CN115762529A - Method for preventing cable from being broken outside by using voice recognition perception algorithm - Google Patents
Method for preventing cable from being broken outside by using voice recognition perception algorithm Download PDFInfo
- Publication number
- CN115762529A CN115762529A CN202211267984.0A CN202211267984A CN115762529A CN 115762529 A CN115762529 A CN 115762529A CN 202211267984 A CN202211267984 A CN 202211267984A CN 115762529 A CN115762529 A CN 115762529A
- Authority
- CN
- China
- Prior art keywords
- sound
- mfcc
- voiceprint
- sound sensor
- data
- 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
- 238000000034 method Methods 0.000 title claims abstract description 23
- 230000008447 perception Effects 0.000 title claims abstract description 9
- 238000010276 construction Methods 0.000 claims abstract description 21
- 230000003068 static effect Effects 0.000 claims abstract description 6
- 230000005236 sound signal Effects 0.000 claims description 13
- 238000000605 extraction Methods 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000009432 framing Methods 0.000 claims description 3
- 230000003449 preventive effect Effects 0.000 claims description 2
- 239000003990 capacitor Substances 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 239000000779 smoke Substances 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
Images
Landscapes
- Burglar Alarm Systems (AREA)
Abstract
The invention relates to a method for preventing cable external damage by using a voice recognition perception algorithm, which determines the distance between construction machinery and a buried cable line through target positioning and determines the type of the construction machinery through target recognition; the purpose of reminding constructors and distribution network personnel is achieved, and corresponding measures are taken to prevent the cable from being broken outside. In the target positioning, three sound sensors are arranged along an underground cable line, namely points A, O and B, and each point is positioned on a straight line, the sensing range threshold of the sound sensor is set to be SO, and when an external force damage source enters the sensing range of the sound sensor, the sound sensor outputs an alarm starting signal outwards; in the target recognition, static characteristics of the MFCC audio are converted into newMFCC characteristic data, so that the characteristics of the audio are dynamically represented, and the accuracy is improved.
Description
Technical Field
The invention relates to the technical field of cable protection, in particular to a method for preventing a cable from being broken by using a voice recognition perception algorithm.
Background
The buried cable of the distribution network line is a common form, the implementation of regional infrastructure is driven along with the development of economy, a large amount of engineering machinery is applied to the construction of specific engineering, in the process, due to the fact that constructors and distribution network personnel are not smooth in communication and other reasons, the damaged buried cable is damaged in a construction region, the damage happens occasionally, the loss of the damaged buried cable is self-evident, and therefore the prevention work needs to be done by the constructors and the distribution network personnel together. In the prior art, one of the methods is that a distribution network worker adds a sensing device at the position of an underground cable to feed abnormal information back to a distribution network middle station to be correspondingly processed, for example, CN213987749U discloses a smoke alarm device for preventing the cable from being broken, which adopts a vibration sensor to give an alarm in a smoke and sound mode after sensing vibration to remind a constructor; for example, CN112581730A discloses a warning method for preventing external damage of cable and a warning device for preventing external damage of cable, wherein a protection tube with optical fiber inside is laid in parallel above the buried cable, and two ends of the protection tube are provided with photoelectric switches to sense when external force is damaged and alarm; however, the above-mentioned technology needs to sense when the location is damaged, and there is a possibility that the buried cable is damaged because the construction machine is not stopped in time.
