Partial discharge detection method and device, electronic equipment and storage medium

Abstract

The invention provides a partial discharge detection method and apparatus, an electronic device and a storage medium. The method comprises the steps of obtaining a discharge sound signal of a to-be-detected power device; performing sliding window filtering on the discharge sound signal based on the discharge periodicity characteristic of the discharge sound signal to obtain a plurality of discharge sound signal components; partial discharge detection is carried out based on the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component, a partial discharge detection result is obtained, the defect that partial discharge detection cannot be accurately and effectively carried out due to environmental interference in a traditional scheme is overcome, and the partial discharge detection accuracy is improved by utilizing the mechanism characteristics of partial discharge. Partial discharge detection is carried out in combination with a sliding window filtering method, most interference factors in the environment can be inhibited, whether partial discharge exists or not can be accurately judged, an accurate and reliable partial discharge detection result is obtained, and the accuracy and credibility of detection are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid detection, and in particular to a partial discharge detection method, device, electronic equipment and storage medium. Background Art

[0002] As electricity demand grows, the safe and stable operation of power grid systems and high-voltage power equipment is of vital importance. However, insulation structure damage caused by aging equipment can cause partial discharge, which can lead to serious power accidents if left unchecked for a long time.

[0003] At present, the most commonly used partial discharge detection methods in industry are ultrasonic detection and ultra-high frequency detection. These methods perform partial discharge detection by collecting high-frequency signals when the equipment is running. However, in actual applications, the collected signals often contain a large amount of environmental interference. These interference signals will make the partial discharge detection process complicated, and the current signal processing methods cannot cope with this situation, that is, they cannot effectively eliminate environmental interference and it is difficult to achieve environmental interference suppression, resulting in the inability to accurately perform partial discharge detection. Summary of the invention

[0004] The present invention provides a partial discharge detection method, device, electronic device and storage medium, which are used to solve the defects in the prior art that a large amount of environmental interference exists and the interference cannot be effectively eliminated, resulting in a complicated partial discharge detection process and inaccurate detection results. The partial discharge detection is performed by using the mechanism characteristics of partial discharge, which can suppress most of the interference factors in the environment, accurately judge whether partial discharge occurs, and obtain accurate and reliable partial discharge detection results.

[0005] The present invention provides a partial discharge detection method, comprising: Acquiring a discharge sound signal of the electric power equipment to be detected; Based on the discharge periodicity characteristics of the discharge sound signal, performing sliding window filtering on the discharge sound signal to obtain multiple discharge sound signal components; Based on the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component, partial discharge detection is performed to obtain a partial discharge detection result.

[0006] According to a partial discharge detection method provided by the present invention, based on the discharge periodicity characteristics of the discharge sound signal, the discharge sound signal is subjected to sliding window filtering to obtain multiple discharge sound signal components, including: Based on the discharge periodicity characteristics, constructing a local discharge characteristic frequency; Based on the local discharge characteristic frequency, the discharge sound signal is subjected to sliding window filtering to obtain a plurality of discharge sound signal components.

[0007] According to a partial discharge detection method provided by the present invention, the partial discharge detection is performed based on the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain the partial discharge detection result, including: Performing a fast Fourier transform on each of the discharge sound signal components to obtain an envelope spectrum of each of the discharge sound signal components; Determining the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component based on the envelope spectrum of each discharge sound signal component; Selecting the local discharge characteristic frequency energy of the target frequency band from the local discharge characteristic frequency energy of each frequency band; Partial discharge detection is performed based on the partial discharge characteristic frequency energy of the target frequency band to obtain a partial discharge detection result.

[0008] According to a partial discharge detection method provided by the present invention, the partial discharge detection is performed based on the partial discharge characteristic frequency energy of the target frequency band to obtain the partial discharge detection result, including: Determine a partial discharge characteristic index based on the partial discharge characteristic frequency energy of the target frequency band; the partial discharge characteristic frequency energy of the target frequency band is the partial discharge characteristic frequency energy of a preset number of frequency bands when the partial discharge characteristic frequency energy is arranged in descending order; A partial discharge detection is performed based on the partial discharge characteristic index to obtain a partial discharge detection result.

