for identifying and locating cable faults caused by external factors such as mechanical excavation, manual construction, and pile driver operations. By analyzing the abnormal vibration feature quantities and their attenuation patterns recorded in the experiment, the fault source and fault location can be determined quickly and accurately.

[0128] Step 105: Match the feature value corresponding to the fault location in a preset database to obtain the fault type corresponding to the fault location.

[0129] Furthermore, in the first embodiment of the present invention, the feature value corresponding to the fault location is matched in a preset database to obtain the fault type corresponding to the fault location, specifically as follows:

[0130] In the first time series, the feature value corresponding to the fault location is obtained;

[0131] The feature value corresponding to the fault location is compared with several data thresholds in a preset database to obtain several similarity scores.

[0132] Obtain the first fault type corresponding to the threshold of data with the highest similarity;

[0133] The first fault type is determined as the fault type corresponding to the fault location.

[0134] In the first embodiment of this invention, a preset database stores vibration data of cables under different operating conditions. The data stored in the preset database is obtained through large-scale experimental studies, which include capturing the vibration characteristics of cables under various predefined states to ensure that the obtained data covers all kinds of operating conditions that cables may encounter. After obtaining vibration data through experimental studies, the data is preprocessed and analyzed to construct a comprehensive vibration feature matching library. The vibration feature matching library contains feature data of various vibration modes and their corresponding cable state information.

[0135] In the first embodiment of the present invention, after determining the fault location, the feature value corresponding to the fault location is obtained and compared with the feature values ​​pre-stored in the database. By setting a threshold or fluctuation range, the feature value corresponding to the fault location can be identified as belonging to which type of fault, thereby determining the fault type of the fault location.

[0136] In the first embodiment of the present invention, compared with sending the feature values ​​extracted by sensors at all different locations into the database for matching when the fault location is not determined, determining the fault location first and then matching the feature values ​​of the fault location with the database can improve the efficiency of data matching and obtain more accurate fault type matching results.

[0137] As an example of the first embodiment of the present invention, see Figure 4 This is a schematic diagram of an embodiment of the vibration attenuation-based underground cable fault location system provided by the present invention. Vibration sensors are used to capture the vibration characteristics of the cable in real time, including key parameters such as frequency, amplitude, and duration. Multiple analog front-end circuits are used to analyze the data collected by each sensor to obtain a first time series. The first time series is transmitted to a data terminal via an analog-to-digital converter and a microcontroller, enabling the data terminal to analyze the cable's vibration response to identify potential abnormal vibration modes and thus perform real-time assessment of the cable's operating status. This underground cable fault location system can continuously monitor and provide fault warnings for the cable by analyzing its vibration response, ensuring the safety and reliability of power transmission.

[0138] As an example of the first embodiment of the present invention, see Figure 5This is a flowchart illustrating an embodiment of the vibration attenuation-based underground cable fault location method provided by the present invention. Vibration data is collected by sensors, and the collected data undergoes adaptive filtering to remove noise. The data is then standardized or normalized to accurately record the cable vibration data in a time-series manner. A variational modal analysis algorithm is applied to decompose the vibration data, outputting multiple intrinsic modes (IMFs). Time-frequency analysis is performed on each IMF to extract features such as frequency and amplitude. A vibration location model is constructed to analyze the vibration type and its attenuation law, and the cable fault is located based on the vibration signal attenuation law to obtain the fault location. After obtaining the fault location, the feature values ​​of the fault location are matched with the feature values ​​of different types of faults recorded in the database to obtain the fault type. This invention does not send all the feature values ​​extracted by sensors at different locations into the database for matching; instead, it sends the unattenuated or minimally attenuated feature values ​​of the fault location into the database for matching, improving the data matching method and obtaining a more accurate fault type.

[0139] In summary, the first embodiment of this invention provides a method for locating underground cable faults based on vibration attenuation. It utilizes several sensors to collect cable vibration data, forming a first time series. An empirical mode decomposition (EMD) method is employed to decompose the first time series into several intrinsic mode function (EMF) components. Based on each EMF component, several characteristic values ​​of the first time series are calculated. Using a preset vibration location model, the fault location is determined based on the position data and characteristic values ​​of each sensor. The characteristic values ​​corresponding to the fault location are matched against a preset database to derive the fault type. This invention can effectively issue early warnings when underground buried cables are subjected to external impacts, ensuring the stable operation of the power system and the normal operation of the social economy. It improves the efficiency and accuracy of underground cable fault detection and significantly reduces economic losses and safety risks caused by fault delays, possessing extremely high practical value and promising prospects for widespread application.

