CN117031154B - Transformer fault analysis method and system based on voiceprint recognition - Google Patents

Transformer fault analysis method and system based on voiceprint recognition Download PDF

Info

Publication number
CN117031154B
CN117031154B CN202310982921.1A CN202310982921A CN117031154B CN 117031154 B CN117031154 B CN 117031154B CN 202310982921 A CN202310982921 A CN 202310982921A CN 117031154 B CN117031154 B CN 117031154B
Authority
CN
China
Prior art keywords
fault
sound
group
suspected fault
waveform
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.)
Active
Application number
CN202310982921.1A
Other languages
Chinese (zh)
Other versions
CN117031154A (en
Inventor
吕泽鹏
张雍赟
王志平
尉镔
许�鹏
宋宏源
李涛
白跃昌
魏志成
杨东平
武兆亮
杨恺晋
张臻伟
梁灏
胡庆娟
闫丽婷
冯俊杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Super High Voltage Substation Branch Of State Grid Shanxi Electric Power Co
Original Assignee
Super High Voltage Substation Branch Of State Grid Shanxi Electric Power Co
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Super High Voltage Substation Branch Of State Grid Shanxi Electric Power Co filed Critical Super High Voltage Substation Branch Of State Grid Shanxi Electric Power Co
Priority to CN202310982921.1A priority Critical patent/CN117031154B/en
Publication of CN117031154A publication Critical patent/CN117031154A/en
Application granted granted Critical
Publication of CN117031154B publication Critical patent/CN117031154B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01RMEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01RMEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/50Testing of electric apparatus, lines, cables or components for short-circuits, continuity, leakage current or incorrect line connections
    • G01R31/62Testing of transformers

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Locating Faults (AREA)

Abstract

The application relates to a transformer fault analysis method and system based on voiceprint recognition, wherein the method comprises the steps of responding to received audio information, analyzing the audio information to obtain a suspected fault sound group and a known sound group; positioning the sound source by using the suspected fault sound group to obtain at least one suspected fault position; displaying the suspected fault position in the three-dimensional model; constructing a transmission path of a suspected fault sound group according to the suspected fault position in the three-dimensional model; correcting the suspected fault sound group by using the transmission path of the suspected fault sound group to obtain the fault sound group and giving a fault report of the transformer according to the fault sound group. According to the transformer fault analysis method and system based on voiceprint recognition, the fault point of the transformer is found by transferring the collected sound from the time domain to the frequency domain for analysis and positioning and orienting the collected sound, so that automatic early warning of the operation fault of the transformer is realized.

