CN113466616A - Method and device for quickly positioning cable fault point - Google Patents
Method and device for quickly positioning cable fault point Download PDFInfo
- Publication number
- CN113466616A CN113466616A CN202110694139.0A CN202110694139A CN113466616A CN 113466616 A CN113466616 A CN 113466616A CN 202110694139 A CN202110694139 A CN 202110694139A CN 113466616 A CN113466616 A CN 113466616A
- Authority
- CN
- China
- Prior art keywords
- sound
- cable fault
- parameters
- time
- noise reduction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/081—Locating faults in cables, transmission lines, or networks according to type of conductors
- G01R31/083—Locating faults in cables, transmission lines, or networks according to type of conductors in cables, e.g. underground
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Locating Faults (AREA)
Abstract
The invention provides a method and a device for quickly positioning a cable fault point, wherein the method for positioning the fault point comprises the following steps: collecting a plurality of second sound signals, extracting frequency spectrum characteristic parameters, inputting the frequency spectrum characteristic parameters into a trained sound classification model, and preliminarily judging whether the second sound signals contain cable fault sound signals; performing noise reduction processing on the second sound signal through a noise reducer, and judging whether the second sound signal subjected to the noise reduction processing contains a cable fault sound signal or not; if the second sound signal after the noise reduction processing contains a cable fault sound signal, inputting the frequency spectrum characteristic parameters in the second sound signal into a pre-constructed sound analysis model to obtain an impact function coefficient characteristic, and obtaining the distance between a cable fault sound source and a sound collection terminal based on the impact function coefficient characteristic and a pre-constructed sound source positioning model so as to obtain the position of the cable fault point.
Description
Technical Field
The invention relates to the technical field of cable fault location, in particular to a method and a device for quickly locating a cable fault point.
Background
The power cable is widely applied to urban underground power grids, internal power supply circuits of industrial and mining enterprises and underwater power transmission lines crossing rivers and sea. Once a cable breaks down, power failure loss can be caused to enterprise production, inconvenience is brought to life of residents, and therefore the cable needs to be found out as soon as possible to repair the cable.
Finding a cable fault generally needs three steps of fault diagnosis, fault location and fault location. The fault diagnosis is to use tools and equipment such as a universal meter and the like to check the connectivity of each phase of the cable and the insulation resistance value of a fault phase, so as to judge the fault property and select a proper test method for the subsequent steps; the fault location is to measure the length of the cable between a fault point and a test point by using an instrument, so as to generally determine the area where the cable fault is located and reduce the fault finding range; fault spotting is the instrumental detection of the intensity or time of arrival of a fault signal in order to gradually approximate and ultimately confirm the location of the fault. At present, in a cable fault fixed point link, a fault point is mainly searched at home and abroad by a method for detecting cable fault discharge sound. The method has two implementation modes: acoustic measurement and acoustic-magnetic synchronization. Both the acoustic measurement method and the acoustic-magnetic synchronization method need to monitor the fault sound information of the cable manually, and due to the fact that human ears are insensitive to the tiny change of sound intensity, the interference of environmental noise and the influence of factors such as the change of sound intensity caused by different cable burial depths, the monitoring difficulty is high, at present, only experienced testers can monitor manually, but the monitoring training time is long, the cost is high, the skill is difficult to train, teach and inherit from the field environment, and the development of the cable fault detection automation level is restricted.
Disclosure of Invention
The present invention aims to provide a method and a device for quickly locating a cable fault point, so as to solve the problems in the background art.