[0140] Example 2

[0141] See Figure 6 This is a schematic diagram of an embodiment of the underground cable fault location device based on vibration attenuation provided by the present invention. The device includes a data acquisition module 201, a decomposition module 202, a calculation module 203, a location module 204, and a matching module 205.

[0142] The acquisition module 201 is used to acquire vibration data of the cable using several sensors to form a first time series; wherein the interval between each sensor is equal;

[0143] The decomposition module 202 is used to decompose the first time series into several intrinsic mode function components using the empirical mode decomposition method;

[0144] The calculation module 203 is used to calculate a number of feature values ​​of the first time series based on each of the intrinsic mode function components; wherein, the feature values ​​correspond one-to-one with the sensors;

[0145] The positioning module 204 is used to determine the fault location based on the position data and characteristic values ​​of each sensor using a preset vibration positioning model;

[0146] The matching module 205 is used to match the feature values ​​corresponding to the fault location in a preset database to obtain the fault type corresponding to the fault location.

[0147] Furthermore, in the second embodiment of the present invention, the empirical mode decomposition method is used to decompose the first time series into several intrinsic mode function components, specifically as follows:

[0148] The first time series is determined as the original signal;

[0149] The original signal is decomposed into an approximate trend term and a residual component in a loop until the residual component is a monotonic function.

[0150] The approximate trend term obtained from each decomposition is determined as the intrinsic mode function component.

[0151] Furthermore, in the second embodiment of the present invention, the original signal is decomposed into an approximate trend term and a residual component, specifically as follows:

[0152] Obtain several maxima and several minima from the original signal;

[0153] Interpolation is performed on several of the maxima to obtain the upper envelope, and interpolation is performed on several of the minima to obtain the lower envelope;

[0154] Calculate the mean of the upper envelope and the lower envelope to obtain an approximate trend term:

[0155] Subtract the approximate trend term from the original signal to obtain the residual component;

[0156] Determine whether the residual component is a monotonic function;

[0157] If the residual component is not a monotonic function, then the residual component is determined as the original signal.

[0158] Furthermore, in the second embodiment of the present invention, the step of calculating several feature values ​​of the first time series based on each of the intrinsic mode function components further includes:

[0159] Based on the intrinsic mode function components, several energy entropies of the first time series are calculated, specifically as follows:

[0160]

[0161] In the formula, H i p is the energy entropy of the i-th sensor; ij is the ratio of the energy contained in the j-th intrinsic mode function component of the i-th sensor to the total energy; m is the number of intrinsic mode function components;

[0162] Alternatively, based on each of the intrinsic mode function components, several waveform factors of the first time series are calculated, specifically as follows:

[0163]

[0164] In the formula, WF i Let a be the waveform factor of the i-th sensor; i (t) represents the vibration signal of the i-th sensor; N represents the signal a. i The total number of sampling points for (t);

[0165] Alternatively, based on each of the intrinsic mode function components, several kurtosis values ​​of the first time series can be calculated, specifically as follows:

[0166]

[0167] In the formula, K i Let a be the kurtosis of the i-th sensor; i (t) represents the vibration signal of the i-th sensor; N represents the signal a. i The total number of sampling points for (t); is the mean of the vibration signal; s is the standard deviation of the vibration signal;

[0168] Alternatively, based on the intrinsic mode function components, several centroid frequencies of the first time series can be calculated, specifically as follows:

[0169]

[0170] In the formula, CF i Let f be the centroid frequency of the i-th sensor; K be the total number of frequency components; f k E represents the k-th frequency component. ik Let be the energy of the frequency component of the i-th sensor in the vibration signal.

[0171] Furthermore, in the second embodiment of the present invention, a preset vibration positioning model is used to determine the fault location based on the position data and feature values ​​of each sensor, specifically as follows:

[0172] Using the position data and feature values ​​of each sensor, a preset vibration positioning model is fitted to obtain the relationship function between the feature values ​​and the position.

[0173] Based on the relationship function, the theoretical characteristic value corresponding to each sensor is calculated according to the position data of each sensor;

[0174] Obtain candidate position data corresponding to the two sensors with the largest theoretical eigenvalues;

[0175] The intermediate value of the two candidate location data is determined as the fault location.

