Cable damage early warning method, device and early warning system

By setting up a cable damage warning device in the upper layer of the cable, collecting and analyzing surface vibration signals, and using a sound classification model to conduct early warning, the problem of high maintenance costs and inability to be warned in the existing technology is solved, and the cable safety is improved and Reduced operation and maintenance costs.

CN119028108AActive Publication Date: 2024-11-26GUANGZHOU POWER SUPPLY BUREAU GUANGDONG POWER GRID CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202411174464.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-26
Publication Date
2024-11-26
Estimated Expiration
2044-08-26

AI Technical Summary

Technical Problem

In the prior art, by setting up cable signs or cable piles in the cable corridor for warning, the maintenance cost is high and it is impossible to warning of external force damage incidents in advance, resulting in major hidden dangers in the safe operation of the cable.

Method used

Set up a cable damage warning device in the upper soil of the cable to collect surface vibration signals, generate feature vector sets through signal separation and time-frequency analysis, and call the preset sound classification model for early warning. When a construction type signal is detected, the warning voice is triggered. And send a notification.

Benefits of technology

Early warning is achieved to avoid external force damage, reduce the number of manual inspections, reduce operation and maintenance costs, and improve cable safety.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119028108A_ABST
    Figure CN119028108A_ABST
Patent Text Reader

Abstract

The invention provides a cable damage early warning method, device and early warning system, and the method comprises the steps: collecting a vibration signal of a ground surface, determining a corresponding feature vector set according to the vibration signal, calling a preset sound classification model, inputting the feature vector set into the sound classification model, obtaining sound information corresponding to the feature vector set, and carrying out the early warning of cable damage. And when the sound information belongs to the construction type, triggering a warning voice, and sending the sound information to each terminal in a preset terminal list. The early warning system is arranged in the upper soil of the underground cable, so that the cable damage early warning device in the early warning system can collect the vibration signals of the ground surface, early warning is carried out in advance in combination with a sound classification model, early warning is carried out in the mode of on-site voice warning and online notification of operation and maintenance personnel, and the good warning effect is achieved; in addition, the number of times of manual inspection can be reduced, and the operation and maintenance cost is reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of power grid safety technology, and in particular to a cable damage early warning method, device and early warning system. Background Art

[0002] With the rapid development of urban construction, municipal construction, road excavation or outdoor operations have been carried out on a large scale. Although the development of these projects has brought prosperity and development to the city, it has also brought some challenges. In particular, during the operation of cable lines, since the laying of cable lines needs to span a large distance, and its operation is closely related to the surrounding environment, accidents of external forces damaging cables occur frequently, which in turn brings great hidden dangers to the safe operation of cables.

[0003] At present, the application of 10KV cable power supply is becoming more and more widespread. Generally speaking, dedicated transformer users receive electricity through switch rooms or overhead lines, and the power is supplied to the user's power room through cables. The incoming cables of dedicated transformer users are often laid by jacking pipes, direct burial, etc. In case of external force damage to the cables, cable signs or cable piles will be set up in the cable corridor to provide reminders on site. However, this method not only has high maintenance costs, but also cannot provide early warning of external force damage events. Summary of the invention

[0004] The purpose of the present application is to solve at least one of the above-mentioned technical defects, especially the technical defect that cable identification signs or cable piles are set up in the cable corridor for warning in the prior art, but this method not only has high maintenance costs, but also cannot provide early warning of external force damage events.

[0005] In a first aspect, the present application provides a cable damage warning method, which is applied to a cable damage warning device in an early warning system, wherein the early warning system is placed in the upper soil of an underground cable; the method comprises:

[0006] Collect vibration signals from the ground surface;

[0007] Determine a corresponding feature vector set according to the vibration signal;

[0008] Calling a preset sound classification model, and inputting the feature vector set into the sound classification model to obtain sound information corresponding to the feature vector set;

[0009] When the sound information belongs to the construction type, a warning voice is triggered, and the sound information is sent to each terminal in the preset terminal list.

[0010] In one embodiment, the method further comprises:

[0011] When no vibration signal from the ground surface is collected within a preset time period, the low power consumption operation mode is activated.

