A partial discharge audio recognition method and system based on power frequency sine similarity

Abstract

The present invention relates to the field of audio recognition technology, and in particular to a partial discharge audio recognition method and system based on industrial frequency sinusoidal similarity. The method collects and preprocesses audio signals within a specific frequency band emitted by a device to be tested, performs window segmentation on an audio waveform fitted by a pure audio signal, arranges sub-waveforms obtained by segmentation, calculates the similarity between each sub-waveform and a preset sinusoidal waveform, thereby eliminating sub-waveforms with low similarity from a sub-waveform time series set, discretizes each sub-waveform, constructs a discrete point matrix according to corresponding positions of discrete points, calculates the maximum slope of the corresponding sub-waveform, determines a pulse waveform according to the maximum slope, and judges whether the audio signal is a partial discharge audio according to the phase position and phase difference of the pulse waveform and the number of pulse waveforms, thereby improving the sensitivity of partial discharge audio recognition and reducing the interference of high-frequency signals on recognition results.

Description

Technical Field

[0001] The present invention relates to the technical field of audio recognition, and in particular to a partial discharge audio recognition method and system based on power frequency sinusoidal similarity. Background Art

[0002] At present, although the power equipment such as switch cabinets in the distribution network has a certain tolerance, insulation aging problems will occur in long-term operation. After insulation defects occur, the power equipment is prone to partial discharge, and the generation of partial discharge will have a huge impact on the insulation condition and operation condition of the power equipment. In severe cases, it will cause accidents, endanger personal safety and cause large-scale power outages in the power grid.

[0003] When partial discharge occurs in power equipment, it is usually accompanied by physical phenomena such as electric pulses, ultrasonic waves, and electromagnetic radiation. There are many methods to detect partial discharge signals based on different physical phenomena. For the phenomenon of sound, ultrasonic detection is usually used. When partial discharge occurs, the frequency band of the emitted sound waves is very wide. The sound wave component with a frequency greater than 20Khz that cannot be recognized by the human ear is ultrasonic wave.

[0004] The ultrasonic detection method is to install an ultrasonic sensor on the equipment casing, convert the sound wave signal into a voltage signal, and obtain the characteristic quantity of the discharge signal, discharge size, etc. through analysis and processing.

[0005] The ultrasonic detection method is to install an ultrasonic sensor on the equipment casing, convert the sound wave signal into a voltage signal, and obtain the characteristic quantity of the discharge signal, discharge size, etc. through a series of complex analysis and processing.

[0006] However, the ultrasonic detection method is costly, has high environmental requirements, and is easily interfered by high-frequency signals. It is not an effective and economical choice under complex on-site operation conditions, and the detection process is complicated. If there is partial discharge in the area around the high-voltage conductor, the application of ultrasonic detection will result in low sensitivity and poor resistance to mechanical vibration interference. For example, the interference caused by natural wind at the detection site will greatly interfere with the detection itself. Summary of the invention

[0007] The present invention provides a partial discharge audio recognition method and system based on industrial frequency sine similarity, which solves the technical problems of low detection sensitivity and susceptibility to interference by high-frequency signals.

[0008] In view of this, a first aspect of the present invention provides a partial discharge audio recognition method based on power frequency sine similarity, comprising the following steps:

[0009] S1. Obtain the audio signal emitted by the device under test within a preset frequency band;

[0010] S2, preprocessing the audio signal to obtain a pure audio signal;

[0011] S3, judging whether the pure audio signal has a power frequency periodicity feature, if it is judged that the pure audio signal has a power frequency periodicity feature, executing the next step, if it is judged that the pure audio signal does not have a power frequency periodicity feature, ending;

[0012] S4, fitting the pure audio signal into an audio waveform, performing window segmentation processing on the audio waveform according to a preset phase interval to obtain a plurality of sub-waveforms, arranging the plurality of sub-waveforms in a time sequence to obtain a sub-waveform time sequence set;

[0013] S5, calculating the similarity between each of the sub-waveforms and a preset sinusoidal waveform to obtain a sub-waveform whose similarity is less than the preset similarity, and removing it from the sub-waveform time series set;

[0014] S6, performing discretization processing on each sub-waveform in the sub-waveform time series set obtained in step S5, so as to obtain a plurality of discrete points, and constructing a discrete point matrix according to corresponding positions of the plurality of discrete points, wherein each element in the discrete point matrix represents a discrete point and its coordinate position;

[0015] S7, calculating the maximum slope of the corresponding sub-waveform based on the discrete point matrix, and determining whether the maximum slope is greater than a preset slope, and if the maximum slope is greater than the preset slope, determining the corresponding sub-waveform as a pulse waveform;

[0016] S8. Obtain the phase corresponding to the pulse waveform. If the phase is within a preset phase range threshold, calculate the phase difference corresponding to the adjacent pulse waveforms, and determine whether all phase differences are within a preset difference range threshold. If it is determined that all phase differences are within the preset difference range threshold, and the number of pulse waveforms is greater than a preset number of waveforms, determine that the audio signal is a partial discharge audio signal.