Description

Transformer fault analysis method and system based on voiceprint recognition
Technical Field
The application relates to the technical field of data analysis, in particular to a transformer fault analysis method and system based on voiceprint recognition.
Background
In recent years, the development of economy and industry makes the demand of China for electric power continuously increase, and an ultra-high voltage large power grid becomes a new trend of electric power development. Through many years of operation, the probability of faults of the transformer is continuously increased, and various faults such as insulation aging, component loosening and the like are at risk.
The traditional transformer detection method comprises dissolved gas analysis, infrared temperature measurement, partial discharge on-line monitoring, frequency response analysis, vibration analysis and the like. However, a detection blind area still exists for faults such as internal discharge, loosening of device screws and the like. Of course, experience personnel find out the fault mode through the operation sound of the transformer, but the mode of monitoring the human ear is easily influenced by subjective, has large error, can not monitor for a long time and can not achieve quantitative scientific analysis.
With the improvement of the intelligent inspection management level of the running state of the extra-high voltage transformer substation equipment, a pickup acquisition device is deployed in part of the extra-high voltage transformer substation, and the pickup acquisition device acquires, stores and plays running sounds of part of the equipment, so that how to acquire, process and analyze the sound signals of the transformer by means of the technology, realize real-time monitoring of the running of the transformer, solve the defect of manual monitoring, and effectively diagnose and evaluate the health state of the transformer and analyze and predict abnormal faults is an important research topic.
Disclosure of Invention
The utility model provides a transformer fault analysis method and system based on voiceprint discernment, through transfer the mode of analysis and location orientation to collecting the sound on the frequency domain to collecting the sound from the time domain to the transformer, realize the automatic early warning to transformer operation trouble.
The above object of the present application is achieved by the following technical solutions:
in a first aspect, the present application provides a method for analyzing a transformer fault based on voiceprint recognition, including:
responding to the received audio information, analyzing the audio information to obtain a suspected fault sound group and a known sound group;
positioning the sound source by using the suspected fault sound group to obtain at least one suspected fault position;
displaying the suspected fault position in the three-dimensional model;
constructing a transmission path of a suspected fault sound group according to the suspected fault position in the three-dimensional model;
correcting the suspected fault sound group by using a transmission path of the suspected fault sound group to obtain a fault sound group; and
and giving a fault report of the transformer according to the fault sound group.
In a possible implementation manner of the first aspect, parsing the audio information includes:
the audio information is transferred from the time domain to the frequency domain;
converting the audio information into a plurality of waveforms in the frequency domain;
combining the obtained waveforms according to the frequency and the amplitude to obtain a known sound group; and
the remaining waveforms are used as suspected fault sound groups.
In a possible implementation manner of the first aspect, locating the sound source using the suspected fault sound group, to obtain at least one suspected fault location includes:
determining the received positions of each waveform in the suspected fault sound group, wherein the number of the received positions is a plurality of;
locating a waveform using a plurality of received positions attributed to the waveform; and
determining a suspected fault position according to the positioning;
wherein one suspected fault location comprises a plurality of locations;
and (3) for the waveform which cannot be positioned, after the association degree with the suspected fault position is determined, attributing the waveform to the suspected fault position with the highest association degree.
In a possible implementation manner of the first aspect, determining the degree of association with the suspected fault location includes:
determining a received position of the waveform subjected to positioning associated with a received position of the waveform incapable of positioning, the received position of the waveform incapable of positioning being located within a received position composition area of the waveform subjected to positioning;
calculating the boundary distance between the received position of the waveform which cannot be positioned and the composition area; and
and selecting a suspected fault position corresponding to the component area with the minimum boundary distance as the suspected fault position of the waveform which cannot be positioned.
In a possible implementation manner of the first aspect, calculating the boundary distance includes:
determining the number of boundaries of the constituent regions;
calculating the distance from the received position of the waveform which cannot be positioned to each boundary of the composition area to obtain a plurality of boundary distance values; and
and accumulating the boundaries of the component areas to obtain the boundary distance.
In a possible implementation manner of the first aspect, correcting the suspected fault sound group using the transmission path of the suspected fault sound group includes:
determining a fixed object in the three-dimensional model according to the transfer path;
combining the suspected fault positions associated with the fixed object and generating a transmission path according to the fixed object;
and deleting the sound waveforms in the suspected fault sound group outside the transmission path to obtain a fault sound group.
In a possible implementation manner of the first aspect, the giving the fault report of the transformer according to the fault sound group includes grouping sound waveforms in the fault sound group according to frequency, the grouping including a high frequency group, a normal group and a low frequency group;
calculating the occurrence frequency of the sound waveforms in the high-frequency group and the low-frequency group in unit time, and giving a fault report of the transformer when the occurrence frequency of the high-frequency group and/or the low-frequency group is greater than the lowest occurrence frequency;
and calculating the similarity of the sound waveforms in the conventional group and the sound waveforms emitted by the corresponding fixed objects when the fixed objects work normally, and giving out a fault report of the transformer when the number of the sound waveforms in the conventional group with the similarity not meeting the requirement is larger than the minimum number.
In a second aspect, the present application provides a transformer fault analysis device based on voiceprint recognition, including:
the analysis and grouping unit is used for responding to the received audio information, analyzing the audio information and obtaining a suspected fault sound group and a known sound group;
the first positioning unit is used for positioning the sound source by using the suspected fault sound group to obtain at least one suspected fault position;
the display unit is used for displaying the suspected fault position in the three-dimensional model;
the path construction unit is used for constructing a transmission path of the suspected fault sound group according to the suspected fault position in the three-dimensional model;
the correction unit is used for correcting the suspected fault sound group by using the transmission path of the suspected fault sound group to obtain a fault sound group; and
and the reporting unit is used for giving a fault report of the transformer according to the fault sound group.
In a third aspect, the present application provides a transformer fault analysis system based on voiceprint recognition, the system comprising:
one or more memories for storing instructions; and
one or more processors configured to invoke and execute the instructions from the memory, to perform the method as described in the first aspect and any possible implementation of the first aspect.
In a fourth aspect, the present application provides a computer-readable storage medium comprising:
a program which, when executed by a processor, performs a method as described in the first aspect and any possible implementation of the first aspect.
In a fifth aspect, the present application provides a computer program product comprising program instructions which, when executed by a computing device, perform a method as described in the first aspect and any possible implementation manner of the first aspect.
In a sixth aspect, the present application provides a chip system comprising a processor for implementing the functions involved in the above aspects, e.g. generating, receiving, transmitting, or processing data and/or information involved in the above methods.
The chip system can be composed of chips, and can also comprise chips and other discrete devices.
In one possible design, the system on a chip also includes memory to hold the necessary program instructions and data. The processor and the memory may be decoupled, provided on different devices, respectively, connected by wire or wirelessly, or the processor and the memory may be coupled on the same device.
Drawings
Fig. 1 is an application schematic diagram of a transformer fault analysis method provided in the present application.
Fig. 2 is a schematic block diagram of a step flow of a transformer fault analysis method provided in the present application.
Fig. 3 is a schematic diagram of a positioning using a sound collector provided herein.