The invention is realized by the following technical scheme: the invention provides a cable fault point quick positioning method, which is characterized in that a monitoring terminal is used for quickly determining a fault section of a cable fault point, and a fault point is quickly determined in the fault section by a fault point positioning method, wherein the fault point positioning method comprises the following steps:
acquiring first acoustic signals under different conditions, and extracting frequency spectrum characteristic parameters under different conditions from the first acoustic signals, wherein the first acoustic signals under different conditions comprise first acoustic signals under a fault condition, and the first acoustic signals under the fault condition comprise cable fault acoustic signals and noise signals;
inputting the frequency spectrum characteristic parameters into a pre-constructed sound classification model for training to obtain a trained sound classification model;
in a plurality of sampling periods T, a plurality of second sound signals from a cable in a fault section are collected through a sound collection terminal, spectrum characteristic parameters in the second sound signals in each sampling period are extracted at the same time, the spectrum characteristic parameters are input into a trained sound classification model, and whether the second sound signals contain cable fault sound signals or not is preliminarily judged;
if the second sound signal contains the cable fault sound signal, performing noise reduction on the second sound signal through a noise reducer, then extracting time-frequency characteristics in the second sound signal after noise reduction, forming a first curve graph, simultaneously extracting the time-frequency characteristics in the first sound signal under a fault condition, forming a second curve graph, comparing the first curve graph with the second curve graph, and judging whether the second sound signal after noise reduction contains the cable fault sound signal, wherein the curve graphs comprise a relation curve graph of various time domain characteristic quantities and time, frequency spectrums at different time positions, a relation curve graph of sound pressure level and time, and a relation curve graph of main peak frequency and time;
if the second sound signal after the noise reduction processing contains a cable fault sound signal, inputting the frequency spectrum characteristic parameters in the second sound signal into a pre-constructed sound analysis model to obtain an impact function coefficient characteristic, and obtaining the distance between a cable fault sound source and a sound collection terminal based on the impact function coefficient characteristic and a pre-constructed sound source positioning model so as to obtain the position of the cable fault point.
Optionally, the spectral feature parameters include one or more of LPC feature parameters, MFCC feature parameters, and PLP feature parameters.
Optionally, the frequency spectrum characteristic parameters include fusion characteristic parameters obtained by fusing LPC characteristic parameters and MFCC characteristic parameters or fusion characteristic parameters obtained by fusing PLP characteristic parameters and MFCC characteristic parameters.
Optionally, constructing the sound classification model further includes: determining a topological structure of a sound classification model, wherein the topological structure comprises an input layer, an output layer and a hidden layer, and the input of the input layer comprises spectral characteristic parameters; the output of the output layer comprises sound signal classification based on spectral parameters.
Optionally, the process of constructing the acoustic analysis model includes: determining a topological structure of a sound analysis model, wherein the topological structure comprises an input layer, an output layer and a hidden layer, and the input of the input layer comprises spectral characteristic parameters; the output of the output layer comprises an impulse function coefficient characteristic;
inputting training data into the topological structure for training, and obtaining a sound analysis model after training is finished, wherein the training data comprises frequency spectrum characteristic parameters extracted from the first sound signal under different conditions.
Optionally, the process of constructing the sound source localization model includes: determining a topological structure of a sound source positioning model, wherein the topological structure comprises an input layer, an output layer and a hidden layer, and the input of the input layer comprises an impact function coefficient characteristic; the output of the output layer includes a distance between a sound source and the sound collection terminal.
Optionally, the noise reducer includes different noise reduction algorithms, one of the noise reduction algorithms is used to perform noise reduction processing on the second acoustic signal, and if the second acoustic signal after the noise reduction processing does not include a cable fault acoustic signal, other noise reduction algorithms are used to perform noise reduction processing on the second acoustic signal.
The second aspect of the present invention provides a cable fault point fast positioning device, including:
the historical data module stores first acoustic signals under different conditions;
the frequency spectrum feature extraction module is used for extracting frequency spectrum features in the acoustic signals;
the time-frequency characteristic analysis module is used for extracting time-frequency characteristics in the acoustic signals and forming corresponding graphs, wherein the graphs comprise graphs of relations between various time-domain characteristic quantities and time, graphs of relations between frequency spectrums at different time positions, between sound pressure levels and time, and graphs of relations between main peak frequencies and time;
the sound classification module is used for generating a sound classification model and inputting training data into the sound classification model for training and learning so as to obtain a sound signal classification condition based on the frequency spectrum parameters;
the sound analysis module is used for generating a sound analysis model and inputting training data into the sound classification model for training and learning so as to obtain the characteristics of the impact function coefficient;
and the sound source positioning module is used for generating a sound source positioning model and inputting training data into the sound source positioning model for training and learning so as to obtain the distance between the sound source and the sound acquisition terminal.