[0017] Preferably, step S2 specifically includes:

[0018] The audio signal is filtered to obtain a pure audio signal.

[0019] Preferably, step S3 specifically includes:

[0020] S301, performing normalization processing on the pure audio signal to obtain a normalized audio signal;

[0021] S302, filtering out data points with an amplitude of 0 in the normalized audio signal to obtain a first data point set, and filtering out data points with one positive and one negative data points adjacent to each other on both sides from the first data point set to obtain a second data point set;

[0022] S303, taking each data point in the second data point set as a segmentation point, segmenting the normalized audio signal to obtain a plurality of data segments, calculating the cross-correlation coefficient between two adjacent data segments, and if all cross-correlation coefficients are greater than a preset coefficient, it is considered that the pure audio signal is a periodic signal feature;

[0023] S304, calculating the signal period according to the sampling frequency of the pure audio signal, if the signal period is within a preset power frequency period threshold, determining that the pure audio signal has power frequency periodicity characteristics, and executing the next step, if determining that the pure audio signal does not have power frequency periodicity characteristics, then ending.

[0024] In a second aspect, the present invention further provides a partial discharge audio recognition system based on power frequency sine similarity, comprising:

[0025] An acquisition module, used to acquire an audio signal emitted by the device under test within a preset frequency band;

[0026] A preprocessing module, used for preprocessing the audio signal to obtain a pure audio signal;

[0027] A power frequency determination module, used to determine whether the pure audio signal has a power frequency periodicity feature, if it is determined that the pure audio signal has a power frequency periodicity feature, then execute the next step, if it is determined that the pure audio signal does not have a power frequency periodicity feature, then end;

[0028] A waveform segmentation module, used for fitting the pure audio signal into an audio waveform, performing window segmentation processing on the audio waveform according to a preset phase interval to obtain a plurality of sub-waveforms, and arranging the plurality of sub-waveforms in a time sequence to obtain a sub-waveform time sequence set;

[0029] A calculation module, used for calculating the similarity between each of the sub-waveforms and a preset sine wave waveform, so as to obtain sub-waveforms whose similarity is less than the preset similarity, and remove them from the sub-waveform time series set;

[0030] A discrete module, used for discretizing each sub-waveform in the sub-waveform time series set to obtain a plurality of discrete points, and constructing a discrete point matrix according to corresponding positions of the plurality of discrete points, wherein each element in the discrete point matrix represents a discrete point and its coordinate position;

[0031] A judgment module, used for calculating the maximum slope of the corresponding sub-waveform based on the discrete point matrix, judging whether the maximum slope is greater than a preset slope, and if the maximum slope is greater than the preset slope, judging the corresponding sub-waveform as a pulse waveform;

[0032] An identification module is used to obtain the phase corresponding to the pulse waveform. If the phase is within a preset phase range threshold, the phase difference corresponding to the adjacent pulse waveforms is calculated to determine whether all phase differences are within a preset difference range threshold. If it is determined that all phase differences are within the preset difference range threshold and the number of pulse waveforms is greater than a preset number of waveforms, the audio signal is determined to be a partial discharge audio signal.

[0033] Preferably, the preprocessing module is specifically used to perform filtering processing on the audio signal to obtain a pure audio signal.

[0034] Preferably, the power frequency determination module specifically includes:

[0035] A normalization module, used for performing normalization processing on the pure audio signal to obtain a normalized audio signal;

[0036] A screening module is used to screen out data points with an amplitude of 0 in the normalized audio signal to obtain a first data point set, and screen out data points with two adjacent data points being one positive and one negative from the first data point set to obtain a second data point set;

[0037] a period determination module, configured to segment the normalized audio signal using each data point in the second data point set as a segmentation point, thereby obtaining a plurality of data segments, and calculating a mutual correlation coefficient between two adjacent data segments, and if all mutual correlation coefficients are greater than a preset coefficient, then the pure audio signal is considered to be a periodic signal feature;

[0038] The power frequency period determination module is used to calculate the signal period according to the sampling frequency of the pure audio signal, and if the signal period is within a preset power frequency period threshold, it is determined that the pure audio signal has a power frequency periodicity feature.