Fig. 4 is a schematic block diagram of a step flow for grouping audio information provided herein.
Fig. 5 is a schematic illustration of a suspected fault location obtained by means of a plurality of locations provided herein.
Fig. 6 is a schematic illustration of a principle of calculating the shortest straight line distance provided in the present application.
Fig. 7 is a schematic block diagram of a procedure for correcting a suspected fault sound group provided in the present application.
Detailed Description
The technical solutions in the present application are described in further detail below with reference to the accompanying drawings.
Referring to fig. 1, in some examples, the method for analyzing a transformer fault based on voiceprint recognition disclosed in the present application is applied to an analysis terminal, where audio information used by the analysis terminal comes from a sound collector disposed on the transformer, the sound collector is attached to an outer surface of the transformer, and the collected sound is sent to the analysis terminal through a wireless network.
Referring to fig. 2, the transformer fault analysis method based on voiceprint recognition disclosed in the present application includes the following steps:
s101, responding to received audio information, analyzing the audio information to obtain a suspected fault sound group and a known sound group;
s102, positioning the sound source by using the suspected fault sound group to obtain at least one suspected fault position;
s103, displaying the suspected fault position in the three-dimensional model;
s104, constructing a transmission path of the suspected fault sound group according to the suspected fault position in the three-dimensional model;
s105, correcting the suspected fault sound group by using a transmission path of the suspected fault sound group to obtain a fault sound group; and
and S106, giving a fault report of the transformer according to the fault sound group.
Specifically, in step S101, after the analysis terminal receives the audio information, the analysis terminal analyzes the audio information, and two analysis results are respectively a suspected fault sound group and a known sound group, where the suspected fault sound group represents a possible fault point inside the transformer; the characterization of the known sound group has been determined to be the sound made by the transformer during operation, and it is noted here that the known sound group includes both normal sounds generated during operation of the transformer and known fault sounds, and that the analysis terminal gives a fault report of the transformer when the known fault sounds occur in the known sound group.
For example, the sound generated during the operation of the transformer is related to the operation state of the transformer, so that the sound possibly generated may be guessed according to the operation state of the transformer, for example:
when the transformer is in overload operation, the current flowing through the coil is strong, and the sound of the transformer becomes lower and louder;
equipment load is started, and a sound of 'chuck' or 'Java' is generated under the influence of higher harmonic waves;
when the transformer has short-circuit fault, the current can be rapidly increased and changed, and the transformer can emit sound of 'whistling'; the transformer and the metal clamping piece thereof can induce potential, and the potential difference is excessively large, so that the sound is changed;
when overvoltage faults occur in the power grid, if ferromagnetic resonance occurs, humming or humming sounds are emitted;
the clamping piece or screw loosens in the transformer, can make some parts of silicon steel sheet collide each other and send abnormal sound.
In step S102, the sound source is positioned by using the suspected fault sound group, so as to obtain at least one suspected fault location. Here, it should be noted that the number of sound collectors disposed on the transformer is plural, and these sound collectors form a uniform linear array, and the calculation method using the uniform linear array is as follows:
assume that in a uniform linear array consisting of 0, 1..m-1 array elements, the sound source signals received by the reference array elements areThen the signal received by the M th array element is +.>Then the fourier transform of a signal of a certain frequency (narrowband) is:
wherein N corresponds to each array element, then the signal received by the uniform linear array at this time is:
then we define the array manifold as:the weight of each array element (sound collector) is the phase delay of the array element to be compensated relative to the reference array element, so the weight is the conjugate transpose of the array manifold. By means of a plurality of array elements (sound collectors) the reception time, the reception angle and the amplitude (amplitude) of the received sound waveform can be determined.
After the completion of step S102, a suspected fault location is obtained, as shown in fig. 3.
In step S103, the suspected fault location is displayed in the three-dimensional model, i.e., the suspected fault location is visualized by the three-dimensional model.
Step S104 is then executed, in which a transmission path of the suspected fault sound group is constructed according to the suspected fault position in the three-dimensional model, the purpose of the constructed transmission path is to correct the suspected fault sound group, and the fault sound group is obtained after the correction is completed.
The purpose of the correction is to exclude audio information that does not belong to the group of faulty sounds. Because during the suspected fault location, too many suspected fault locations are obtained, and there are some accurate and inaccurate suspected fault locations, for example, some suspected fault locations may appear in the transformer oil, it is obvious that the suspected fault locations are inaccurate.
The reasons for this are missing data acquisition, data acquisition errors, calculation errors, etc. It is therefore necessary to construct a transmission path of the suspected malfunctioning sound group and then correct the suspected malfunctioning sound group using the transmission path.
The specific principle of correction is that firstly a fuzzy position determination is carried out, and then an accurate position determination is carried out by means of a three-dimensional model. Since a certain component may generate a sound when a fault occurs in the transformer, the sound generation position is fixed, but when the reverse position determination is performed through sound collection, a certain amount of blurring exists, and the reasons of blurring include data collection missing, data collection error, calculation error and the like.
Finally, step S106 is performed, in which a fault report of the transformer is given according to the fault sound group, and the fault report includes information such as the number of the transformer (through the sound collector), the fault location (through the three-dimensional model), and the fault type (through the known fault sound).
In some examples, referring to fig. 4, parsing the audio information includes the steps of:
s201, audio information is transferred from a time domain to a frequency domain;
s202, converting the audio information into a plurality of waveforms in a frequency domain;
s203, combining the obtained waveforms according to the frequency and the amplitude to obtain a known sound group; and
and S204, taking the residual waveform as a suspected fault sound group.
The purpose of transferring the audio information from the time domain to the frequency domain is to obtain a more detailed audio information content. Under a simple summary, the time domain is an overall summary of the time-varying signal over the time axis. When a signal is subjected to time domain analysis, time domain parameters of some signals are sometimes the same, but it cannot be said that the signals are identical. Since the signal varies not only with time but also with information about frequency, phase, etc., it is necessary to further analyze the frequency structure of the signal and describe the signal in the frequency domain.
After the method is transferred to the frequency domain, the audio information is transferred into a plurality of waveforms, the obtained waveforms are combined according to the frequency and the amplitude to obtain a known sound group, the known sound group is a part of all the audio information, and after the known sound group is eliminated, the data content of the audio information can be reduced.
In the process of obtaining the known sound group, the known sound group is directly screened out through the sound generated by the parts in the transformer in normal operation and the sound generated by the parts in the transformer in operation under the known fault state. However, this screening method can only perform basic screening, because the sound generated by part of the faults is not fixed and is influenced by environmental factors, operation factors and the like.
And finally, taking the residual waveform as a suspected fault sound group, and completing the division work of the known sound group and the suspected fault sound group.
In some examples, locating the sound source using the suspected fault sound group to obtain the at least one suspected fault location includes the steps of:
s301, determining the received positions of each waveform in the suspected fault sound group, wherein the number of the received positions is a plurality of;
s302, locating a waveform by using a plurality of received positions belonging to the waveform; and
s303, determining a suspected fault position according to the positioning;
wherein one suspected fault location comprises a plurality of locations;
and (3) for the waveform which cannot be positioned, after the association degree with the suspected fault position is determined, attributing the waveform to the suspected fault position with the highest association degree.
Specifically, in step S301, the received position of each waveform in the suspected faulty sound group is first determined, where the received position refers to the position of the sound collector, and in this step, the number of received positions is a plurality.