Compared with the prior art, the invention has the following beneficial effects:
the invention provides a method and a device for quickly positioning a cable fault point, which are used for judging the distance by adopting a machine learning method based on the acoustic characteristics of sound signals, particularly extracting the frequency spectrum characteristics of the sound signals near a cable acquired on site, judging whether the sound acquired on the site contains cable fault sound by utilizing a pre-constructed sound classification model according to the frequency spectrum characteristics, comparing by utilizing a curve chart formed by time-frequency characteristics, judging whether the sound signals subjected to noise reduction contain the cable fault sound, obtaining the impact function coefficient characteristics by utilizing a pre-constructed sound analysis model according to the frequency spectrum characteristics, and finally determining the distance information between a cable fault sound source and a sound acquisition terminal by utilizing a sound source positioning model, so that the distance measurement is carried out without a third-party sensor, the cost is reduced, and the problem that a plurality of microphones on the traditional sound acquisition terminal are limited by microphone array hardware layout concentration is solved Inaccurate distance measurement.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only preferred embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive exercise.
Fig. 1 is a flowchart of a method for quickly locating a cable fault point according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, exemplary embodiments according to the present invention will be described in detail below with reference to the accompanying drawings. It is to be understood that the described embodiments are merely a subset of embodiments of the invention and not all embodiments of the invention, with the understanding that the invention is not limited to the example embodiments described herein. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the invention described herein without inventive step, shall fall within the scope of protection of the invention.
In the following description, numerous specific details are set forth in order to provide a more thorough understanding of the present invention. It will be apparent, however, to one skilled in the art, that the present invention may be practiced without one or more of these specific details. In other instances, well-known features have not been described in order to avoid obscuring the invention.
It is to be understood that the present invention may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. As used herein, the term "and/or" includes any and all combinations of the associated listed items.
In order to provide a thorough understanding of the present invention, a detailed structure will be set forth in the following description in order to explain the present invention. Alternative embodiments of the invention are described in detail below, however, the invention may be practiced in other embodiments that depart from these specific details.
Referring to fig. 1, a first aspect of the present invention provides a method for quickly locating a cable fault point, where a monitoring terminal is used to quickly determine a fault section of the cable fault point, the monitoring terminals are arranged on a power cable at intervals, and each monitoring terminal has a fault traveling wave collecting capability, and is capable of collecting fault traveling wave information sent by the cable fault point and recording the time for collecting the fault traveling wave, a background server is used to determine a fault section range of the cable fault point by a double-end traveling wave locating method, and then a fault point is quickly determined in the fault section by a fault point locating method, where the fault point locating method includes the following steps:
step S1, collecting first acoustic signals under different conditions, and extracting frequency spectrum characteristic parameters under different conditions from the first acoustic signals, wherein the first acoustic signals under different conditions comprise first acoustic signals under fault conditions, and the first acoustic signals under fault conditions comprise cable fault acoustic signals and noise signals;
step S2, inputting the frequency spectrum characteristic parameters into a pre-constructed sound classification model for training to obtain a trained sound classification model;
step S3, in a plurality of sampling periods T, a plurality of second sound signals from the cable in the fault section are collected through a sound collection terminal, meanwhile, the frequency spectrum characteristic parameters in the second sound signals in each sampling period are extracted, the frequency spectrum characteristic parameters are input into a trained sound classification model, and whether the second sound signals contain cable fault sound signals or not is preliminarily judged;
step S4, if the second sound signal contains the cable fault sound signal, the noise reducer is used for carrying out noise reduction on the second sound signal, then the time-frequency characteristic in the second sound signal after noise reduction is extracted, a first curve graph is formed, meanwhile, the time-frequency characteristic in the first sound signal under the fault condition is extracted, a second curve graph is formed, the first curve graph is compared with the second curve graph, whether the second sound signal after noise reduction contains the cable fault sound signal or not is judged, wherein the curve graphs comprise the relation curve graphs of various time domain characteristic quantities and time, the frequency spectrums of different time positions, the relation curve graphs of sound pressure levels and time, and the relation curve graphs of main peak frequency and time;
step S5, if the second sound signal after the noise reduction processing contains a cable fault sound signal, inputting the frequency spectrum characteristic parameter in the second sound signal into a sound analysis model which is constructed in advance to obtain an impact function coefficient characteristic, and obtaining the distance between a cable fault sound source and a sound collection terminal based on the impact function coefficient characteristic and a sound source positioning model which is constructed in advance so as to obtain the cable fault point position.