[0012] In one embodiment, the step of determining a corresponding feature vector set according to the vibration signal comprises:

[0013] Performing signal separation on the vibration signal to obtain a background sound signal and a disruptive sound signal;

[0014] Time-frequency analysis is performed on the background sound signal and the destruction sound signal respectively to extract energy distribution features in the background sound signal and the destruction sound signal, and a feature vector set is generated according to the energy distribution features of the background sound signal and the energy distribution features of the destruction sound signal.

[0015] In one embodiment, the step of performing signal separation on the vibration signal comprises:

[0016] Preprocessing the vibration signal to obtain a signal to be analyzed;

[0017] The signal to be analyzed is subjected to independent component analysis to separate background sound signals and destruction sound signals from the signal to be analyzed.

[0018] In one embodiment, the step of generating a feature vector set includes:

[0019] Determine the energy value of the background sound signal in each frequency band according to the energy distribution characteristics of the background sound signal, so as to generate a first feature vector according to the energy value of the background sound signal in each frequency band;

[0020] Determining the energy value of the destruction sound signal in each frequency band according to the energy distribution characteristics of the destruction sound signal, so as to generate a second feature vector according to the energy value of the destruction sound signal in each frequency band;

[0021] A feature vector set is formed by the first feature vector and the second feature vector.

[0022] In one of the embodiments, the sound classification model is trained based on a Transformer neural network architecture.

[0023] In one embodiment, the training process of the sound classification model includes:

[0024] Obtaining a training data set and its corresponding sample data set; wherein the training data set includes single-class data samples and combined data samples;

[0025] Determine the Transformer model;

[0026] The Transformer model is iteratively trained based on the training data set and the sample data set to obtain a trained Transformer model, and the trained Transformer model is determined as a sound classification model.

[0027] In a second aspect, the present application provides a cable damage warning device, the device comprising a wireless module, an on-site alarm module, a vibration monitoring module, and a voiceprint monitoring module;

[0028] The vibration monitoring module is used to collect vibration signals from the ground surface and upload the vibration signals to the soundprint monitoring module;

[0029] The voiceprint monitoring module is used to determine a corresponding feature vector set according to the received vibration signal, and call a preset sound classification model, input the feature vector set into the sound classification model to obtain sound information corresponding to the feature vector set, and send the sound information to the wireless module and the on-site alarm module when the sound information belongs to the construction type;

[0030] The on-site alarm module is used to trigger a warning voice when receiving the sound information;

[0031] The wireless module is used to send the received sound information to each terminal in the preset terminal list.

[0032] In one of the embodiments, the device further includes a power management module; the power management module is used to start a low power consumption operation mode when the vibration monitoring module does not collect a vibration signal within a preset time period.

[0033] In a third aspect, the present application provides an early warning system, the system comprising a housing, an upper cover, a vibration spring, and a cable damage early warning device as described in any one of the above embodiments;

[0034] The cable damage warning device is placed inside the housing, and the upper cover is used to close the housing to protect the cable damage warning device;

[0035] The cable damage warning device is used to execute the cable damage warning method as described in any one of the above embodiments;

[0036] The vibration spring is fixed to the bottom of the shell and is used to receive vibration signals from the ground surface.

[0037] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0038] The cable damage warning method, device and warning system provided by the present application are applied to the cable damage warning device in the warning system, and the warning system is placed in the upper soil of the underground cable; including: collecting the vibration signal of the surface, and then determining the corresponding feature vector set according to the vibration signal, and then calling the preset sound classification model, and inputting the feature vector set into the sound classification model to obtain the sound information corresponding to the feature vector set, when the sound information belongs to the construction type, triggering the warning voice, and sending the sound information to each terminal in the preset terminal list. By setting the early warning system in the upper soil of the underground cable, the cable damage early warning device in the early warning system can collect the vibration signal of the surface, so as to combine the sound classification model for early warning, and warn in the form of on-site voice warning and online notification of operation and maintenance personnel, so as to achieve a better warning effect, so that on the one hand, external force damage can be avoided to the greatest extent, and on the other hand, the on-site environment of the cable can be notified to the operation and maintenance personnel in advance, and the number of manual inspections can be reduced, and the operation and maintenance costs can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0040] Figure 1 A schematic diagram of a process flow of a cable damage warning method provided in an embodiment of the present application;