Since waveforms in the suspected faulty sound group with only one received location cannot complete subsequent localization work. In step S302, a waveform is located using a plurality of received positions assigned to the waveform, where locating refers to determining a generation position of the waveform.
Finally, in step S303, the suspected fault location is determined according to the positioning, which is based on the specific principle that the positioning obtained in step S302 has a certain degree of aggregation, and the degree of aggregation points to an area, which is the suspected fault location.
That is, one suspected fault location includes a plurality of locations. For example, where at least three locations are set to meet the concentration requirement to be considered as suspected fault locations, one implementation is to construct a sphere (dashed circle in fig. 5) and then move the sphere over space, where the sphere is considered as a suspected fault location when at least three locations are within the sphere.
For the waveform which cannot be positioned, after the association degree with the suspected fault position is determined, the waveform is attributed to the suspected fault position with the highest association degree, and the specific steps are as follows:
s401, determining the received position of the waveform which is associated with the received position of the waveform and is subjected to positioning, wherein the received position of the waveform which is not subjected to positioning is positioned in a received position composition area of the waveform which is subjected to positioning;
s402, calculating the boundary distance between the received position of the waveform which cannot be positioned and the composition area; and
s403, selecting a suspected fault position corresponding to the composition area with the minimum boundary distance as the suspected fault position of the waveform which cannot be positioned.
Specifically, in step S401, the received position of the waveform for which positioning is completed, which is associated with the received position of the waveform for which positioning is impossible, is first determined, in such a manner that the received position of the waveform for which positioning is impossible is located within the received position composition area of the waveform for which positioning is completed.
For example, for a suspected fault location, all the associated received locations (sound collectors) are sequentially connected to form a closed graph; the received position of the waveform where no positioning can be performed is regarded as a point which is within the closed figure, as shown in fig. 6.
This approach mainly considers that the waveform that cannot be located attenuates during the propagation after it is generated from its generation point, so a shortest propagation path is selected to determine its attribution. Meanwhile, when a waveform that cannot be located is associated with a plurality of suspected fault positions, it is necessary to calculate the linear distance of the receiving position thereof to each of the associated suspected fault positions, and then select one of the suspected fault positions having the shortest linear distance (the summation of S1 to S6 in fig. 6).
The step of calculating the boundary distance is as follows:
determining the number of boundaries of the constituent regions;
calculating the distance from the received position of the waveform which cannot be positioned to each boundary of the composition area to obtain a plurality of boundary distance values; and
and accumulating the boundaries of the component areas to obtain the boundary distance.
In some examples, referring to fig. 7, correcting the suspected malfunctioning sound group using the transmission path of the suspected malfunctioning sound group includes:
s501, determining a fixed object in a three-dimensional model according to a transmission path;
s502, combining suspected fault positions associated with the fixed object and generating a transmission path according to the fixed object;
s503, deleting the sound waveform in the suspected fault sound group outside the transmission path to obtain a fault sound group.
Specifically, in steps S501 to S503, a fixed object in the three-dimensional model, which refers to a specific component or a specific region in the transformer, is determined by the transfer path first. Because in the actual operation of the transformer, a sound with certain characteristics is generated when a specific component fails or a certain area (where a plurality of strongly related components exist) fails.
After the fixed object is obtained, the suspected fault positions associated with the fixed object are combined and a transmission path is generated according to the fixed object, wherein the transmission path is a virtual sound transmission path. And finally deleting the sound waveforms in the suspected fault sound group outside the transmission path to obtain the fault sound group.
The specific mode of deletion calculates the coincidence ratio of the transmission path of each specific waveform and the transmission path generated according to the fixed object, for example, the coincidence ratio is set to be 80%, and when the coincidence ratio of the transmission path of one specific waveform and the transmission path generated according to the fixed object is less than 80%, the waveform is discarded.
Rejection means that the waveform is not generated at a fixed object.
Also, merging suspected fault locations associated with a stationary object may also facilitate more accurate determination of where the stationary object is, as there are more sound samples.
In some examples, the specific decision is as follows:
giving a fault report of the transformer according to the fault sound group comprises grouping sound waveforms in the fault sound group according to frequencies, wherein the grouping comprises a high-frequency group, a normal group and a low-frequency group;
calculating the occurrence frequency of the sound waveforms in the high-frequency group and the low-frequency group in unit time, and giving a fault report of the transformer when the occurrence frequency of the high-frequency group and/or the low-frequency group is greater than the lowest occurrence frequency;
and calculating the similarity of the sound waveforms in the conventional group and the sound waveforms emitted by the corresponding fixed objects when the fixed objects work normally, and giving out a fault report of the transformer when the number of the sound waveforms in the conventional group with the similarity not meeting the requirement is larger than the minimum number.
It will be appreciated that the frequency of the sound generated by the transformer during normal operation is within a relatively fixed range, and that the transformer may fail when sounds of frequencies outside the relatively fixed range occur during operation.
Thus first giving a fault report of the transformer from the fault sound group comprises grouping the sound waveforms in the fault sound group according to frequency, the grouping comprising a high frequency group, a normal group and a low frequency group. Wherein the conventional group corresponds to the sound mentioned in the foregoing having a frequency in a relatively fixed range, and the high frequency group and the low frequency group correspond to the sound having a frequency outside the relatively fixed range, respectively.
For the high frequency group and the low frequency group, specific decision rules are: and calculating the occurrence frequency of the sound waveforms in the high-frequency group and the low-frequency group in unit time, and giving a fault report of the transformer when the occurrence frequency of the high-frequency group and/or the low-frequency group is greater than the lowest occurrence frequency.
Specifically, the high-frequency group and the low-frequency group are determined mainly according to the frequency or time of occurrence, for example, the high-frequency group and the low-frequency group occur only once, and then the problem that the data abnormality is not reproduced in the latter is possibly considered; when the high frequency group and the low frequency group occur a plurality of times at that time, it is necessary to make a determination as to whether or not an abnormal condition has occurred in the operation of the transformer.
For the regular group, the specific decision rules are:
and calculating the similarity of the sound waveforms in the conventional group and the sound waveforms emitted by the corresponding fixed objects when the fixed objects work normally, and giving out a fault report of the transformer when the number of the sound waveforms in the conventional group with the similarity not meeting the requirement is larger than the minimum number.
Specifically, the frequency of occurrence cannot be simply used for determining the sound waveforms in the conventional group, and the sound waveforms emitted by the fixed object during normal operation are also selected for use, so that the overall feedback on the normal operation cannot be realized.
Therefore, in the application, the method of similarity comparison is used for judging, the similarity has two dimensions of amplitude and frequency, similar sound waveforms can be deleted from the conventional group through the amplitude and the frequency, and then the number judgment method is used for predicting whether the operation of the transformer is abnormal or not for the remaining sound waveforms in the conventional group.
The application also provides a transformer fault analysis device based on voiceprint recognition, which comprises:
the analysis and grouping unit is used for responding to the received audio information, analyzing the audio information and obtaining a suspected fault sound group and a known sound group;
the first positioning unit is used for positioning the sound source by using the suspected fault sound group to obtain at least one suspected fault position;
the display unit is used for displaying the suspected fault position in the three-dimensional model;