In some embodiments of step S1, the spectral feature parameters extracted from the first acoustic signal, whose first acoustic signal is derived from pre-collected noisy cable fault sounds of different fault conditions in real environment or artificially synthesized noisy cable fault sounds, for example by adding noise to the fault sounds with different signal-to-noise ratios, include one or more of LPC feature parameters, MFCC feature parameters, PLP feature parameters.
In specific implementation, taking PLP feature parameter extraction as an example, the PLP feature parameter extraction steps are as follows:
carrying out short-time Fourier transform on the first sound signal S (n) to obtain a frequency spectrum x (omega), and calculating a power spectrum p (omega) of the frequency spectrum x (omega);
critical band analysis is performed on the power spectrum p (ω):
Ω(ω)＝6ln{ω/1200π+[(ω/1200π)2+1]1/2}
where ω is the angular frequency and Ω is the Bark domain frequency.
Performing convolution operation on the power spectrum p (omega) and the simulated hearing frequency critical curve psi (omega) to obtain a critical band power spectrum theta omegai]Wherein the formula of the critical band curve is as follows:
carrying out equal loudness curve pre-emphasis processing;
performing intensity-loudness conversion;
and performing inverse Fourier transform to obtain a signal short-time autocorrelation function R (tau), obtaining a 12-order linear prediction coefficient according to a Durbin formula, and performing spectral analysis on the 12-order linear prediction coefficient to obtain 12-dimensional PLP characteristic parameters and first-order difference characteristic parameters thereof, namely 24-dimensional PLP characteristic parameters.
In addition, it should be noted that, extraction of LPC characteristic parameters and MFCC characteristic parameters is performed by the conventional method, and this embodiment is not described in detail again
In some embodiments of step S1, the extracting the spectral feature parameters from the first acoustic signal includes fused feature parameters obtained by fusing LPC feature parameters and MFCC feature parameters or fused feature parameters obtained by fusing PLP feature parameters and MFCC feature parameters.
In specific implementation, for example, after the PLP characteristic parameter and the MFCC characteristic parameter are fused, the fusion specifically includes:
setting a characteristic vector V, and performing serial fusion on the PLP characteristic parameters and the MFCC characteristic parameters:
V＝(αβ)
in the formula, α is a PLP characteristic parameter, and β is an MFCC characteristic parameter.
In some embodiments of step S2, the constructed sound classification model may use one or more of a deep confidence neural network, a feed-forward neural network, a convolutional neural network, or a cyclic neural network, and the present embodiment is preferably described by taking the deep confidence network as an example.
When a sound classification model is constructed through a deep belief neural network, firstly, determining a topological structure of the sound classification model, wherein the topological structure comprises an input layer, an output layer and a hidden layer, and the input of the input layer comprises spectrum characteristic parameters; the output of the output layer includes the sound signal classification based on the spectrum parameters, and the hidden layer is generally 3 to 8 layers, and the present embodiment is 4 layers.
It can be understood that, because the cable fault sound has different expression forms under different fault types, the sound signal classification results of different faults can be obtained after the spectrum parameters are input into the sound classification model for continuous training.
And the training process of the deep confidence network comprises the following two steps: pre-training and fine-tuning.
(1) And (4) pre-training. And in the RBM parameter learning process, the whole pre-training process is unsupervised. And each RBM for training is a new representation of input data, the input data is assumed to be a vector v, an output vector obtained after activation of an RBM hidden layer unit is assumed to be h, the output vector h is used as input training data of the next RBM to train a new RBM, and the obtained weight of each RBM can independently extract features from the output of the previous layer.