[0009] According to a partial discharge detection method provided by the present invention, the partial discharge detection is performed based on the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain the partial discharge detection result, and then the method further includes: When the partial discharge detection result indicates that the power equipment to be detected has a partial discharge phenomenon, normalizing the partial discharge characteristic frequency energy of each frequency band to obtain a signal weight of each frequency band; Based on the signal weights of the frequency bands, signal reconstruction is performed on the discharge sound signal components to obtain a reconstructed sound signal; The partial discharge type is identified based on the reconstructed sound signal to obtain the partial discharge type of the electric power equipment to be detected.

[0010] According to a partial discharge detection method provided by the present invention, the partial discharge type identification is performed based on the reconstructed sound signal to obtain the partial discharge type of the power equipment to be detected, including: Respectively extracting time domain features and frequency domain features of the reconstructed sound signal to obtain a time domain partial discharge spectrum and a frequency domain Fourier transform feature; Based on the time-domain partial discharge spectrum and the frequency-domain Fourier transform characteristics, partial discharge type identification is performed to obtain the partial discharge type of the power equipment to be detected.

[0011] According to a partial discharge detection method provided by the present invention, the partial discharge type identification is performed based on the time domain partial discharge spectrum and the frequency domain Fourier transform characteristics to obtain the partial discharge type of the power equipment to be detected, including: Performing feature fusion on the time-domain partial discharge spectrum and the frequency-domain Fourier transform feature to obtain a fused feature spectrum; Based on the fused characteristic spectrum, partial discharge type identification is performed to obtain the partial discharge type of the power equipment to be detected.

[0012] The present invention also provides a partial discharge detection device, comprising: A signal acquisition unit, used to acquire a discharge sound signal of the electric power equipment to be detected; A sliding window filtering unit, configured to perform sliding window filtering on the discharge sound signal based on the discharge periodicity characteristics of the discharge sound signal to obtain a plurality of discharge sound signal components; The discharge detection unit is used to perform partial discharge detection based on the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain a partial discharge detection result.

[0013] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements any of the above-mentioned partial discharge detection methods when executing the computer program.

[0014] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the partial discharge detection method as described above is implemented.

[0015] The local discharge detection method, device, electronic device and storage medium provided by the present invention obtain the discharge sound signal of the power equipment to be detected, perform sliding window filtering on the discharge sound signal according to the discharge periodicity characteristics of the discharge sound signal, obtain multiple discharge sound signal components, and perform local discharge detection according to the local discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain a local discharge detection result, thereby overcoming the defect of the traditional scheme that the local discharge detection cannot be accurately and effectively performed due to environmental interference, utilizing the mechanism characteristics of the local discharge and combining the sliding window filtering method to perform local discharge detection, which can suppress most of the interference factors in the environment, accurately judge whether there is local discharge, obtain accurate and reliable local discharge detection results, and improve the accuracy and credibility of the detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0017] Figure 1 It is a flow chart of the partial discharge detection method provided by the present invention; Figure 2 It is a schematic diagram of the framework of the partial discharge detection process provided by the present invention; Figure 3 It is an overall framework diagram of the partial discharge detection method provided by the present invention; Figure 4 It is a structural schematic diagram of a partial discharge detection device provided by the present invention; Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are 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.

[0019] With the sustained and rapid development of modern industry and economy, the demand for electricity in society is increasing. In order to ensure the steady operation of industry and economy and the stability and order of people's lives, the sufficient and stable power supply has become an indispensable key factor. However, the power grid system and high-voltage power equipment will inevitably age during long-term use, and their insulation structure will also be damaged due to various reasons, such as mechanical stress and environmental factors. Although electrical equipment can continue to operate in the case of partial damage to the insulation structure, non-penetrating breakdown often occurs at these damaged parts, resulting in the occurrence of partial discharge.

[0020] Partial discharge is an important sign of insulation degradation in power equipment. If the equipment is in this state for a long time, it will aggravate the damage of the insulation structure and eventually lead to penetrating breakdown, which may cause serious power accidents, such as large-scale power outages or even fires, causing immeasurable losses to society and enterprises. Therefore, monitoring and analyzing partial discharge of high-voltage electrical equipment is an important means to ensure the safe and stable operation of the power grid system and high-voltage power equipment.