the path construction unit is used for constructing a transmission path of the suspected fault sound group according to the suspected fault position in the three-dimensional model;
the correction unit is used for correcting the suspected fault sound group by using the transmission path of the suspected fault sound group to obtain a fault sound group; and
and the reporting unit is used for giving a fault report of the transformer according to the fault sound group.
Further, the method further comprises the following steps:
the first processing unit is used for converting the audio information from the time domain to the frequency domain;
a second processing unit for converting the audio information into a plurality of waveforms in a frequency domain;
the combination unit is used for combining the obtained waveforms according to the frequency and the amplitude to obtain a known sound group; and
and the third processing unit is used for taking the residual waveform as a suspected fault sound group.
Further, the method further comprises the following steps:
a received position determining unit, configured to determine a received position of each waveform in the suspected fault sound group, where the number of received positions is a plurality of received positions;
a second positioning unit for positioning a waveform using a plurality of received positions assigned to the waveform; and
the third positioning unit is used for determining the suspected fault position according to positioning;
wherein one suspected fault location comprises a plurality of locations;
and (3) for the waveform which cannot be positioned, after the association degree with the suspected fault position is determined, attributing the waveform to the suspected fault position with the highest association degree.
Further, the method further comprises the following steps:
a fourth positioning unit configured to determine a received position of a waveform for which positioning is completed in association with a received position of a waveform for which positioning is impossible, the received position of the waveform for which positioning is impossible being located within a received position composition area of the waveform for which positioning is completed;
the first calculating unit is used for calculating the boundary distance between the received position of the waveform which cannot be positioned and the composition area; and
and the selecting unit is used for selecting the suspected fault position corresponding to the composition area with the minimum boundary distance as the suspected fault position of the waveform which cannot be positioned.
Further, the method further comprises the following steps:
a first determining unit configured to determine the number of boundaries of the constituent regions;
the second calculation unit is used for calculating the distance from the received position of the waveform which cannot be positioned to each boundary of the composition area to obtain a plurality of boundary distance values; and
and the accumulation unit is used for accumulating the boundaries of the component areas to obtain the boundary distance.
Further, the method further comprises the following steps:
a second determining unit configured to determine a fixed object in the three-dimensional model based on the transfer path;
the fourth processing unit is used for merging suspected fault positions associated with the fixed object and generating a transmission path according to the fixed object;
and a fifth processing unit, configured to delete the sound waveform in the suspected fault sound group located outside the transmission path, to obtain a fault sound group.
Further, the method further comprises the following steps:
a grouping unit for giving a fault report of the transformer according to the fault sound group, including grouping sound waveforms in the fault sound group according to frequency, the grouping including a high frequency group, a normal group and a low frequency group;
the first analysis and reporting unit is used for calculating the occurrence frequency of the sound waveforms in the high-frequency group and the low-frequency group in unit time, and giving a fault report of the transformer when the occurrence frequency of the high-frequency group and/or the low-frequency group is greater than the lowest occurrence frequency;
and the second analysis and reporting unit is used for calculating the similarity between the sound waveforms in the conventional group and the sound waveforms emitted by the corresponding fixed object when the fixed object works normally, and giving out fault reports of the transformer when the number of the sound waveforms in the conventional group with the similarity not meeting the requirement is larger than the minimum number.
In one example, the unit in any of the above apparatuses may be one or more integrated circuits configured to implement the above methods, for example: one or more application specific integrated circuits (applicationspecific integratedcircuit, ASIC), or one or more digital signal processors (digital signal processor, DSP), or one or more field programmable gate arrays (fieldprogrammable gate array, FPGA), or a combination of at least two of these integrated circuit forms.
For another example, when the units in the apparatus may be implemented in the form of a scheduler of processing elements, the processing elements may be general-purpose processors, such as a central processing unit (central processing unit, CPU) or other processor that may invoke the program. For another example, the units may be integrated together and implemented in the form of a system-on-a-chip (SOC).
Various objects such as various messages/information/devices/network elements/systems/devices/actions/operations/processes/concepts may be named in the present application, and it should be understood that these specific names do not constitute limitations on related objects, and that the named names may be changed according to the scenario, context, or usage habit, etc., and understanding of technical meaning of technical terms in the present application should be mainly determined from functions and technical effects that are embodied/performed in the technical solution.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown 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 units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
It should also be understood that in various embodiments of the present application, first, second, etc. are merely intended to represent that multiple objects are different. For example, the first time window and the second time window are only intended to represent different time windows. Without any effect on the time window itself, the first, second, etc. mentioned above should not impose any limitation on the embodiments of the present application.
It is also to be understood that in the various embodiments of the application, terms and/or descriptions of the various embodiments are consistent and may be referenced to one another in the absence of a particular explanation or logic conflict, and that the features of the various embodiments may be combined to form new embodiments in accordance with their inherent logic relationships.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a computer-readable storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned computer-readable storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The present application also provides a computer program product comprising instructions that, when executed, cause the fault analysis system to perform operations of the fault analysis system corresponding to the above-described method.
The application also provides a transformer fault analysis system based on voiceprint recognition, the system comprises:
one or more memories for storing instructions; and
one or more processors configured to invoke and execute the instructions from the memory to perform the method as described above.
The present application also provides a chip system comprising a processor for implementing the functions involved in the above, e.g. generating, receiving, transmitting, or processing data and/or information involved in the above method.
The chip system can be composed of chips, and can also comprise chips and other discrete devices.
The processor referred to in any of the foregoing may be a CPU, microprocessor, ASIC, or integrated circuit that performs one or more of the procedures for controlling the transmission of feedback information described above.
In one possible design, the system on a chip also includes memory to hold the necessary program instructions and data. The processor and the memory may be decoupled, and disposed on different devices, respectively, and connected by wired or wireless means, so as to support the chip system to implement the various functions in the foregoing embodiments. In the alternative, the processor and the memory may be coupled to the same device.
Optionally, the computer instructions are stored in a memory.
Alternatively, the memory may be a storage unit in the chip, such as a register, a cache, etc., and the memory may also be a storage unit in the terminal located outside the chip, such as a ROM or other type of static storage device, a RAM, etc., that may store static information and instructions.
It is to be understood that the memory in this application may be either volatile memory or nonvolatile memory, or may include both volatile and nonvolatile memory.
The non-volatile memory may be a ROM, programmable ROM (PROM), erasable programmable ROM (erasablePROM, EPROM), electrically erasable programmable EPROM (EEPROM), or flash memory.
The volatile memory may be RAM, which acts as external cache. There are many different types of RAM, such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), and direct memory bus RAM.
The embodiments of the present invention are all preferred embodiments of the present application, and are not intended to limit the scope of the present application in this way, therefore: all equivalent changes in structure, shape and principle of this application should be covered in the protection scope of this application.