Disclosure of Invention
The invention provides a method for preventing a cable from being broken by using a voice recognition perception algorithm, which aims at the defects of the background art, when a construction machine enters an underground cable area, an alarm is sent to constructors and/or distribution network personnel, the underground cable in the area is warned before construction, and the possibility of preventing the cable from being damaged in advance is realized; the technical scheme adopted by the invention is as follows:
a method for preventing cable external damage by using a voice recognition perception algorithm collects data of construction machinery, executes two steps of target positioning and target recognition, and further reminds constructors and/or distribution network personnel to take preventive measures, wherein:
the target positioning is realized by arranging a plurality of sound sensors on an underground cable line and positioning the construction machinery by utilizing a time delay estimation calculation method; the method comprises the following steps:
arranging three sound sensors along a buried cable line, wherein the three sound sensors are respectively a point A, a point O and a point B, and each point is positioned on a straight line, wherein the distance AO = OB between two adjacent sound sensors is marked as d; the external force failure source is point S;
setting the sensing range threshold of the sound sensor as SO, and outputting an alarm starting signal to the outside by the sound sensor when an external force damage source enters the sensing range of the sound sensor;
wherein SA, SO, SB are calculated by formulas 1 to 3:
in the formula: v is the speed of sound;
receiving the time delay difference of the sound sensor for A and O;
receiving the time delay difference of the sound source signals for B and O;
the target recognition is to acquire various typical mechanical voiceprint data, analyze and extract various signal characteristics, perform model training to obtain a model base, and when a sound sensor receives sound data, recognize the data and compare the data with the model base to realize voiceprint recognition of different machines; the method comprises the following steps:
collecting voiceprint data including various typical construction machine types; preprocessing by a windowing and framing method to remove background noise in the image; establishing a model base by an MFCC feature extraction method; after the sound sensor receives the voiceprint to be detected, preprocessing and MFCC feature extraction are carried out to obtain extracted voiceprint, the extracted voiceprint is compared with the model library to obtain a recognition result, and therefore a certain type of machine name is output.
Further, the MFCC feature extraction method is used for processing voiceprint data and voiceprint data to be measured, and is to process the acquired audio signal, that is, to: firstly, preprocessing collected audio signals, then obtaining frequency domain signals (DFT) from the preprocessed time domain signals through Fourier transformation, then processing the DFT through Mel (Mel) filter bank, logarithmic energy and Discrete Cosine Transform (DCT), finally converting the audio signals into Mel cepstrum domain, and further respectively obtaining first order differential coefficients delta MFCC and second order differential coefficients delta 2 The MFCC is weighted and then is combined with the MFCC to obtain newMFCC characteristic data; wherein the newMFCC is calculated by equation 4:
in the formula: a and b are weights, and 0<b<a<1, taking 1/3 of a, and taking 1/6 of b; MFCC is a static characteristic of audio; Δ MFCC is a dynamic property; delta 2 MFCC is a balance factorAnd (5) performing secondary treatment.
Furthermore, the sound sensors all adopt the same specification, and the distance between two adjacent sound sensors is a linear distance.
Further, the sound sensor adopts a condenser microphone.
The beneficial effects of the implementation of the invention are as follows: the distance between the construction machine and the underground cable line is determined through target positioning, and the type of the construction machine is determined through target identification; the purpose of reminding constructors and distribution network personnel is achieved, and corresponding measures are taken to prevent the cable from being broken. In the target positioning, three sound sensors are arranged along an underground cable line, namely points A, O and B, the points are positioned on a straight line, the sensing range threshold of the sound sensor is set to be SO, and when an external force damage source enters the sensing range of the sound sensor, the sound sensor outputs an alarm starting signal outwards; in the target identification, the static characteristics of the MFCC audio are converted into newMFCC characteristic data, so that the characteristics of dynamically representing the audio are achieved, and the accuracy is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the prior art descriptions will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1: a flow chart;
FIG. 2 is a schematic diagram: a target location schematic;
FIG. 3: a target identification schematic diagram;
FIG. 4: MFCC is converted to a newMFCC schematic.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention provides a method for preventing cable external damage by using a voice recognition perception algorithm, which is shown in the attached drawings 1 to 3:
comprises two steps of target positioning and target identification,
the target positioning is realized by arranging a plurality of sound sensors on an underground cable line and positioning the construction machinery by utilizing a time delay estimation calculation method; the target location is:
arranging three sound sensors along a buried cable line, wherein the three sound sensors are respectively a point A, a point O and a point B, and each point is positioned on a straight line, wherein the distance AO = OB between two adjacent sound sensors is marked as d; the external force failure source is point S;
setting a sensing range threshold value of the sound sensor as SO, and outputting an alarm starting signal to the outside by the sound sensor when an external force damage source enters the sensing range of the sound sensor;
wherein SA, SO, SB are calculated by formulas 1 to 3:
in the formula: v is the speed of sound;
receiving the time delay difference of the sound sensors for A and O;
the delay difference of the sound source signals received for B and O.
The sound sensors are all in the same specification, and the distance between every two adjacent sound sensors is a linear distance;
the sound sensor adopts a capacitor microphone; the principle of the condenser microphone is that a vibrating film is made into a capacitor, one polar plate of the capacitor is fixed, and the other polar plate generates displacement according to the vibration of sound, so that the capacity of the film capacitor is changed, and the condenser microphone is more suitable for collecting remote sound.