(2) And (5) fine adjustment process. And (3) taking the RBM pre-training weight parameter as an initial parameter of the network, and carrying out supervised fine adjustment on all weights through random gradient descent (SGD) until the training is finished.
In some embodiments of step S3, the sound collecting terminal is a microphone array collecting terminal, and the collecting terminal may be a handheld collecting terminal, and the operator collects sound signals from one side of the cable fault section through the handheld collecting terminal, and can collect multiple sets of sound signals through the microphone array on the collecting terminal in a period T, and optionally, at least one sound signal sound is collected in each period T, and the collected sound signals are used as second sound signals, and the spectral feature parameter extraction is performed by the spectral feature parameter extraction method in step S1, and the extracted spectral feature parameters are input into the sound classification model to obtain a sound classification result, so as to further determine whether the second sound signals contain cable fault sounds.
In some embodiments of step S4, the noise reducer includes different noise reduction algorithms, where the noise reduction algorithms are currently commonly used acoustic noise reduction algorithms, one of the noise reduction algorithms is used to perform noise reduction on the second acoustic signal, then time-frequency features in the noise-reduced second acoustic signal are extracted, a first graph is formed, time-frequency features in the first acoustic signal under a fault condition are extracted, a second graph is formed, the first graph is compared with the second graph, it is determined whether all of the noise-reduced second acoustic signal include cable fault acoustic signals, if a part of the noise-reduced second acoustic signal includes cable fault acoustic signals, it is indicated that an abnormality occurs in the noise reduction process of the second acoustic signal, and then other noise reduction algorithms are used to perform noise reduction on the second acoustic signal.
In some embodiments of step S5, the constructed sound analysis model may use one or more of a deep confidence neural network, a feed-forward neural network, a convolutional neural network, or a cyclic neural network, and the embodiment is preferably described by taking the deep confidence network as an example.
The process of constructing the sound analysis model comprises the following steps: determining a topological structure of a sound analysis model, wherein the topological structure comprises an input layer, an output layer and a hidden layer, and the input of the input layer comprises spectral characteristic parameters; the output of the output layer comprises an impulse function coefficient characteristic;
training data including the spectral feature parameters extracted from the first acoustic signal under different conditions is input into the topology for training, and a sound analysis model is obtained after the training is finished, and the training method may not be described in detail with reference to the training method in step S1.
And inputting the extracted spectral characteristic parameters of the second acoustic signal into the trained acoustic analysis model to obtain the impact function coefficient characteristics of the second acoustic signal.
It should be noted that, the coefficient characteristics of the impulse function are usually expressed in a frequency domain, the specific dimension is related to the frequency domain resolution, if the audio frequency is 16K sampling rate, due to the symmetry of the frequency domain expression, 8K sampling points are usually selected, that is, the reverberation impulse function is expressed by a vector of 8K dimension; if the audio is at 8K sample rate, 4K samples are typically selected. Taking the sampling rate of 16K as an example, the characteristics of the coefficient of the reverberation impulse function are represented by vector with 8000 dimensions.
In other embodiments of step S5, the sound source localization model may be constructed by using one or more combinations of a deep confidence neural network, a feed-forward neural network, a convolutional neural network, or a cyclic neural network, and the embodiment is preferably described by taking the deep confidence network as an example.
The process of constructing the sound source positioning model comprises the following steps: determining a topological structure of a sound source positioning model, wherein the topological structure comprises an input layer, an output layer and a hidden layer, and the input of the input layer comprises an impact function coefficient characteristic; the output of the output layer includes a distance between a sound source and the sound collection terminal. The training data includes the characteristics of the coefficients of the impulse function at different distances from the sound source to the sound collecting terminal, and the training method may be described in detail with reference to the training method in step S1.
Inputting the impact function coefficient characteristics of the second acoustic signal into the trained sound source positioning model, so that the distance information between the fault point sound source and the acquisition terminal can be obtained, wherein the distance information can be a determined distance value, and the sound source positioning model is a regression model at the moment, so that the number of nodes of the output layer can be 1, and the determined distance value is represented.