Claims (7)

1. The transformer fault analysis method based on voiceprint recognition is characterized by comprising the following steps of:
responding to the received audio information, analyzing the audio information to obtain a suspected fault sound group and a known sound group;
positioning the sound source by using the suspected fault sound group to obtain at least one suspected fault position;
displaying the suspected fault position in the three-dimensional model;
constructing a transmission path of a suspected fault sound group according to the suspected fault position in the three-dimensional model;
correcting the suspected fault sound group by using a transmission path of the suspected fault sound group to obtain a fault sound group; and
giving a fault report of the transformer according to the fault sound group;
positioning the sound source by using the suspected fault sound group to obtain at least one suspected fault position, wherein the method comprises the following steps:
determining the received positions of each waveform in the suspected fault sound group, wherein the number of the received positions is a plurality of;
locating a waveform using a plurality of received positions attributed to the waveform; and
determining a suspected fault position according to the positioning;
wherein one suspected fault location comprises a plurality of locations;
for the waveform which cannot be positioned, after the association degree with the suspected fault position is determined, the waveform is attributed to the suspected fault position with the highest association degree;
determining the degree of association with the suspected fault location includes:
determining a received position of the waveform subjected to positioning associated with a received position of the waveform incapable of positioning, the received position of the waveform incapable of positioning being located within a received position composition area of the waveform subjected to positioning;
calculating the boundary distance between the received position of the waveform which cannot be positioned and the composition area; and
selecting a suspected fault position corresponding to the composition area with the minimum boundary distance as the suspected fault position of the waveform which cannot be positioned;
calculating the boundary distance includes:
determining the number of boundaries of the constituent regions;
calculating the distance from the received position of the waveform which cannot be positioned to each boundary of the composition area to obtain a plurality of boundary distance values; and
and accumulating the boundaries of the component areas to obtain the boundary distance.
2. The method for analyzing transformer faults based on voiceprint recognition of claim 1 in which parsing the audio information includes:
the audio information is transferred from the time domain to the frequency domain;
converting the audio information into a plurality of waveforms in the frequency domain;
combining the obtained waveforms according to the frequency and the amplitude to obtain a known sound group; and
the remaining waveforms are used as suspected fault sound groups.
3. The method for analyzing a transformer fault based on voiceprint recognition according to claim 1 or 2, wherein correcting the suspected fault sound group using the transmission path of the suspected fault sound group comprises:
determining a fixed object in the three-dimensional model according to the transfer path;
combining the suspected fault positions associated with the fixed object and generating a transmission path according to the fixed object;
and deleting the sound waveforms in the suspected fault sound group outside the transmission path to obtain a fault sound group.
4. The method of claim 1, wherein the step of reporting the fault of the transformer based on the fault sound group comprises grouping sound waveforms in the fault sound group according to frequency, the grouping including a high frequency group, a normal group, and a low frequency group;
calculating the occurrence frequency of the sound waveforms in the high-frequency group and the low-frequency group in unit time, and giving a fault report of the transformer when the occurrence frequency of the high-frequency group and/or the low-frequency group is greater than the lowest occurrence frequency;
and calculating the similarity of the sound waveforms in the conventional group and the sound waveforms emitted by the corresponding fixed objects when the fixed objects work normally, and giving out a fault report of the transformer when the number of the sound waveforms in the conventional group with the similarity not meeting the requirement is larger than the minimum number.
5. A transformer fault analysis device based on voiceprint recognition, comprising:
the analysis and grouping unit is used for responding to the received audio information, analyzing the audio information and obtaining a suspected fault sound group and a known sound group;
the first positioning unit is used for positioning the sound source by using the suspected fault sound group to obtain at least one suspected fault position;
the display unit is used for displaying the suspected fault position in the three-dimensional model;
the path construction unit is used for constructing a transmission path of the suspected fault sound group according to the suspected fault position in the three-dimensional model;
the correction unit is used for correcting the suspected fault sound group by using the transmission path of the suspected fault sound group to obtain a fault sound group; and
a reporting unit for giving a fault report of the transformer according to the fault sound group;
a received position determining unit, configured to determine a received position of each waveform in the suspected fault sound group, where the number of received positions is a plurality of received positions;
a second positioning unit for positioning a waveform using a plurality of received positions assigned to the waveform;
the third positioning unit is used for determining the suspected fault position according to positioning;
wherein one suspected fault location comprises a plurality of locations;
for the waveform which cannot be positioned, after the association degree with the suspected fault position is determined, the waveform is attributed to the suspected fault position with the highest association degree;
a fourth positioning unit configured to determine a received position of a waveform for which positioning is completed in association with a received position of a waveform for which positioning is impossible, the received position of the waveform for which positioning is impossible being located within a received position composition area of the waveform for which positioning is completed;
the first calculating unit is used for calculating the boundary distance between the received position of the waveform which cannot be positioned and the composition area; and
the selecting unit is used for selecting a suspected fault position corresponding to the composition area with the minimum boundary distance as the suspected fault position of the waveform which cannot be positioned;
a first determining unit configured to determine the number of boundaries of the constituent regions;
the second calculation unit is used for calculating the distance from the received position of the waveform which cannot be positioned to each boundary of the composition area to obtain a plurality of boundary distance values; and
and the accumulation unit is used for accumulating the boundaries of the component areas to obtain the boundary distance.
6. A transformer fault analysis system based on voiceprint recognition, the system comprising:
one or more memories for storing instructions; and
one or more processors to invoke and execute the instructions from the memory to perform the method of any of claims 1-4.
7. A computer-readable storage medium, the computer-readable storage medium comprising:
program which, when executed by a processor, performs the method according to any one of claims 1 to 4.
CN202310982921.1A 2023-08-07 2023-08-07 Transformer fault analysis method and system based on voiceprint recognition Active CN117031154B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310982921.1A CN117031154B (en) 2023-08-07 2023-08-07 Transformer fault analysis method and system based on voiceprint recognition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310982921.1A CN117031154B (en) 2023-08-07 2023-08-07 Transformer fault analysis method and system based on voiceprint recognition