[0041] Figure 2 A schematic flow chart of the steps of determining a corresponding feature vector set according to a vibration signal provided in an embodiment of the present application;

[0042] Figure 3 One of the structural schematic diagrams of a cable damage warning device provided in an embodiment of the present application;

[0043] Figure 4 A second structural schematic diagram of a cable damage warning device provided in an embodiment of the present application;

[0044] Figure 5 A schematic diagram of the structure of an early warning system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0046] In one of the embodiments, the present application provides a cable damage warning method. The following embodiments illustrate the method by applying it to a cable damage warning device in a warning system. It is understandable that when cable signs or cable piles are set for warning, the cable is often damaged before the operation and maintenance personnel can learn of the cable fault information. In addition, more cable signs or cable piles need to be built to achieve a slightly better warning effect, resulting in a high construction cost. In addition, maintenance personnel are required to conduct regular inspections during this process. Based on this, the present application provides a cable damage warning method, and places the warning system that executes the method in the upper soil of the underground cables. Figure 1 As shown, the cable damage early warning method includes:

[0047] Step S101: collecting vibration signals from the ground surface.

[0048] Among them, the vibration signal refers to the signal generated by the vibration of the object.

[0049] In this step, since the early warning system is placed in the upper soil of the underground cable, when there is construction activity in the field environment where the underground cable is located, the vibration signal on the surface can be collected and analyzed. Moreover, sound propagates fastest in solids, and the vibration signal can be quickly obtained by collecting the vibration signal on the surface, thereby improving the response speed of the cable damage early warning method.

[0050] Furthermore, when early warning of underground cables is needed, the areas can be divided according to the distribution of the underground cables, and an early warning system can be set up in the divided areas. When the early warning system is set up, the early warning system is started. At this time, the early warning system will detect and collect surface vibration signals in real time.

[0051] It is understandable that the density of the early warning system distribution and the specific location of the early warning system can be adjusted and set according to the actual environment and actual needs, and this application does not impose specific restrictions on this. For example, when the accuracy of the early warning is required to be high, the distribution density of the early warning system can be increased to enhance the strength of the vibration signal.

[0052] Specifically, when there is construction on site where underground cables are located, construction noise will generally be generated, which mainly comes from the following aspects:

[0053] (1) Mechanical noise: The mechanical noise of various types of construction machinery and equipment on site, such as excavators, pile drivers, cranes, graders, and transportation vehicles on site, can usually be regarded as point sound sources compared to the entire construction site.

[0054] (2) On-site operation noise: During the construction process, the collision, cutting and grinding noises between construction machinery and building materials or the ground, such as the sound of a pile driver hitting the ground, the sound of a cutting machine cutting steel bars, and the sound of a grinder rubbing against the ground, may reach an instantaneous sound pressure level of over 130 dB(A).

[0055] (3) Collision noise: noise caused by collision during transportation or vehicle loading and unloading, scattered noise caused by loading and unloading scaffolding and formwork, as well as the knocking sound of workers, the impact sound of metal materials, etc., which are usually instantaneous noise.

[0056] (4) Construction worker noise: the noise generated by workers talking, shouting, using hand tools, etc. at the construction site.

[0057] Step S102: Determine a corresponding feature vector set according to the vibration signal.

[0058] The feature vector set describes the distribution of the vibration signal on different frequency components.

[0059] In this step, when the vibration signal is collected, it is necessary to perform separation and component analysis on the vibration signal, and then obtain a feature vector set corresponding to the vibration signal to match the input specification of the sound classification model.

[0060] Step S103: calling a preset sound classification model, and inputting the feature vector set into the sound classification model to obtain sound information corresponding to the feature vector set.

[0061] The sound information refers to classification information and damage level information. The sound classification model is used to analyze the sound features in the feature vector set to determine the classification information and damage level information corresponding to the feature vector set.

[0062] In this step, a pre-trained sound classification model is called, and then the feature vector set is input into the sound classification model so that the sound classification model performs inference based on the feature vector set and outputs the analysis result of the feature vector set, namely, the sound information.