The second aspect of the present invention provides a cable fault point fast positioning device, including:
the historical data module stores first acoustic signals under different conditions;
the frequency spectrum feature extraction module is used for extracting frequency spectrum features in the acoustic signals;
the time-frequency characteristic analysis module is used for extracting time-frequency characteristics in the acoustic signals and forming corresponding graphs, wherein the graphs comprise graphs of relations between various time-domain characteristic quantities and time, graphs of relations between frequency spectrums at different time positions, between sound pressure levels and time, and graphs of relations between main peak frequencies and time;
the sound classification module is used for generating a sound classification model and inputting training data into the sound classification model for training and learning so as to obtain a sound signal classification condition based on the frequency spectrum parameters;
the sound analysis module is used for generating a sound analysis model and inputting training data into the sound classification model for training and learning so as to obtain the characteristics of the impact function coefficient;
and the sound source positioning module is used for generating a sound source positioning model and inputting training data into the sound source positioning model for training and learning so as to obtain the distance between the sound source and the sound acquisition terminal.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the apparatus embodiment, since it is substantially similar to the method embodiment, it is relatively simple to describe, and reference may be made to some descriptions of the method embodiment for relevant points. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (8)
1. A cable fault point quick positioning method is characterized in that a fault section of a cable fault point is quickly determined through a monitoring terminal, and a fault point is quickly determined in the fault section through a fault point positioning method, wherein the fault point positioning method comprises the following steps:
acquiring first acoustic signals under different conditions, and extracting frequency spectrum characteristic parameters under different conditions from the first acoustic signals, wherein the first acoustic signals under different conditions comprise first acoustic signals under a fault condition, and the first acoustic signals under the fault condition comprise cable fault acoustic signals and noise signals;
inputting the frequency spectrum characteristic parameters into a pre-constructed sound classification model for training to obtain a trained sound classification model;
in a plurality of sampling periods T, a plurality of second sound signals from a cable in a fault section are collected through a sound collection terminal, spectrum characteristic parameters in the second sound signals in each sampling period are extracted at the same time, the spectrum characteristic parameters are input into a trained sound classification model, and whether the second sound signals contain cable fault sound signals or not is preliminarily judged;
if the second sound signal contains the cable fault sound signal, performing noise reduction on the second sound signal through a noise reducer, then extracting time-frequency characteristics in the second sound signal after noise reduction, forming a first curve graph, simultaneously extracting the time-frequency characteristics in the first sound signal under a fault condition, forming a second curve graph, comparing the first curve graph with the second curve graph, and judging whether the second sound signal after noise reduction contains the cable fault sound signal, wherein the curve graphs comprise a relation curve graph of various time domain characteristic quantities and time, frequency spectrums at different time positions, a relation curve graph of sound pressure level and time, and a relation curve graph of main peak frequency and time;
if the second sound signal after the noise reduction processing contains a cable fault sound signal, inputting the frequency spectrum characteristic parameters in the second sound signal into a pre-constructed sound analysis model to obtain an impact function coefficient characteristic, and obtaining the distance between a cable fault sound source and a sound collection terminal based on the impact function coefficient characteristic and a pre-constructed sound source positioning model so as to obtain the position of the cable fault point.
2. The method as claimed in claim 1, wherein the spectral characteristic parameters include one or more of LPC characteristic parameters, MFCC characteristic parameters, and PLP characteristic parameters.
3. The method as claimed in claim 2, wherein the spectral feature parameters include fused feature parameters obtained by fusing LPC feature parameters and MFCC feature parameters or fused feature parameters obtained by fusing PLP feature parameters and MFCC feature parameters.
4. The method for rapidly positioning a cable fault point according to claim 2 or 3, wherein constructing the sound classification model further comprises: determining a topological structure of a sound classification model, wherein the topological structure comprises an input layer, an output layer and a hidden layer, and the input of the input layer comprises spectral characteristic parameters; the output of the output layer comprises sound signal classification based on spectral parameters.