[0039] In a third aspect, the present invention further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0040] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0041] It can be seen from the above technical solutions that the present invention has the following advantages:

[0042] The present invention acquires and pre-processes audio signals within a specific frequency band emitted by a device under test to obtain a pure audio signal, performs window segmentation on an audio waveform fitted to the pure audio signal, arranges the sub-waveforms obtained by segmentation in time sequence, calculates the similarity between each sub-waveform and a preset sine wave waveform, thereby eliminating sub-waveforms with low similarity in a sub-waveform time series set, discretizes each sub-waveform, constructs a discrete point matrix according to corresponding positions of the discrete points obtained by discretization, calculates the maximum slope of the corresponding sub-waveform, determines a pulse waveform according to the maximum slope, and determines whether the audio signal is a partial discharge audio according to the phase position and phase difference of the pulse waveform and the number of pulse waveforms, thereby improving the sensitivity of partial discharge audio recognition and reducing the interference of high-frequency signals on the recognition result. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A flowchart of a partial discharge audio recognition method based on power frequency sine similarity provided by an embodiment of the present invention;

[0044] Figure 2 A structural schematic diagram of a partial discharge audio recognition system based on power frequency sine similarity provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] For easier understanding, see Figure 1 The present invention provides a partial discharge audio recognition method based on power frequency sine similarity, comprising the following steps:

[0047] S1. Obtain an audio signal emitted by the device under test within a preset frequency band.

[0048] In an example, the preset frequency band is 1Khz-20Khz, and the audio signal can be acquired in real time using a sound sensor and an AD acquisition module. The number of data points collected per second can be known from the sampling frequency, and the maximum amplitude and amplitude range of the data can be known from the data channel bit number of the AD acquisition module.

[0049] S2. Preprocess the audio signal to obtain a pure audio signal.

[0050] Specifically, the audio signal is filtered to obtain a pure audio signal.

[0051] S3, judging whether the pure audio signal has the periodicity characteristic of the power frequency, if it is judged that the pure audio signal has the periodicity characteristic of the power frequency, executing the next step, if it is judged that the pure audio signal does not have the periodicity characteristic of the power frequency, ending.

[0052] It should be noted that the audio data collected by the recording equipment includes not only the audio data generated by the voltage and current signals, but also various noises and murmurs. Partial discharge occurs when the voltage and current signals of the power frequency cycle are near the peak value. Therefore, the premise for determining whether partial discharge occurs is that the audio data must have power frequency periodicity. Since the voltage and current signals are power frequency periodic signals, it can be known that the measured audio signal should also have power frequency periodicity.

[0053] S4, fitting the pure audio signal into an audio waveform, performing window segmentation processing on the audio waveform according to a preset phase interval to obtain a plurality of sub-waveforms, arranging the plurality of sub-waveforms in time sequence to obtain a sub-waveform time sequence set.

[0054] In an example, the preset phase intervals are 0° to 72°, 108° to 250°, and 288° to 360°.

[0055] S5. Calculate the similarity between each sub-waveform and the preset sine wave waveform to obtain sub-waveforms with similarity less than the preset similarity, and remove them from the sub-waveform time series set.

[0056] It is understandable that in order to exclude other signals that also have industrial frequency periodicity, such as 50hz square wave signals, triangle wave signals, etc., which will affect the judgment of current and voltage signals that also have industrial frequency periodicity, it is necessary to perform a similarity analysis on the waveform of the audio signal and the standard sine wave. If the similarity meets the threshold condition, it meets the requirements for pulse recognition.

[0057] In one example, the preset amplitude of the sine wave waveform is 1, the frequency is 50 Hz, the interval is 0 to 360°, and the preset similarity is 0.75.

[0058] S6. Discretize each sub-waveform in the sub-waveform time series set obtained in step S5 to obtain a number of discrete points, and construct a discrete point matrix according to the corresponding positions of the discrete points. Each element in the discrete point matrix represents a discrete point and its coordinate position.

[0059] It is understandable that after the sub-waveform is discretized, the coordinate position of each discrete point can be obtained based on a pre-established coordinate system, and the corresponding discrete points can be constructed into a discrete point matrix according to their coordinate positions.