[0063] Step S104: When the sound information belongs to the construction type, a warning voice is triggered, and the sound information is sent to each terminal in the preset terminal list.

[0064] The construction type is any one or more combinations of the preset construction type list.

[0065] In this step, the classification information in the sound information is first obtained, and based on the classification information, it is determined whether the sound information belongs to any one or more combinations in the preset construction type list. If not, the cable damage warning is terminated and the vibration information on the surface is continued to be collected; if so, a warning voice is triggered to remind the surrounding construction teams to pay attention to the location of the underground cables to avoid damaging the underground cables. In addition, the sound information will be sent to each terminal in the terminal list. This allows the operation and maintenance personnel to understand the specific construction type (classification information) and degree of damage (damage level information) around the underground cables.

[0066] It is understandable that the terminal list can be set by the operation and maintenance personnel. For example, the early warning system in different areas can set the terminal list according to the person in charge of the corresponding area. When the early warning system believes that there is a construction team around, it can send specific information to the corresponding terminal according to the predetermined terminal list to achieve the purpose of notifying the operation and maintenance personnel in advance.

[0067] In one example, the communication method for sending the sound information adopts wireless communication, and the sound information can be sent by setting an antenna in a 4G / 5G communication manner.

[0068] In another embodiment, a light emitting device may be provided on the early warning system. When the sound information belongs to the construction type, the on-site prompting measure may not only trigger the warning voice but also flash the light of the light emitting device as a warning.

[0069] The cable damage warning method, device and warning system provided by the present application are applied to the cable damage warning device in the warning system, and the warning system is placed in the upper soil of the underground cable; including: collecting the vibration signal of the surface, and then determining the corresponding feature vector set according to the vibration signal, and then calling the preset sound classification model, and inputting the feature vector set into the sound classification model to obtain the sound information corresponding to the feature vector set, when the sound information belongs to the construction type, triggering the warning voice, and sending the sound information to each terminal in the preset terminal list. By setting the early warning system in the upper soil of the underground cable, the cable damage early warning device in the early warning system can collect the vibration signal of the surface, so as to combine the sound classification model for early warning, and warn in the form of on-site voice warning and online notification of operation and maintenance personnel, so as to achieve a better warning effect, so that on the one hand, external force damage can be avoided to the greatest extent, and on the other hand, the on-site environment of the cable can be notified to the operation and maintenance personnel in advance, and the number of manual inspections can be reduced, and the operation and maintenance costs can be reduced.

[0070] In one embodiment, the cable damage early warning method further includes:

[0071] When no vibration signal from the ground surface is collected within a preset time period, the low power consumption operation mode is activated.

[0072] The preset period refers to a preset time length. For example, the preset period can be set to 40 minutes, 1 hour, or 3 hours, etc.

[0073] In this embodiment, when no vibration signal from the ground surface is collected within the preset interval, it indicates that there is a high probability that there is no construction activity in the on-site environment of the underground cable. At this time, the power consumption of the cable damage warning device can be reduced by reducing the operating frequency of some modules in the cable damage warning device, thereby entering a low-power operation mode. In this way, the low-power operation mode can be entered when the risk of damage is low, and resource consumption can be reduced while ensuring the safety of underground cables, thereby improving system resource utilization.

[0074] like Figure 2 As shown, in one embodiment, the step of determining the corresponding feature vector set according to the vibration signal includes:

[0075] Step S201: Separate the vibration signal to obtain a background sound signal and a disruptive sound signal.

[0076] Among them, background sound signals refer to sound signals that are unrelated to construction activities, such as normal sound signals such as the sound of passing cars, human voices, and thunder. Destructive sound signals refer to sound signals generated by the construction team during construction activities.

[0077] In this step, since the collected vibration signal can be roughly divided into two categories: background sound and destructive sound, the vibration signal is separated by using signal separation technology to obtain background sound signal and destructive sound signal. Specifically, the signal separation technology includes but is not limited to independent component analysis, deep learning, non-negative matrix decomposition or adaptive filter, etc. The specific signal separation technology can be selected according to specific needs, and this application does not make specific restrictions on this.