The target recognition is to acquire various typical mechanical voiceprint data, analyze and extract various signal characteristics, perform model training to obtain a model base, and when a sound sensor receives the sound data, recognize the data and compare the data with the model base to realize voiceprint recognition of different machines; the target identification is:
collecting voiceprint data including various typical construction machine types; preprocessing by a windowing and framing method to remove background noise in the image; establishing a model base by an MFCC feature extraction method; after receiving the voiceprint to be detected, the sound sensor performs preprocessing and MFCC feature extraction to obtain an extracted voiceprint, and the extracted voiceprint is compared with the model library to obtain a recognition result so as to output a certain type of mechanical name.
The MFCC feature extraction method is used for processing voiceprint data and to-be-detected voiceprint data, and is used for processing an acquired audio signal, namely: firstly, preprocessing collected audio signals, then obtaining frequency domain signals (DFT) from the preprocessed time domain signals through Fourier transformation, then processing the DFT through Mel (Mel) filter bank, logarithmic energy and Discrete Cosine Transform (DCT), finally converting the audio signals into Mel cepstrum domain, and further respectively obtaining first order differential coefficients delta MFCC and second order differential coefficients delta 2 The MFCC is weighted and then is combined with the MFCC to obtain newMFCC characteristic data; wherein the newMFCC is calculated by equation 4:
in the formula: a and b are weights, and 0<b<a<1, taking 1/3 of a, and 1/6 of b; MFCC is a static characteristic of audio; Δ MFCC is a dynamic property; delta 2 MFCC is the balance factor.
Since MFCCs can only characterize audio static properties, they cannot dynamically characterize audio. Thus, the newMFCC is a MFCC-based weighted dynamic feature parameter, i.e. the MFCC parameter and the weighted first and second order difference coefficients are combined into a new vector newMFCC as the feature parameter of the audio signal.
Firstly, training a UBM model by using an expectation maximization algorithm EM (expectation maximization algorithm), namely training a Gaussian mixture model GMM (Gaussian mixture model) by using a training set containing a large amount of sound signal data; then, the mean value is re-estimated by using MLE or MAP by using the sound signal data of the voiceprint to be measured, and a new target GMM model, namely a GMM-UBM model is obtained.
The invention collects the audio signal generated by certain construction machinery entering the sensing range through a plurality of sound sensors arranged on the buried cable line, and outputs an alarm starting signal to the outside to send out an alarm to remind constructors and/or distribution network personnel; determining the distance between the construction machine and the underground cable line through target positioning, namely arranging a plurality of sound sensors on the underground cable line, positioning the construction machine by using a time delay estimation calculation method, setting a sensing range threshold of the sound sensors as SO, and outputting an alarm starting signal to the outside by the sound sensors when an external force damage source enters the sensing range of the sound sensors; the type of the construction machinery is determined through target recognition, namely, various typical mechanical voiceprint data are collected, signal characteristics are analyzed and extracted, model training is carried out to obtain a model base, and when a sound sensor receives the sound data, the data are recognized and compared with the model base to realize voiceprint recognition of different machinery; the purpose of reminding constructors and distribution network personnel is achieved, and corresponding measures are taken to prevent the cable from being broken outside.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and should not be taken as limiting the scope of the present invention, which is intended to cover any modifications, equivalents, improvements, etc. within the spirit and scope of the present invention.