5. The method for rapidly positioning the cable fault point according to claim 1, wherein the process of constructing the acoustic analysis model comprises the following steps: determining a topological structure of a sound analysis model, wherein the topological structure comprises an input layer, an output layer and a hidden layer, and the input of the input layer comprises parameters of spectral features; the output of the output layer comprises an impulse function coefficient characteristic;
inputting training data into the topological structure for training, and obtaining a sound analysis model after training is finished, wherein the training data comprises frequency spectrum characteristic parameters extracted from the first sound signal under different conditions.
6. The method for rapidly positioning the cable fault point according to claim 1, wherein the process of constructing the sound source positioning model comprises the following steps: determining a topological structure of a sound source positioning model, wherein the topological structure comprises an input layer, an output layer and a hidden layer, and the input of the input layer comprises an impact function coefficient characteristic; the output of the output layer includes a distance between a sound source and the sound collection terminal.
7. The method according to claim 1, wherein the noise reducer comprises different noise reduction algorithms, one of the noise reduction algorithms is used to perform noise reduction processing on the second acoustic signal, and if the noise-reduced second acoustic signal does not comprise a cable fault acoustic signal, other noise reduction algorithms are used to perform noise reduction processing on the second acoustic signal.
8. A cable fault point quick locating device is characterized by comprising:
the historical data module stores first acoustic signals under different conditions;
the spectral feature extraction module is used for extracting spectral features in the acoustic signals;
the time-frequency characteristic analysis module is used for extracting time-frequency characteristics in the acoustic signals and forming corresponding graphs, wherein the graphs comprise graphs of relations between various time-domain characteristic quantities and time, graphs of relations between frequency spectrums at different time positions, between sound pressure levels and time, and graphs of relations between main peak frequencies and time;
the sound classification module is used for generating a sound classification model and inputting training data into the sound classification model for training and learning so as to obtain a sound signal classification condition based on the frequency spectrum parameters;
the sound analysis module is used for generating a sound analysis model and inputting training data into the sound classification model for training and learning so as to obtain the characteristics of the impact function coefficient;
and the sound source positioning module is used for generating a sound source positioning model and inputting training data into the sound source positioning model for training and learning so as to obtain the distance between the sound source and the sound acquisition terminal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110694139.0A CN113466616A (en) | 2021-06-22 | 2021-06-22 | Method and device for quickly positioning cable fault point |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110694139.0A CN113466616A (en) | 2021-06-22 | 2021-06-22 | Method and device for quickly positioning cable fault point |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113466616A true CN113466616A (en) | 2021-10-01 |
Family
ID=77869368
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110694139.0A Pending CN113466616A (en) | 2021-06-22 | 2021-06-22 | Method and device for quickly positioning cable fault point |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113466616A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115267331A (en) * | 2022-09-29 | 2022-11-01 | 天津安力信通讯科技有限公司 | Signal orientation positioning method and system based on multi-spectral analysis |
| CN115860714A (en) * | 2022-11-28 | 2023-03-28 | 珠海德瑞斯科技有限公司 | Power equipment safe operation management system and method based on industrial Internet |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107507625A (en) * | 2016-06-14 | 2017-12-22 | 讯飞智元信息科技有限公司 | Sound source distance determines method and device |
| CN109116196A (en) * | 2018-07-06 | 2019-01-01 | 山东科汇电力自动化股份有限公司 | A kind of power cable fault discharging sound intelligent identification Method |
| CN109357749A (en) * | 2018-09-04 | 2019-02-19 | 南京理工大学 | A kind of power equipment audio signal analysis method based on DNN algorithm |