[0060] S7. Calculate the maximum slope of the corresponding sub-waveform based on the discrete point matrix, and determine whether the maximum slope is greater than a preset slope. If the maximum slope is greater than the preset slope, determine the corresponding sub-waveform as a pulse waveform.

[0061] It can be understood that the discrete point matrix is ​​composed of discrete points of the sub-waveform. The corresponding slope is calculated through the discrete points on the first row vector and the last row vector, and the maximum slope value is determined. If the maximum slope is greater than the preset slope, the corresponding sub-waveform is determined to be a pulse waveform. The pulse waveform is a representative waveform, where the preset slope is generally 1.3 to 1.5.

[0062] S8. Obtain the phase corresponding to the pulse waveform. If the phase is within the preset phase range threshold, calculate the phase difference corresponding to adjacent pulse waveforms, and determine whether all phase differences are within the preset difference range threshold. If it is determined that all phase differences are within the preset difference range threshold, and the number of pulse waveforms is greater than the preset number of waveforms, then determine that the audio signal is a partial discharge audio signal.

[0063] It should be noted that, based on historical experience, partial discharge is prone to occur near the phase of 90° and 270° of the industrial frequency sinusoidal periodic signal, and once it occurs, it will last for a period of time. Therefore, the preset phase range threshold can be set to 90° and 270°. If the phase corresponding to the pulse waveform is within the range of ±2° of 90° or 270°, and it is judged that the numerical value of the phase difference corresponding to adjacent pulse waveforms is not much different, it can be considered that the pulses are evenly distributed, and if the number of pulse waveforms is greater than 30 at this time, the audio signal can be considered to be a partial discharge audio signal.

[0064] In a specific embodiment, step S3 specifically includes:

[0065] S301 : performing normalization processing on a pure audio signal to obtain a normalized audio signal.

[0066] The normalization process is to limit the range of data values ​​to between -1 and 1, so that the range of data value changes can be clearly observed.

[0067] S302, filtering out data points with an amplitude of 0 in the normalized audio signal to obtain a first data point set, and filtering out data points with one positive and one negative data points adjacent to each other on both sides from the first data point set to obtain a second data point set.

[0068] S303, taking each data point in the second data point set as a segmentation point, segmenting the normalized audio signal to obtain a plurality of data segments, and calculating the mutual correlation coefficient between two adjacent data segments. If all the mutual correlation coefficients are greater than a preset coefficient, it is considered that the pure audio signal is a periodic signal feature.

[0069] S304, calculating the signal period according to the sampling frequency of the pure audio signal. If the signal period is within the preset power frequency period threshold, it is determined that the pure audio signal has the power frequency periodicity characteristic, and the next step is executed. If it is determined that the pure audio signal does not have the power frequency periodicity characteristic, the process ends.

[0070] It should be noted that the present invention provides a partial discharge audio recognition method based on industrial frequency sinusoidal similarity, which obtains a pure audio signal by collecting and preprocessing the audio signal in a specific frequency band emitted by the device to be tested, and performs window segmentation on the audio waveform fitted by the pure audio signal, and arranges the sub-waveforms obtained by segmentation in time sequence, calculates the similarity between each sub-waveform and a preset sinusoidal waveform, thereby eliminating sub-waveforms with low similarity in the sub-waveform time series set, and discretizes each sub-waveform, and constructs a discrete point matrix according to the corresponding positions of the discrete points, and calculates the maximum slope of the corresponding sub-waveform, determines the pulse waveform according to the maximum slope, and judges whether the audio signal is a partial discharge audio according to the phase position and phase difference of the pulse waveform and the number of pulse waveforms, thereby improving the sensitivity of partial discharge audio recognition and reducing the interference of high-frequency signals on the recognition results.

[0071] The above is an embodiment of a partial discharge audio recognition method based on power frequency sine similarity provided by the present invention. The following is an embodiment of a partial discharge audio recognition system based on power frequency sine similarity provided by the present invention.

[0072] See also Figure 2 The present invention provides a partial discharge audio recognition system based on power frequency sine similarity, comprising:

[0073] The acquisition module 100 is used to acquire the audio signal emitted by the device under test within a preset frequency band;

[0074] The preprocessing module 200 is used to preprocess the audio signal to obtain a pure audio signal;

[0075] The preprocessing module is specifically used to filter the audio signal to obtain a pure audio signal.