[0176] Furthermore, in the second embodiment of the present invention, the vibration positioning model is specifically as follows:

[0177] f(x) = a·e -bx +c

[0178] In the formula, f(x) is the characteristic value of the sensor; x is the position data of the sensor; a, b and c are the parameters to be fitted.

[0179] Furthermore, in the second embodiment of the present invention, the feature value corresponding to the fault location is matched in a preset database to obtain the fault type corresponding to the fault location, specifically as follows:

[0180] In the first time series, the feature value corresponding to the fault location is obtained;

[0181] The feature value corresponding to the fault location is compared with several data thresholds in a preset database to obtain several similarity scores.

[0182] Obtain the first fault type corresponding to the threshold of data with the highest similarity;

[0183] The first fault type is determined as the fault type corresponding to the fault location.

[0184] Furthermore, in the second embodiment of the present invention, vibration data of the cable is collected using several sensors to form a first time series, specifically as follows:

[0185] Obtain the starting point position and preset interval distance of the initial sensor installation;

[0186] Calculate the position data of several sensors based on the installation starting point position of the initial sensor and the preset interval distance;

[0187] Based on the position data of each sensor, each sensor is positioned along the longitudinal direction of the cable;

[0188] Vibration data of the cable are collected using the sensors described above to form a first time series.

[0189] Furthermore, in the second embodiment of the present invention, vibration data of the cable is collected using each of the sensors to form a first time series, specifically as follows:

[0190] After collecting vibration data of the cable using the sensors, the vibration data is denoised using an adaptive filtering algorithm.

[0191] The noise-reduced vibration data is then standardized to form the first time series.

[0192] In summary, the second embodiment of this invention provides a vibration attenuation-based underground cable fault location device. Based on the organic integration of modules, it utilizes several sensors to collect cable vibration data, forming a first time series. An empirical mode decomposition method is used to decompose the first time series into several intrinsic mode function components. Based on each intrinsic mode function component, several characteristic values ​​of the first time series are calculated. Using a preset vibration location model, the fault location is determined based on the position data and characteristic values ​​of each sensor. The characteristic values ​​corresponding to the fault location are matched against a preset database to obtain the fault type corresponding to the fault location. This invention can effectively issue early warnings when underground direct-buried cables are subjected to external impacts, ensuring the stable operation of the power system and the normal operation of the social economy; it improves the efficiency and accuracy of underground cable fault detection; and it significantly reduces economic losses and safety risks caused by fault delays, possessing extremely high practical value and promising prospects for widespread application.

[0193] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for locating a fault in an underground cable based on vibration attenuation, characterized in that, The utility model relates to a kind of method for locating the fault position of cable, comprising: Collect vibration data of cable using several sensors to form first time series;Wherein, interval distance of each sensor is equal; Adopt empirical mode decomposition method, the first time series is decomposed into several intrinsic mode function components; According to each intrinsic mode function component, the characteristic value of the first time series is calculated;Wherein, the characteristic value and the sensor one-to-one correspond; Determine the fault position according to the position data and characteristic value of each sensor using preset vibration positioning model; The characteristic value corresponding to the fault position is matched in the preset database, and the fault type corresponding to the fault position is obtained; Wherein, the utility model determines the fault position according to the position data and characteristic value of each sensor using preset vibration positioning model, specifically: Using the position data and characteristic value of each sensor, the preset vibration positioning model is fitted to obtain the relationship function between characteristic value and position; Based on the relationship function, the theoretical characteristic value corresponding to each sensor is calculated according to the position data of each sensor; The candidate position data corresponding to the two sensors with the maximum theoretical characteristic value is obtained; The middle value of the two candidate position data is determined as the fault position; The vibration positioning model, specifically: wherein is a characteristic value of the sensor; is position data of the sensor; , and are parameters to be fitted, respectively.

2. The vibration attenuation based underground cable fault location method of claim 1, wherein, The first time series is decomposed into several intrinsic mode function components using empirical mode decomposition method, specifically: The first time series is determined as original signal; The original signal is decomposed into approximate trend item and residual component in a loop until the residual component is monotonic function; Approximate trend item obtained in each decomposition is determined as intrinsic mode function component.