[0021] In the process of partial discharge, a variety of physical and chemical phenomena will be generated, such as ultrasound, chemical reaction, optical signal, pulse current and electromagnetic radiation. At present, the detection technologies for these phenomena mainly include ultrasonic detection method, chemical detection method, optical detection method, pulse current detection method, ultra-high frequency detection method, etc. Ultrasonic and ultra-high frequency detection methods are widely used in industry. These methods detect partial discharge by collecting high-frequency signals when the power equipment is running.

[0022] However, in the actual detection process, the collected signals often contain a lot of environmental interference, which makes the detection of partial discharge complicated and difficult. The current signal processing methods cannot effectively deal with this situation. That is, the current signal processing methods, including filtering methods, modal decomposition methods, wavelet transform methods, etc., although they can remove environmental interference signals to a certain extent, they all have their own limitations. Specifically, the local discharge characteristics are generally in the high frequency band, but the environmental interference may also be in the high frequency band. Therefore, the interference signal cannot be removed by filtering methods, and the interference removal effect is not good; the modal decomposition method has problems such as modal overlap, endpoint effect and high computational complexity; the wavelet transform rule needs to select a suitable wavelet basis function, otherwise it is impossible to accurately distinguish the frequency information of each component in the signal. In short, the current method cannot effectively eliminate the interference factors in the environment, resulting in difficult and inaccurate partial discharge detection.

[0023] In this regard, the present invention provides a local discharge detection method, which aims to utilize the mechanism characteristics of local discharge and adopt an adaptive sliding window filtering method to calculate the characteristic frequency energy of local discharge, and perform local discharge detection based on this to achieve accurate judgment of local discharge. It solves the problem in traditional schemes that local discharge detection cannot be accurately and effectively performed due to strong environmental interference, achieves effective suppression of most interference factors in the environment, and improves detection accuracy.

[0024] Figure 1 is a flow chart of the partial discharge detection method provided by the present invention, such as Figure 1 As shown, the method includes: Step 110, obtaining a discharge sound signal of the electric power equipment to be detected; Step 120, based on the discharge periodicity characteristics of the discharge sound signal, performing sliding window filtering on the discharge sound signal to obtain multiple discharge sound signal components; Step 130 , performing partial discharge detection based on the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain a partial discharge detection result.

[0025] Specifically, considering that in the current partial discharge detection process, the collected signals contain various environmental interferences, which makes the detection process complicated and difficult, and the detection results are not accurate, in the embodiments of the present invention, when performing partial discharge detection, the primary consideration is also the environmental interference problem, that is, how to remove environmental interference and effectively extract partial discharge characteristics to achieve accurate judgment of partial discharge. Different from the signal processing methods used in traditional solutions, namely filtering methods, modal decomposition methods, wavelet transform methods, etc., in the embodiments of the present invention, starting from the mechanism of partial discharge, partial discharge detection is performed in combination with the mechanism characteristics of partial discharge, which can not only effectively suppress a large number of interference factors in the environment, but also improve the accuracy of detection.

[0026] It is understandable that in actual applications, before performing partial discharge detection, it is first necessary to determine the power equipment to be detected and the signal used to detect the equipment. The power equipment to be detected here can be a high-voltage power equipment that is being put into use in the power grid system, which can be one or more units, and the embodiment of the present invention does not specifically limit this. After determining the power equipment to be detected, in order to detect and analyze the partial discharge phenomenon of the high-voltage power equipment, in the embodiment of the present invention, the discharge sound signal of the power equipment can be obtained. The discharge sound signal is the sound signal generated during its operation. These sound signals may include sound waves, ultrasonic waves, etc. generated during the discharge process. By analyzing these sound signals, important information about partial discharge can be obtained, such as the location, intensity, frequency, etc. of the discharge.

[0027] Furthermore, partial discharge detection can be performed on the power equipment to be detected based on the acquired discharge sound signal to determine whether partial discharge occurs, thereby obtaining partial discharge detection results. In actual detection, considering that the partial discharge phenomenon is periodic, periodic feature extraction can be performed in the embodiment of the present invention to obtain discharge periodic features. Since partial discharge is complex, its discharge periodicity may be affected by AC voltage, internal defects, the performance of insulating materials, etc. Therefore, when extracting periodic features, the discharge sound signal can be analyzed by frequency, voltage, etc., so as to obtain the periodic features of partial discharge through analysis, thereby obtaining discharge periodic features.