[0076] The power frequency determination module 300 is used to determine whether the pure audio signal has the power frequency periodicity feature. If it is determined that the pure audio signal has the power frequency periodicity feature, the next step is executed; if it is determined that the pure audio signal does not have the power frequency periodicity feature, the step ends;

[0077] The waveform segmentation module 400 is used to fit the pure audio signal into an audio waveform, perform window segmentation processing on the audio waveform according to a preset phase interval to obtain a plurality of sub-waveforms, and arrange the plurality of sub-waveforms in a time sequence to obtain a sub-waveform time sequence set;

[0078] A calculation module 500 is used to calculate the similarity between each sub-waveform and a preset sine wave waveform, so as to obtain sub-waveforms with a similarity less than the preset similarity, and remove them from the sub-waveform time series set;

[0079] The discrete module 600 is used to discretize each sub-waveform in the sub-waveform time series set to obtain a plurality of discrete points, and construct a discrete point matrix according to the corresponding positions of the plurality of discrete points, wherein each element in the discrete point matrix represents a discrete point and its coordinate position;

[0080] A determination module 700 is used to calculate the maximum slope of the corresponding sub-waveform based on the discrete point matrix, and determine whether the maximum slope is greater than a preset slope. If the maximum slope is greater than the preset slope, the corresponding sub-waveform is determined to be a pulse waveform;

[0081] The identification module 800 is used to obtain the phase corresponding to the pulse waveform. If the phase is within the preset phase range threshold, the phase difference corresponding to the adjacent pulse waveforms is calculated to determine whether all phase differences are within the preset difference range threshold. If it is determined that all phase differences are within the preset difference range threshold and the number of pulse waveforms is greater than the preset number of waveforms, the audio signal is determined to be a partial discharge audio signal.

[0082] In a specific embodiment, the power frequency determination module specifically includes:

[0083] A normalization module, used for normalizing the pure audio signal to obtain a normalized audio signal;

[0084] A screening module is used to screen out data points with an amplitude of 0 in the normalized audio signal to obtain a first data point set, and screen out data points with one positive and one negative data points adjacent to each other on both sides from the first data point set to obtain a second data point set;

[0085] a period determination module, for segmenting the normalized audio signal using each data point in the second data point set as a segmentation point, thereby obtaining a plurality of data segments, and calculating a mutual correlation coefficient between two adjacent data segments; if all mutual correlation coefficients are greater than a preset coefficient, the pure audio signal is considered to be a periodic signal feature;

[0086] The power frequency period judgment module is used to calculate the signal period according to the sampling frequency of the pure audio signal. If the signal period is within a preset power frequency period threshold, it is determined that the pure audio signal has the power frequency periodicity feature.

[0087] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0088] The present invention also provides a computer-readable storage medium on which a computer program is stored, characterized in that the computer program implements the steps of the above method when executed by a processor.

[0089] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0090] In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0091] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple 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.

[0092] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0093] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for executing all or part of the steps of the method described in each embodiment of the present invention through a computer device (which can be a personal computer, a server, or a network device, etc.). The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (full name in English: Read-Only Memory, English abbreviation: ROM), random access memory (full name in English: Random Access Memory, English abbreviation: RAM), disk or optical disk, etc. Various media that can store program codes.

[0094] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A partial discharge audio recognition method based on power frequency sine similarity, characterized in that: The following steps are involved: S1. Obtain the audio signal emitted by the device under test within a preset frequency band; S2, preprocessing the audio signal to obtain a pure audio signal; S3, judging whether the pure audio signal has a power frequency periodicity feature, if it is judged that the pure audio signal has a power frequency periodicity feature, executing the next step, if it is judged that the pure audio signal does not have a power frequency periodicity feature, ending; Specifically include: S301, performing normalization processing on the pure audio signal to obtain a normalized audio signal; S302, filtering out data points with an amplitude of 0 in the normalized audio signal to obtain a first data point set, and filtering out data points with one positive and one negative data points adjacent to each other on both sides from the first data point set to obtain a second data point set; S303, taking each data point in the second data point set as a segmentation point, segmenting the normalized audio signal to obtain a plurality of data segments, calculating the cross-correlation coefficient between two adjacent data segments, and if all cross-correlation coefficients are greater than a preset coefficient, it is considered that the pure audio signal is a periodic signal feature; S304, calculating the signal period according to the sampling frequency of the pure audio signal, if the signal period is within the preset power frequency period threshold, it is determined that the pure audio signal has the power frequency periodicity feature, and the next step is executed; if it is determined that the pure audio signal does not have the power frequency periodicity feature, the process ends; S4, fitting the pure audio signal into an audio waveform, performing window segmentation processing on the audio waveform according to a preset phase interval to obtain a plurality of sub-waveforms, arranging the plurality of sub-waveforms in a time sequence to obtain a sub-waveform time sequence set; S5, calculating the similarity between each of the sub-waveforms and a preset sinusoidal waveform to obtain a sub-waveform whose similarity is less than the preset similarity, and removing it from the sub-waveform time series set; S6, performing discretization processing on each sub-waveform in the sub-waveform time series set obtained in step S5, so as to obtain a plurality of discrete points, and constructing a discrete point matrix according to corresponding positions of the plurality of discrete points, wherein each element in the discrete point matrix represents a discrete point and its coordinate position; S7, calculating the maximum slope of the corresponding sub-waveform based on the discrete point matrix, and determining whether the maximum slope is greater than a preset slope, and if the maximum slope is greater than the preset slope, determining the corresponding sub-waveform as a pulse waveform; S8. Obtain the phase corresponding to the pulse waveform. If the phase is within a preset phase range threshold, calculate the phase difference corresponding to the adjacent pulse waveforms, and determine whether all phase differences are within a preset difference range threshold. If it is determined that all phase differences are within the preset difference range threshold, and the number of pulse waveforms is greater than a preset number of waveforms, determine that the audio signal is a partial discharge audio signal.