Claims (4)
1. A method for preventing cable external damage by using a voice recognition perception algorithm collects data of construction machinery, and executes two steps of target positioning and target recognition, so as to remind constructors and/or distribution network personnel to take preventive measures; the method is characterized in that a plurality of sound sensors are arranged on an underground cable line for positioning the target, and a time delay estimation calculation method is used for positioning the construction machinery, and the method comprises the following steps:
arranging three sound sensors along a buried cable line, wherein the three sound sensors are respectively a point A, a point O and a point B, and each point is positioned on a straight line, wherein the distance AO = OB between two adjacent sound sensors is marked as d; the external force failure source is point S;
setting the sensing range threshold of the sound sensor as SO, and outputting an alarm starting signal to the outside by the sound sensor when an external force damage source enters the sensing range of the sound sensor;
wherein SA, SO, SB are calculated by formulas 1 to 3:
in the formula: v is the speed of sound;
receiving the time delay difference of the sound sensor for A and O;
receiving the time delay difference of the sound source signals for B and O;
the target recognition is to acquire various typical mechanical voiceprint data, analyze and extract various signal characteristics, perform model training to obtain a model base, and when a sound sensor receives the sound data, recognize the data and compare the data with the model base to realize voiceprint recognition of different machines, and the method comprises the following steps:
acquiring voiceprint data including various typical construction machine types; preprocessing by a windowing and framing method to remove background noise in the image; establishing a model base by an MFCC feature extraction method; after receiving the voiceprint to be detected, the sound sensor performs preprocessing and MFCC feature extraction to obtain an extracted voiceprint, and the extracted voiceprint is compared with the model library to obtain a recognition result so as to output a certain type of mechanical name.
2. The method for preventing cable break-out by using voice recognition perception algorithm as claimed in claim 1, wherein said MFCC feature extraction method is used for processing the voiceprint data and the voiceprint under test, and the acquired audio signal is processed by: firstly, preprocessing the collected audio signal, then obtaining a frequency domain signal DFT through Fourier change of the preprocessed time domain signal, then processing the frequency domain signal DFT through a Mel filter bank, logarithmic energy and Discrete Cosine Transform (DCT), finally converting the audio signal into a Mel cepstrum domain, and further respectively obtaining a first-order differential coefficient delta MFCC and a second-order differential coefficient delta 2 The MFCC is weighted and then is combined with the MFCC to obtain newMFCC characteristic data; wherein the newMFCC is calculated by equation 4:
in the formula: a and b are weights, and 0<b<a<1, taking 1/3 of a, and taking 1/6 of b; MFCC is a static characteristic of audio; Δ MFCC is a dynamic property; delta 2 MFCC is the balance factor.
3. The method as claimed in claim 2, wherein the sound sensors are all of the same specification, and the distance between two adjacent sound sensors is a linear distance.
4. The method of claim 2, wherein the sound sensor is a condenser microphone.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211267984.0A CN115762529A (en) | 2022-10-17 | 2022-10-17 | Method for preventing cable from being broken outside by using voice recognition perception algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211267984.0A CN115762529A (en) | 2022-10-17 | 2022-10-17 | Method for preventing cable from being broken outside by using voice recognition perception algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115762529A true CN115762529A (en) | 2023-03-07 |
Family
ID=85351654
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211267984.0A Pending CN115762529A (en) | 2022-10-17 | 2022-10-17 | Method for preventing cable from being broken outside by using voice recognition perception algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115762529A (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108682098A (en) * | 2018-07-12 | 2018-10-19 | 国网江苏省电力有限公司扬州供电分公司 | A kind of cable external force damage prevention monitoring and alarming system and its working method |
| 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 |
| CN109117732A (en) * | 2018-07-16 | 2019-01-01 | 国网江西省电力有限公司电力科学研究院 | A kind of transmission line of electricity relates to the identification of bird failure bird kind figure sound and control method |
| CN110400571A (en) * | 2019-08-08 | 2019-11-01 | Oppo广东移动通信有限公司 | Audio-frequency processing method, device, storage medium and electronic equipment |