[0078] Step S202: performing time-frequency analysis on the background sound signal and the destruction sound signal respectively to extract energy distribution features in the background sound signal and the destruction sound signal.

[0079] Among them, the energy distribution characteristics are used to describe the energy distribution of the signal at different frequencies.

[0080] In this step, the background sound signal and the destruction sound signal can be converted from the time domain to the frequency domain using methods such as Fourier transform or wavelet transform, so as to perform time domain analysis and observe the distribution of the signal at different frequencies.

[0081] Furthermore, in the embodiment, the energy distribution characteristics of the signal are particularly concerned during the time-frequency analysis. The energy distribution characteristics reflect the strength of the signal at different frequencies, which is crucial for identifying and distinguishing different types of sound signals. For example, the background sound signal may have higher energy in the low-frequency range, while the disruptive sound signal may show more energy concentration in the high-frequency range. By analyzing these energy distribution characteristics, the composition of the sound signal can be better understood, and important information can be provided for subsequent sound signal processing and sound recognition tasks.

[0082] Step S203: Generate a feature vector set according to the energy distribution characteristics of the background sound signal and the energy distribution characteristics of the destruction sound signal.

[0083] In one embodiment, the step of performing signal separation on the vibration signal includes:

[0084] Preprocess the vibration signal to obtain the signal to be analyzed;

[0085] The signal to be analyzed is subjected to independent component analysis to separate the background sound signal and the destructive sound signal from the signal to be analyzed.

[0086] In this embodiment, the vibration signal is firstly centralized, that is, the corresponding mean is subtracted from each variable value, so that the calculation can be simplified to complete the preprocessing of the vibration signal, and obtain the preprocessed vibration signal, that is, the signal to be analyzed, thereby improving the processing performance of the subsequent independent cost analysis. Then, the independent component analysis is performed on the signal to be analyzed to find the independent components in the signal to be analyzed, and the background sound signal and the destructive sound signal are separated from the signal to be analyzed based on the two categories of background sound and destructive sound.

[0087] It can be understood that independent component analysis can effectively identify independent components in the signal to be analyzed, so that the background sound signal and the damage sound signal that were originally mixed together can be clearly distinguished, so that different sound signals can be further analyzed later, thereby improving the accuracy of the final output sound information and improving the accuracy of the cable damage warning.

[0088] In one example, the independent component analysis method adopted in this embodiment is the FastICA (Fast Independent Component Analysis) method.

[0089] In one embodiment, the step of generating a feature vector set includes:

[0090] Determine the energy value of the background sound signal in each frequency band according to the energy distribution characteristics of the background sound signal, so as to generate a first feature vector according to the energy value of the background sound signal in each frequency band;

[0091] Determine the energy value of the destructive sound signal in each frequency band according to the energy distribution characteristics of the destructive sound signal, so as to generate a second feature vector according to the energy value of the destructive sound signal in each frequency band;

[0092] A feature vector set is formed by the first feature vector and the second feature vector.

[0093] The first eigenvector is used to describe the intensity of the background sound signal in different frequency bands, and the second eigenvector is used to describe the intensity of the destruction sound signal in different frequency bands.

[0094] In this embodiment, the energy values ​​of the background sound signal and the destruction sound signal in each frequency band are calculated respectively according to the energy distribution characteristics of the background sound signal and the energy distribution characteristics of the destruction sound signal, and the energy values ​​in each frequency band are respectively composed of a first eigenvector corresponding to the background sound signal and a second eigenvector corresponding to the destruction sound signal.

[0095] Exemplarily, assuming that the energy distribution characteristics of the background sound signal involve frequency bands a, b and c, at this time, the total energy value A of frequency band a, the total energy value B of frequency band b and the total energy value C of frequency band c are calculated in sequence according to the energy distribution characteristics, then the first eigenvector corresponding to the background sound signal is [A, B, C].

[0096] It can be understood that the energy value on each frequency band is calculated through the energy distribution characteristics, and then the first eigenvector and the second eigenvector are constructed, so that the sound classification model can better identify the sound information and determine its category information, thereby improving the accuracy of the cable damage warning and reducing the probability of false alarm.

[0097] In one embodiment, the sound classification model is trained based on a Transformer neural network architecture.