[0028] Next, the discharge sound signal can be processed according to the discharge periodicity characteristics to remove environmental interference and extract signal components representing partial discharge. Here, the discharge sound signal is specifically subjected to sliding window filtering using the discharge periodicity characteristics to obtain multiple discharge sound signal components; that is, under the guidance of the discharge periodicity characteristics, the discharge sound signal is subjected to sliding window filtering according to a certain bandwidth to obtain signal components corresponding to each bandwidth window, that is, multiple discharge sound signal components.

[0029] After that, the energy index of the corresponding frequency band can be calculated according to each discharge sound signal component, so as to obtain the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component; here, specifically, the signal strength or signal energy distribution in each frequency band is calculated according to the signal component obtained by sliding window filtering, that is, the energy index (Partial discharge characteristic frequency energy index) of each frequency band is calculated, and the energy index can reflect the signal strength or signal energy distribution in the corresponding frequency band, so as to obtain the partial discharge characteristic frequency energy of each frequency band. Each signal component corresponds to a frequency band, and the partial discharge characteristic frequency energy of each frequency band can be a specific energy value (such as calculating the mean square value and sum value of the signal energy in the frequency band), or it can be an energy distribution, which is not specifically limited in the embodiment of the present invention.

[0030] Furthermore, local discharge detection can be performed based on this local discharge characteristic frequency energy to obtain local discharge detection results; that is, local discharge-related information can be effectively identified based on the local discharge characteristic frequency energy of each frequency band, such as identifying and distinguishing local discharge signals from interference signals, so that local discharge detection can be performed on power equipment to obtain accurate local discharge detection results, thereby achieving accurate and effective local discharge monitoring and discharge abnormality diagnosis of power equipment, as well as monitoring of the degree of insulation structure degradation, which is conducive to the safe and stable operation of the power grid system, as well as the preventive maintenance and fault prediction of high-voltage power equipment.

[0031] In the embodiment of the present invention, the mechanism characteristics of partial discharge are utilized, combined with an adaptive sliding window filtering method, the characteristic frequency energy of partial discharge in different frequency bands is traversed, and partial discharge detection is performed accordingly, so that most interference factors in the environment can be suppressed, and whether partial discharge occurs can be accurately determined. Compared with traditional signal denoising methods, it is simpler and more efficient, has higher accuracy, stronger practicality and generalization, and has a certain degree of interpretability. It provides a practical solution to the urgent and thorny problem of partial discharge detection in the current power grid system, that is, the problem of a large amount of environmental interference in the signal.

[0032] The local discharge detection method provided by the present invention performs sliding window filtering on the discharge sound signal according to the discharge periodicity characteristics of the discharge sound signal of the power equipment to obtain multiple discharge sound signal components, and performs local discharge detection according to the local discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain a local discharge detection result, thereby overcoming the defect of the traditional scheme that the local discharge detection cannot be accurately and effectively performed due to environmental interference. The local discharge detection is performed by utilizing the mechanism characteristics of the local discharge in combination with the sliding window filtering method, which can suppress most of the interference factors in the environment, accurately judge whether there is local discharge, obtain accurate and reliable local discharge detection results, and improve the accuracy and credibility of the detection.

[0033] Based on the above embodiment, step 120 includes: Based on the periodic characteristics of discharge, the characteristic frequency of partial discharge is constructed; Based on the local discharge characteristic frequency, the discharge sound signal is subjected to sliding window filtering to obtain multiple discharge sound signal components.

[0034] Specifically, in step 120, the process of performing sliding window filtering on the discharge sound signal according to the discharge periodicity characteristics of the discharge sound signal to obtain multiple discharge sound signal components may specifically include the following steps: First, the frequency range can be determined based on the discharge periodicity characteristics of the discharge sound signal obtained by analyzing the discharge sound signal to obtain the local discharge characteristic frequency; here, specifically in order to effectively suppress the interference signal in the discharge sound signal and extract the characteristics of the local discharge for subsequent detection, the energy index can be constructed in the embodiment of the present invention, that is, the local discharge characteristic frequency energy is proposed, and before calculating this, one or more characteristic frequency ranges must be determined based on the periodicity characteristics of the local discharge, so as to obtain the local discharge characteristic frequency. These ranges can be obtained by understanding the local discharge characteristics and analyzing the signal.