Publications (2)

Publication Number Publication Date
CN117031154A CN117031154A (en) 2023-11-10
CN117031154B true CN117031154B (en) 2024-03-22

Family

ID=88638490

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310982921.1A Active CN117031154B (en) 2023-08-07 2023-08-07 Transformer fault analysis method and system based on voiceprint recognition

Country Status (1)

Country Link
CN (1) CN117031154B (en)

Citations (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
SE9404209D0 (en) * 1994-12-05 1994-12-05 Abb Research Ltd Method and apparatus for locating partial discharges of an electric high voltage apparatus
EP1686673A1 (en) * 2005-01-27 2006-08-02 Hojgaard Sound IS Acoustic detection of power network failures
KR20080099429A (en) * 2007-05-09 2008-11-13 (주)태광이엔시 Acoustic emission measurement intelligent device), diagonos system, and a method thereof
KR20140033944A (en) * 2012-09-11 2014-03-19 엘에스전선 주식회사 System and method for monitoring-diagnose wind power transformer
CN108760034A (en) * 2018-05-21 2018-11-06 广西电网有限责任公司电力科学研究院 A kind of transformer vibration noise source positioning system and method
CN110672950A (en) * 2019-10-08 2020-01-10 深圳海岸语音技术有限公司 Power equipment fault sound image detection system and method
KR20200060179A (en) * 2018-11-22 2020-05-29 고려대학교 세종산학협력단 Transformer fault diagnosis with sound information
CN111739557A (en) * 2020-06-19 2020-10-02 浙江讯飞智能科技有限公司 Equipment fault positioning method, device, equipment and storage medium
KR102264773B1 (en) * 2020-11-30 2021-06-15 주식회사 국제기술인증원 Fault Diagnosis System for Diagnosis and Control of Temperature, Partial Discharge and Noise of Transformers Using Sensors
CN113283310A (en) * 2021-05-07 2021-08-20 国网浙江省电力有限公司武义县供电公司 System and method for detecting health state of power equipment based on voiceprint features
CN113362856A (en) * 2021-06-21 2021-09-07 国网上海市电力公司 Sound fault detection method and device applied to power Internet of things
CN114280413A (en) * 2021-11-29 2022-04-05 山东信通电子股份有限公司 Method and device for positioning abnormal fault sound of power transmission channel
WO2022100323A1 (en) * 2020-11-10 2022-05-19 国网新疆电力有限公司电力科学研究院 Perception system for operating state of large transformer based on sound-vibration integration
CN115291056A (en) * 2022-07-29 2022-11-04 国网安徽省电力有限公司超高压分公司 Transformer working state identification system and method based on voiceprint identification model
CN115358110A (en) * 2022-07-25 2022-11-18 国网江苏省电力有限公司淮安供电分公司 Transformer fault detection system based on acoustic sensor array
CN115586410A (en) * 2022-11-03 2023-01-10 国网安徽省电力有限公司电力科学研究院 Unmanned aerial vehicle-based transformer voiceprint inspection device and method
KR20230062189A (en) * 2021-10-29 2023-05-09 주식회사 싸이콤 Method for diagnosing power facility based on artificial intelligence and device using the same
CN116189711A (en) * 2023-04-26 2023-05-30 四川省机场集团有限公司 Transformer fault identification method and device based on acoustic wave signal monitoring
CN116223040A (en) * 2023-03-07 2023-06-06 华北电力大学(保定) Carrier roller fault acoustic signal feature extraction and positioning method
CN117095695A (en) * 2023-10-19 2023-11-21 国网山西省电力公司超高压变电分公司 Wide-area voiceprint compression acquisition method and system for transformer body