2. The method for identifying partial discharge audio based on power frequency sine similarity according to claim 1 is characterized in that: Step S2 specifically includes: The audio signal is filtered to obtain a pure audio signal.

3. A partial discharge audio recognition system based on power frequency sine similarity, characterized in that: include: An acquisition module, used to acquire an audio signal emitted by the device under test within a preset frequency band; A preprocessing module, used for preprocessing the audio signal to obtain a pure audio signal; A power frequency determination module, used to determine whether the pure audio signal has a power frequency periodicity feature, if it is determined that the pure audio signal has a power frequency periodicity feature, then execute the next step, if it is determined that the pure audio signal does not have a power frequency periodicity feature, then end; The power frequency determination module specifically includes: A normalization module, used for performing normalization processing on the pure audio signal to obtain a normalized audio signal; A screening module is used to screen out data points with an amplitude of 0 in the normalized audio signal to obtain a first data point set, and screen out data points with two adjacent data points being one positive and one negative from the first data point set to obtain a second data point set; a period determination module, configured to segment the normalized audio signal using each data point in the second data point set as a segmentation point, thereby obtaining a plurality of data segments, and calculating a mutual correlation coefficient between two adjacent data segments, and if all mutual correlation coefficients are greater than a preset coefficient, then the pure audio signal is considered to be a periodic signal feature; A power frequency period determination module, used to calculate the signal period according to the sampling frequency of the pure audio signal, and if the signal period is within a preset power frequency period threshold, determine that the pure audio signal has a power frequency periodicity feature; A waveform segmentation module, used for fitting the pure audio signal into an audio waveform, performing window segmentation processing on the audio waveform according to a preset phase interval to obtain a plurality of sub-waveforms, and arranging the plurality of sub-waveforms in a time sequence to obtain a sub-waveform time sequence set; A calculation module, used for calculating the similarity between each of the sub-waveforms and a preset sine wave waveform, so as to obtain sub-waveforms whose similarity is less than the preset similarity, and remove them from the sub-waveform time series set; A discrete module, used for discretizing each sub-waveform in the sub-waveform time series set to obtain a plurality of discrete points, and constructing a discrete point matrix according to corresponding positions of the plurality of discrete points, wherein each element in the discrete point matrix represents a discrete point and its coordinate position; A judgment module, used for calculating the maximum slope of the corresponding sub-waveform based on the discrete point matrix, judging whether the maximum slope is greater than a preset slope, and if the maximum slope is greater than the preset slope, judging the corresponding sub-waveform as a pulse waveform; An identification module is used to obtain the phase corresponding to the pulse waveform. If the phase is within a preset phase range threshold, the phase difference corresponding to the adjacent pulse waveforms is calculated to determine whether all phase differences are within a preset difference range threshold. If it is determined that all phase differences are within the preset difference range threshold and the number of pulse waveforms is greater than a preset number of waveforms, the audio signal is determined to be a partial discharge audio signal.

4. The partial discharge audio recognition system based on power frequency sine similarity according to claim 3 is characterized in that: The preprocessing module is specifically used to perform filtering processing on the audio signal to obtain a pure audio signal.

5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 2 are implemented.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 2 are implemented.

Images