| CN111157855A (en) * | 2019-12-31 | 2020-05-15 | 山东信通电子股份有限公司 | Method for judging transmission line fault and server |
| CN111540347A (en) * | 2020-05-12 | 2020-08-14 | 山东科华电力技术有限公司 | Cable tunnel monitoring method and system based on audio |
| CN113707176A (en) * | 2021-09-02 | 2021-11-26 | 国网安徽省电力有限公司铜陵供电公司 | Transformer fault detection method based on acoustic signal and deep learning technology |
-
2021
- 2021-06-22 CN CN202110694139.0A patent/CN113466616A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107507625A (en) * | 2016-06-14 | 2017-12-22 | 讯飞智元信息科技有限公司 | Sound source distance determines method and device |
| CN109116196A (en) * | 2018-07-06 | 2019-01-01 | 山东科汇电力自动化股份有限公司 | A kind of power cable fault discharging sound intelligent identification Method |
| CN109357749A (en) * | 2018-09-04 | 2019-02-19 | 南京理工大学 | A kind of power equipment audio signal analysis method based on DNN algorithm |
| CN111157855A (en) * | 2019-12-31 | 2020-05-15 | 山东信通电子股份有限公司 | Method for judging transmission line fault and server |
| CN111540347A (en) * | 2020-05-12 | 2020-08-14 | 山东科华电力技术有限公司 | Cable tunnel monitoring method and system based on audio |
| CN113707176A (en) * | 2021-09-02 | 2021-11-26 | 国网安徽省电力有限公司铜陵供电公司 | Transformer fault detection method based on acoustic signal and deep learning technology |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115267331A (en) * | 2022-09-29 | 2022-11-01 | 天津安力信通讯科技有限公司 | Signal orientation positioning method and system based on multi-spectral analysis |
| CN115267331B (en) * | 2022-09-29 | 2022-12-16 | 天津安力信通讯科技有限公司 | Signal orientation positioning method and system based on multi-spectral analysis |
| CN115860714A (en) * | 2022-11-28 | 2023-03-28 | 珠海德瑞斯科技有限公司 | Power equipment safe operation management system and method based on industrial Internet |
| CN115860714B (en) * | 2022-11-28 | 2023-08-08 | 珠海德瑞斯科技有限公司 | Power equipment safe operation management system and method based on industrial Internet |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106408088B (en) | A kind of rotating machinery method for diagnosing faults based on deep learning theory | |
| CN109033612B (en) | Transformer fault diagnosis method based on vibration noise and BP neural network | |
| CN106323452B (en) | A kind of detection method and detection device of equipment abnormal sound | |
| CN106546892A (en) | The recognition methodss of shelf depreciation ultrasonic audio and system based on deep learning | |
| CN113466616A (en) | Method and device for quickly positioning cable fault point | |
| CN102563361A (en) | Device and method for detecting and positioning leakage of gas transmission pipeline based on conventional data and sound wave signals | |
| CN109409308A (en) | A method of the birds species identification based on birdvocalization | |
| CN109033226B (en) | Vibration sensor arrangement optimization method based on complex network theory | |
| CN111980676B (en) | Method and processing device for evaluating well cementation quality by array acoustic logging | |
| CN110286155B (en) | Damage detection method and system for multilayer composite material | |
| CN107367676A (en) | Partial discharge indicator based on audio intelligent identification | |
| WO2021033501A1 (en) | Seismic observation device, seismic observation method, and recording medium on which seismic observation program is recorded | |
| CN116778956A (en) | Transformer acoustic feature extraction and fault identification method | |
| CN112052712B (en) | Power equipment state monitoring and fault identification method and system | |
| CN105352726A (en) | Fault diagnosis method for gear | |
| CN117349597B (en) | Intelligent background noise reduction method for intelligent digital accurate pointing instrument | |
| CN104568137B (en) | The decision method of specific noise source contribution rate in noise testing | |
| Wang et al. | Identification of the defective transmission devices using the wavelet transform | |
| CN107741263B (en) | Liquid level estimation method based on frequency domain information fusion | |
| CN107271184B (en) | A kind of the kernel regression decomposition method and system of rolling bearing fault diagnosis | |
| Wang et al. | Research on fault diagnosis method of wind turbine bearing based on deep belief network | |
| TWI793626B (en) | Acoustic diagnostic method for detecting structural degradation and system applying the same | |
| Darraz et al. | Rotating machine diagnosis using acoustic imaging and artificial intelligence | |
| CN105889052A (en) | Noise separation method for fault diagnosis of slurry pump | |
| CN101526417A (en) | Method for diagnosing structural damage by musical tone rule |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20211001 |