[0098] In this embodiment, the Transformer neural network architecture is a deep learning model mainly used to process sequence data. Its core idea is to use the attention mechanism to capture long-distance dependencies in sequence data, and it has good generalization ability.

[0099] In one embodiment, the training process of the sound classification model includes:

[0100] Obtain a training data set and its corresponding sample data set; wherein the training data set includes single-class data samples and combined data samples;

[0101] Determine the Transformer model;

[0102] The Transformer model is iteratively trained based on the training data set and the sample data set to obtain a trained Transformer model, and the trained Transformer model is determined as a sound classification model.

[0103] Specifically, the training data set includes multiple sound samples, including single-category data samples and combined data samples. Single-category data samples refer to sound samples with only one category of information, and combined data samples refer to sound samples mixed with two or more category information. The sample data set includes the sound information of each data sample in the training data set.

[0104] In this embodiment, the Transformer model is iteratively trained using a training data set and its corresponding sample data set until a preset stop condition is met, all network parameters of the trained Transformer model are fixed, and the trained Transformer model is determined as a sound classification model. Furthermore, the stop condition may be reaching a preset number of iterations, or the change in the model parameter between two iterations is less than a preset threshold, and this application does not impose any specific restrictions on this.

[0105] In an example, the workflow of the Transformer model in this embodiment is as follows: first, a single-layer fully connected layer is used to process a vector of length 40 into a vector of length 80 for preliminary feature encoding; secondly, the encoder in the transformer is used for feature extraction, the model dimension is 80, two multi-head attention mechanisms are used, and the internal augmentation is increased to 256 bits; finally, two layers of fully connected layers are used for classification, and the dimension is increased to 600 to obtain a 5-bit output.

[0106] It is understandable that in the process of using the Transformer architecture to train the sound classification model, the attention mechanism of the Transformer model can capture long-distance dependencies and effectively process complex patterns in time series data; at the same time, the Transformer model has parallel processing capabilities, fast training speed, and can converge quickly; and the Transformer model has strong generalization capabilities and can adapt well to different sound classification tasks, thereby improving the accuracy and robustness of the model.

[0107] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0108] The cable damage warning device provided in an embodiment of the present application is described below. The cable damage warning device described below and the cable damage warning method described above can be referenced to each other.

[0109] like Figure 3 As shown, the present application provides a cable damage warning device 300, which includes a wireless module 301, an on-site alarm module 302, a vibration monitoring module 303 and a voiceprint monitoring module 304;

[0110] The vibration monitoring module 303 is used to collect vibration signals from the ground surface and upload the vibration signals to the soundprint monitoring module 304;

[0111] The voiceprint monitoring module 302 is used to determine the corresponding feature vector set according to the received vibration signal, and call the preset sound classification model, input the feature vector set into the sound classification model to obtain the sound information corresponding to the feature vector set, and send the sound information to the wireless module 301 and the on-site alarm module 302 when the sound information belongs to the construction type;

[0112] The on-site alarm module 302 is used to trigger a warning voice when receiving sound information;

[0113] The wireless module 301 is used to send the received sound information to each terminal in the preset terminal list.

[0114] The vibration monitoring module 303 includes a vibration sensor for receiving and collecting vibration signals of the vibrating springs.

[0115] In this embodiment, when the voiceprint monitoring model 302 determines the sound information and judges that the sound information belongs to the construction type, the sound information is sent to the wireless module 301 and the on-site alarm module 302, and then the on-site alarm module 302 can issue a warning voice to remind the construction equipment and construction personnel that there are underground cables here and they are prohibited from damaging them. The wireless module 301 can send the sound information to the terminal of the corresponding operation and maintenance personnel through the set antenna in the form of 4G / 5G communication to remind the cable maintenance personnel.

[0116] in addition, Figure 3 The connection relationship between the modules represents the main data interaction flow and does not constitute a limitation on the specific data interaction in the cable damage early warning device.