3. The vibration attenuation based underground cable fault location method of claim 2, wherein, The original signal is decomposed into approximate trend item and residual component, specifically: Obtain several maximum points and several minimum points in the original signal; Interpolate several maximum points to obtain upper envelope, and interpolate several minimum points to obtain lower envelope; Calculate the mean value of the upper envelope and the lower envelope to obtain approximate trend item: Subtract the approximate trend item from the original signal to obtain residual component; Determine whether the residual component is monotonic function; If the residual component is not monotonic function, the residual component is determined as original signal.

4. The vibration attenuation based underground cable fault location method of claim 1, wherein, According to each intrinsic mode function component, the characteristic value of the first time series is calculated, further comprising: According to each intrinsic mode function component, the energy entropy of the first time series is calculated, specifically: wherein is the energy entropy of the i-th sensor; is the ratio of the energy contained in the j-th eigenmode function component of the i-th sensor to the total energy; is the number of eigenmode function components. Or, according to each intrinsic mode function component, the waveform factor of the first time series is calculated, specifically: wherein is the waveform factor of the i-th sensor; is the vibration signal of the i-th sensor; is the total number of sampling points of the signal is the total number of sampling points of the signal Or, according to each intrinsic mode function component, the kurtosis of the first time series is calculated, specifically: wherein is the kurtosis of the i-th sensor; is the vibration signal of the i-th sensor; is the signal is the total number of sampling points of the signal is the mean of the vibration signal; is the standard deviation of the vibration signal; Or, according to each intrinsic mode function component, the barycentric frequency of the first time series is calculated, specifically: wherein is the center frequency of the i-th sensor; is the total number of frequency components; is the k-th frequency component; is the energy of the frequency component of the i-th sensor in the vibration signal.

5. The vibration attenuation based underground cable fault location method of claim 1, wherein, The characteristic value corresponding to the fault position is matched in the preset database, and the fault type corresponding to the fault position is obtained, specifically: In the first time series, the characteristic value corresponding to the fault position is obtained; The characteristic value corresponding to the fault position is compared with a plurality of data thresholds in a preset database, and a plurality of similarities are obtained; A first fault type corresponding to a data threshold with the highest similarity is obtained; The first fault type is determined as the fault type corresponding to the fault position.

6. The vibration attenuation based underground cable fault location method of claim 1, wherein, The vibration data of the cable is collected by using a plurality of sensors to form a first time sequence, specifically: An installation starting point position of a starting sensor and a preset interval distance are obtained; The position data of a plurality of sensors are calculated according to the installation starting point position of the starting sensor and the preset interval distance; Each of the sensors is arranged along the longitudinal direction of the cable according to the position data of each of the sensors; The vibration data of the cable is collected by using each of the sensors to form a first time sequence.

7. The vibration attenuation based underground cable fault location method of claim 6, wherein, The vibration data of the cable is collected by using each of the sensors to form a first time sequence, specifically: After the vibration data of the cable is collected by using each of the sensors, an adaptive filtering algorithm is used to denoise the vibration data; The denoised vibration data is subjected to data standardization processing to form a first time sequence.

8. A device for locating a fault in an underground cable based on attenuation of vibrations, characterized in that It comprises: a collection module, a decomposition module, a calculation module, a positioning module and a matching module; The collection module is used to collect the vibration data of the cable by using a plurality of sensors to form a first time sequence; wherein the interval distances of each of the sensors are equal; The decomposition module is used to decompose the first time sequence into a plurality of intrinsic mode function components by using an empirical mode decomposition method; The calculation module is used to calculate a plurality of characteristic values of the first time sequence according to each of the intrinsic mode function components; wherein the characteristic values and the sensors are one-to-one corresponding; The positioning module is used to determine a fault position according to the position data and the characteristic values of each of the sensors by using a preset vibration positioning model; The matching module is used to match the characteristic value corresponding to the fault position in a preset database to obtain the fault type corresponding to the fault position; The vibration positioning model is used to determine the fault position according to the position data and the characteristic values of each of the sensors, specifically: The position data and the characteristic values of each of the sensors are used to fit the preset vibration positioning model to obtain a relationship function between the characteristic values and the positions; The theoretical characteristic values corresponding to each of the sensors are calculated according to the position data of each of the sensors based on the relationship function; The candidate position data corresponding to the two sensors with the maximum theoretical characteristic values are obtained; The middle value of the two candidate position data is determined as the fault position; The vibration positioning model is specifically: wherein is a characteristic value of the sensor; is position data of the sensor; , and are parameters to be fitted, respectively.

Images