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10348554B2 (en) * 2016-04-25 2019-07-09 Cisco Technology, Inc. Hybrid fibre coaxial fault locationing in cable network environments
US20230024104A1 (en) * 2021-07-23 2023-01-26 Nec Laboratories America, Inc Identification of false transformer humming using machine learning

Patent Citations (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
SE9404209D0 (en) * 1994-12-05 1994-12-05 Abb Research Ltd Method and apparatus for locating partial discharges of an electric high voltage apparatus
EP1686673A1 (en) * 2005-01-27 2006-08-02 Hojgaard Sound IS Acoustic detection of power network failures
KR20080099429A (en) * 2007-05-09 2008-11-13 (주)태광이엔시 Acoustic emission measurement intelligent device), diagonos system, and a method thereof
KR20140033944A (en) * 2012-09-11 2014-03-19 엘에스전선 주식회사 System and method for monitoring-diagnose wind power transformer
CN108760034A (en) * 2018-05-21 2018-11-06 广西电网有限责任公司电力科学研究院 A kind of transformer vibration noise source positioning system and method
KR20200060179A (en) * 2018-11-22 2020-05-29 고려대학교 세종산학협력단 Transformer fault diagnosis with sound information
CN110672950A (en) * 2019-10-08 2020-01-10 深圳海岸语音技术有限公司 Power equipment fault sound image detection system and method
CN111739557A (en) * 2020-06-19 2020-10-02 浙江讯飞智能科技有限公司 Equipment fault positioning method, device, equipment and storage medium
WO2022100323A1 (en) * 2020-11-10 2022-05-19 国网新疆电力有限公司电力科学研究院 Perception system for operating state of large transformer based on sound-vibration integration
KR102264773B1 (en) * 2020-11-30 2021-06-15 주식회사 국제기술인증원 Fault Diagnosis System for Diagnosis and Control of Temperature, Partial Discharge and Noise of Transformers Using Sensors
CN113283310A (en) * 2021-05-07 2021-08-20 国网浙江省电力有限公司武义县供电公司 System and method for detecting health state of power equipment based on voiceprint features
CN113362856A (en) * 2021-06-21 2021-09-07 国网上海市电力公司 Sound fault detection method and device applied to power Internet of things
KR20230062189A (en) * 2021-10-29 2023-05-09 주식회사 싸이콤 Method for diagnosing power facility based on artificial intelligence and device using the same
CN114280413A (en) * 2021-11-29 2022-04-05 山东信通电子股份有限公司 Method and device for positioning abnormal fault sound of power transmission channel
CN115358110A (en) * 2022-07-25 2022-11-18 国网江苏省电力有限公司淮安供电分公司 Transformer fault detection system based on acoustic sensor array
CN115291056A (en) * 2022-07-29 2022-11-04 国网安徽省电力有限公司超高压分公司 Transformer working state identification system and method based on voiceprint identification model
CN115586410A (en) * 2022-11-03 2023-01-10 国网安徽省电力有限公司电力科学研究院 Unmanned aerial vehicle-based transformer voiceprint inspection device and method
CN116223040A (en) * 2023-03-07 2023-06-06 华北电力大学(保定) Carrier roller fault acoustic signal feature extraction and positioning method
CN116189711A (en) * 2023-04-26 2023-05-30 四川省机场集团有限公司 Transformer fault identification method and device based on acoustic wave signal monitoring
CN117095695A (en) * 2023-10-19 2023-11-21 国网山西省电力公司超高压变电分公司 Wide-area voiceprint compression acquisition method and system for transformer body

Also Published As

Publication number Publication date
CN117031154A (en) 2023-11-10

Similar Documents

Publication Publication Date Title
CN107102244B (en) A kind of discharge source localization method of GIS ultrahigh frequency local discharge on-line monitoring device
EP2156203B1 (en) Method and device to predict a state of a power system in the time domain
US11448682B2 (en) Trending functions for partial discharge
AU2010358396B2 (en) Apparatus and method for monitoring an electric power transmission system through partial discharges analysis
US20130096853A1 (en) Systems and methods for monitoring electrical contacts
CN109342883A (en) A kind of local ageing fault detecting and positioning method for cable
JP5456582B2 (en) Transformer soundness diagnosis method, soundness diagnosis device, and soundness diagnosis program
Wang et al. Reflectometry-based cable insulation aging diagnosis and prognosis
CN113296029A (en) Distribution transformer voiceprint monitoring method, device, equipment and storage medium
CN106656368A (en) Communication system monitoring method and apparatus
CN117031154B (en) Transformer fault analysis method and system based on voiceprint recognition
US11231999B2 (en) Detection of electric power system anomalies in streaming measurements
CN117095695B (en) Wide-area voiceprint compression acquisition method and system for transformer body
EP3793333A1 (en) Filament current control method and apparatus
CN114019422A (en) Transformer fault monitoring system based on ATT-BilSTM
CN111314110B (en) Fault early warning method for distributed system
CN111866921A (en) Method, device and equipment for searching service fault of 5G base station and storage medium
CN111507618A (en) Method and device for checking longitude and latitude of cell and storage medium
CN117113833A (en) Verification method and system of verification device
CN107132500A (en) A kind of synchronous phasor measurement unit on-line calibration method and apparatus
KR101623354B1 (en) Quality inspection method based on integrated quality inspection syetem
CN116436030A (en) New energy station broadband oscillation on-site monitoring control method and device
CN114675143A (en) Partial discharge measurement method and device
CN112020087B (en) Tunnel fault monitoring method and device and computing equipment
CN113779112A (en) Electric energy quality analysis system based on spatial information and big data mapping algorithm

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
GR01 Patent grant
GR01 Patent grant