[0117] It is understandable that by setting up an early warning system in the upper soil of the underground cable, the cable damage warning device in the early warning system can collect vibration signals from the surface to provide early warning in combination with the sound classification model, and provide early warning in the form of on-site voice warnings and online notifications to operation and maintenance personnel to achieve a better warning effect. On the one hand, this can avoid external force damage to the greatest extent, and on the other hand, it can notify the operation and maintenance personnel of the on-site environment of the cable in advance, and it can also reduce the number of manual inspections and reduce operation and maintenance costs.

[0118] like Figure 4 As shown, in one embodiment, the cable damage warning device 300 further includes a power management module 305; the power management module 305 is used to start the low power consumption operation mode when the vibration monitoring module 303 does not collect a vibration signal within a preset time period.

[0119] In one embodiment, the voiceprint monitoring module includes:

[0120] A signal separation submodule is used to separate the vibration signal to obtain a background sound signal and a damaging sound signal;

[0121] The time-frequency analysis submodule is used to perform time-frequency analysis on the background sound signal and the destruction sound signal respectively to extract the energy distribution characteristics of the background sound signal and the destruction sound signal, and generate a feature vector set according to the energy distribution characteristics of the background sound signal and the energy distribution characteristics of the destruction sound signal.

[0122] In one embodiment, the signal separation submodule includes:

[0123] A preprocessing unit, used for preprocessing the vibration signal to obtain a signal to be analyzed;

[0124] The component analysis unit is used to perform independent component analysis on the signal to be analyzed so as to separate the background sound signal and the destructive sound signal from the signal to be analyzed.

[0125] In one embodiment, the time-frequency analysis submodule includes:

[0126] A first generating unit, configured to determine the energy value of the background sound signal in each frequency band according to the energy distribution characteristics of the background sound signal, so as to generate a first feature vector according to the energy value of the background sound signal in each frequency band;

[0127] A second generating unit is used to determine the energy value of the destruction sound signal in each frequency band according to the energy distribution characteristics of the destruction sound signal, so as to generate a second feature vector according to the energy value of the destruction sound signal in each frequency band;

[0128] The vector set generating unit is used to form a feature vector set using the first feature vector and the second feature vector.

[0129] In one embodiment, the cable damage warning device further comprises:

[0130] The data set acquisition module is used to acquire the training data set and its corresponding sample data set; wherein the training data set includes single-class data samples and combined data samples;

[0131] Model determination module, used to determine the Transformer model;

[0132] The model training module is used to iteratively train the Transformer model based on the training data set and the sample data set to obtain a trained Transformer model, and determine the trained Transformer model as a sound classification model.

[0133] The division of each module in the above cable damage warning device is only for illustration. In other embodiments, the cable damage warning device can be divided into different modules as needed to complete all or part of the functions of the above cable damage warning device. Each module in the above cable damage warning device can be implemented in whole or in part by software, hardware and a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0134] like Figure 5 As shown, the present application provides an early warning system, the system comprising a housing 1, an upper cover 2, a vibration spring 4, and a cable damage early warning device 3 as in any of the above embodiments;

[0135] The cable damage warning device 3 is placed inside the housing 1, and the upper cover 2 is used to close the housing 1 to protect the cable damage warning device 3;

[0136] The cable damage warning device 3 is used to execute the cable damage warning method as described in any of the above embodiments;

[0137] The vibration spring 4 is fixed to the bottom of the housing 1 and is used to receive vibration signals from the ground surface.

[0138] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the present application scheme, and does not constitute a limitation on the early warning system to which the present application scheme is applied. For example, the vibrating spring 3 can be replaced by any device with a vibration response. The specific early warning system may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0139] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish an entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the term "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only include those elements, but also include other elements that are not clearly listed, or also include elements inherent to such process, method, article or equipment. In the absence of more restrictions, the elements limited by the sentence "comprise one..." do not exclude the existence of other identical elements in the process, method, article or equipment including the elements. Herein, the singular "one", "one" and "described / the" may also include plural forms, unless the context clearly indicates another way. It should also be understood that the terms "include / comprise" or "have" etc. specify the existence of stated features, wholes, steps, operations, components, parts or combinations thereof, but do not exclude the possibility of the existence or addition of one or more other features, wholes, steps, operations, components, parts or combinations thereof. At the same time, the term "and / or" used in this specification includes any and all combinations of the relevant listed items.

[0140] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can refer to each other.