| CN111292510A (en) * | 2020-01-16 | 2020-06-16 | 广州华铭电力科技有限公司 | Recognition early warning method for urban cable damaged by external force |
| CN112951245A (en) * | 2021-03-09 | 2021-06-11 | 江苏开放大学（江苏城市职业学院） | Dynamic voiceprint feature extraction method integrated with static component |
| CN213814882U (en) * | 2020-11-23 | 2021-07-27 | 国网宁夏电力有限公司宁东供电公司 | Distribution lines prevents outer broken warning device |
| US20210337307A1 (en) * | 2020-04-27 | 2021-10-28 | Orcam Technologies Ltd. | Processing audio and video |
| CN113705418A (en) * | 2021-08-24 | 2021-11-26 | 中国人民解放军96901部队26分队 | Infrasound signal identification method, system and equipment based on MFCC and HMM |
| CN114594490A (en) * | 2022-01-26 | 2022-06-07 | 国网江苏省电力有限公司南京供电分公司 | Laser external damage prevention method and device for power line |
-
2022
- 2022-10-17 CN CN202211267984.0A patent/CN115762529A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN108682098A (en) * | 2018-07-12 | 2018-10-19 | 国网江苏省电力有限公司扬州供电分公司 | A kind of cable external force damage prevention monitoring and alarming system and its working method |
| CN109117732A (en) * | 2018-07-16 | 2019-01-01 | 国网江西省电力有限公司电力科学研究院 | A kind of transmission line of electricity relates to the identification of bird failure bird kind figure sound and control method |
| CN110400571A (en) * | 2019-08-08 | 2019-11-01 | Oppo广东移动通信有限公司 | Audio-frequency processing method, device, storage medium and electronic equipment |
| CN111292510A (en) * | 2020-01-16 | 2020-06-16 | 广州华铭电力科技有限公司 | Recognition early warning method for urban cable damaged by external force |
| US20210337307A1 (en) * | 2020-04-27 | 2021-10-28 | Orcam Technologies Ltd. | Processing audio and video |
| CN213814882U (en) * | 2020-11-23 | 2021-07-27 | 国网宁夏电力有限公司宁东供电公司 | Distribution lines prevents outer broken warning device |
| CN112951245A (en) * | 2021-03-09 | 2021-06-11 | 江苏开放大学（江苏城市职业学院） | Dynamic voiceprint feature extraction method integrated with static component |
| CN113705418A (en) * | 2021-08-24 | 2021-11-26 | 中国人民解放军96901部队26分队 | Infrasound signal identification method, system and equipment based on MFCC and HMM |
| CN114594490A (en) * | 2022-01-26 | 2022-06-07 | 国网江苏省电力有限公司南京供电分公司 | Laser external damage prevention method and device for power line |
Non-Patent Citations (1)
| Title |
|---|
| 黄山;吴振升;任志刚;刘弘景;桂媛;: "电力智能巡检机器人研究综述", 电测与仪表, no. 02, 31 December 2020 (2020-12-31) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109357167B (en) | Gas pipeline leakage point detection device and detection method | |
| US8073690B2 (en) | Speech recognition apparatus and method recognizing a speech from sound signals collected from outside | |
| CN109949823B (en) | DWPT-MFCC and GMM-based in-vehicle abnormal sound identification method | |
| CN102163427B (en) | Method for detecting audio exceptional event based on environmental model | |
| CN106323452B (en) | A kind of detection method and detection device of equipment abnormal sound | |
| CN110875060A (en) | Voice signal processing method, device, system, equipment and storage medium | |
| CN102563361A (en) | Device and method for detecting and positioning leakage of gas transmission pipeline based on conventional data and sound wave signals | |
| CN107170465B (en) | Audio quality detection method and audio quality detection system | |
| CN110816588B (en) | Locomotive wheel out-of-roundness detection method, device, equipment and system | |
| CN101829689B (en) | Drift fault recognition method of hot-rolling strip steel based on sound signals | |
| CN102915728B (en) | Sound segmentation device and method and speaker recognition system | |
| JPH1183618A (en) | Acoustic monitoring device | |
| US11480213B2 (en) | Bearing monitoring method and system | |
| CN111739542B (en) | Method, device and equipment for detecting characteristic sound | |
| CN113314144A (en) | Voice recognition and power equipment fault early warning method, system, terminal and medium | |
| CN116778956A (en) | Transformer acoustic feature extraction and fault identification method | |
| CN115932659A (en) | Transformer fault detection method based on voiceprint characteristics | |
| CN112052712B (en) | Power equipment state monitoring and fault identification method and system | |
| CN115762529A (en) | Method for preventing cable from being broken outside by using voice recognition perception algorithm | |
| CN113466616A (en) | Method and device for quickly positioning cable fault point | |
| CN116168727A (en) | Transformer abnormal sound detection method, system, equipment and storage medium | |
| CN114781466B (en) | Fault diagnosis method and system based on harmonic fundamental frequency of rotary mechanical vibration signal | |
| CN117292713A (en) | Transformer voiceprint monitoring method and system based on end Bian Yun cooperation | |
| JP2002323371A (en) | Sound diagnostic device and sound diagnostic method | |
| CN112049787A (en) | Non-contact water pump fault detection method and system |
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 |