[0141] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A cable damage early warning method, characterized in that: A cable damage warning device used in an early warning system, wherein the early warning system is placed in the upper soil of an underground cable; The method comprises: Collect vibration signals from the ground surface; Determine a corresponding feature vector set according to the vibration signal; Calling a preset sound classification model, and inputting the feature vector set into the sound classification model to obtain sound information corresponding to the feature vector set; When the sound information belongs to the construction type, a warning voice is triggered, and the sound information is sent to each terminal in the preset terminal list.

2. The cable damage early warning method according to claim 1, characterized in that: The method further comprises: When no vibration signal from the ground surface is collected within a preset time period, the low power consumption operation mode is activated.

3. The cable damage early warning method according to claim 1, characterized in that: The step of determining a corresponding feature vector set according to the vibration signal comprises: Performing signal separation on the vibration signal to obtain a background sound signal and a disruptive sound signal; Time-frequency analysis is performed on the background sound signal and the destruction sound signal respectively to extract energy distribution features in the background sound signal and the destruction sound signal, and a feature vector set is generated according to the energy distribution features of the background sound signal and the energy distribution features of the destruction sound signal.

4. The cable damage early warning method according to claim 3, characterized in that: The step of performing signal separation on the vibration signal comprises: Preprocessing the vibration signal to obtain a signal to be analyzed; The signal to be analyzed is subjected to independent component analysis to separate background sound signals and destruction sound signals from the signal to be analyzed.

5. The cable damage early warning method according to claim 3, characterized in that: The step of generating a feature vector set comprises: Determine the energy value of the background sound signal in each frequency band according to the energy distribution characteristics of the background sound signal, so as to generate a first feature vector according to the energy value of the background sound signal in each frequency band; Determining the energy value of the destruction sound signal in each frequency band according to the energy distribution characteristics of the destruction sound signal, so as to generate a second feature vector according to the energy value of the destruction sound signal in each frequency band; A feature vector set is formed by the first feature vector and the second feature vector.

6. The cable damage early warning method according to any one of claims 1 to 5, characterized in that: The sound classification model is trained based on the Transformer neural network architecture.

7. The cable damage early warning method according to claim 6, characterized in that: The training process of the sound classification model includes: Obtaining a training data set and its corresponding sample data set; wherein the training data set includes single-class data samples and combined data samples; Determine the Transformer model; The Transformer model is iteratively trained based on the training data set and the sample data set to obtain a trained Transformer model, and the trained Transformer model is determined as a sound classification model.

8. A cable damage warning device, characterized in that: The device includes a wireless module, an on-site alarm module, a vibration monitoring module and a voiceprint monitoring module; The vibration monitoring module is used to collect vibration signals from the ground surface and upload the vibration signals to the soundprint monitoring module; The voiceprint monitoring module is used to determine a corresponding feature vector set according to the received vibration signal, and call a preset sound classification model, input the feature vector set into the sound classification model to obtain sound information corresponding to the feature vector set, and send the sound information to the wireless module and the on-site alarm module when the sound information belongs to the construction type; The on-site alarm module is used to trigger a warning voice when receiving the sound information; The wireless module is used to send the received sound information to each terminal in the preset terminal list.

9. The cable damage warning device according to claim 8, characterized in that: The device also includes a power management module; the power management module is used to start a low power consumption operation mode when the vibration monitoring module does not collect a vibration signal within a preset time period.

10. An early warning system, characterized in that: The system comprises a housing, an upper cover, a vibration spring and the cable damage warning device as claimed in claim 8 or 9; The cable damage warning device is placed inside the housing, and the upper cover is used to close the housing to protect the cable damage warning device; The cable damage warning device is used to execute the cable damage warning method according to any one of claims 1 to 7; The vibration spring is fixed to the bottom of the shell and is used to receive vibration signals from the ground surface.

Citation Information

Patent Citations

  • Sound signal detection system and method of distributed optical fiber sensor

    CN112033522A

  • Distributed optical fiber acoustic wave signal identification method and device, electronic equipment and storage medium

    CN117711118A

  • System for Detecting an Intrusion Attempt Inside a Perimeter Defined by a Fence

    US20140375453A1