[0035] Next, the discharge sound signal can be subjected to sliding window filtering according to the local discharge frequency to obtain multiple discharge sound signal components; here, specifically, the discharge sound signal can be subjected to sliding window filtering according to a certain bandwidth based on the local discharge characteristic frequency, that is, the discharge sound signal is subjected to sliding window filtering according to a certain bandwidth in the frequency dimension to obtain signal components corresponding to each bandwidth window, that is, multiple discharge sound signal components.

[0036] Based on the above embodiment, step 130 includes: Performing a fast Fourier transform on each discharge sound signal component to obtain an envelope spectrum of each discharge sound signal component; Determining the local discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component based on the envelope spectrum of each discharge sound signal component; Selecting the local discharge characteristic frequency energy of the target frequency band from the local discharge characteristic frequency energy of each frequency band; The partial discharge detection is performed based on the partial discharge characteristic frequency energy in the target frequency band to obtain the partial discharge detection result.

[0037] Specifically, in step 130, the process of performing partial discharge detection according to the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain the partial discharge detection result may specifically include: Figure 2 FIG. 1 is a schematic diagram of the framework of the partial discharge detection process provided by the present invention, such as Figure 2 As shown, after multiple sound signal components are obtained through sliding window filtering, in an embodiment of the present invention, the envelope spectrum of each signal component can be solved by fast Fourier transform. Here, specifically, the envelope of each discharge sound signal component is subjected to fast Fourier transform (FastFourier Transform) to obtain the envelope spectrum of each discharge sound signal component.

[0038] Next, the local discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component can be calculated based on the envelope spectrum obtained in the previous step; that is, the energy index can be calculated based on the envelope spectrum of each discharge sound signal component and the local discharge characteristic frequency. Specifically, the signal strength or signal energy distribution in each frequency band can be calculated to obtain the local discharge characteristic frequency energy of each frequency band.

[0039] Then, the local discharge characteristic frequency energy of the target frequency band can be selected from the local discharge characteristic frequency energies of each frequency band; that is, representative frequency bands, such as those with large energy and characteristic energy distribution, are selected from the local discharge characteristic frequency energies of each frequency band obtained by calculation, and the local discharge characteristic frequency energies of these frequency bands are used as data actually used in subsequent detection, namely, the local discharge characteristic frequency energy of the target frequency band.

[0040] Afterwards, partial discharge detection can be performed based on the partial discharge characteristic frequency energy of these target frequency bands to obtain partial discharge detection results; that is, partial discharge-related information can be effectively identified based on the partial discharge characteristic frequency energy of each frequency band, thereby performing partial discharge detection on the power equipment and obtaining accurate partial discharge detection results.

[0041] Based on the above embodiment, the process of performing fast Fourier transform on each discharge sound signal component can be expressed by the following formula: In the formula, is the envelope of each discharge sound signal component, represents the fast Fourier transform, is the envelope spectrum of each discharge sound signal component.

[0042] Based on the above embodiment, according to the envelope spectrum of each discharge sound signal component, the process of determining the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component can be expressed by the following formula: In the formula, represents the characteristic frequency of partial discharge, is the harmonic order of the partial discharge characteristic frequency, Indicates the characteristic frequency energy of partial discharge in the corresponding frequency band.

[0043] Based on the above embodiment, partial discharge detection is performed based on the partial discharge characteristic frequency energy in the target frequency band to obtain partial discharge detection results, including: Based on the partial discharge characteristic frequency energy of the target frequency band, the partial discharge characteristic index is determined; the partial discharge characteristic frequency energy of the target frequency band is the partial discharge characteristic frequency energy of the first preset number of frequency bands when the partial discharge characteristic frequency energy is arranged in descending order; Partial discharge detection is performed based on the partial discharge characteristic index to obtain a partial discharge detection result.

[0044] Specifically, the process of performing partial discharge detection according to the partial discharge characteristic frequency energy in the target frequency band to obtain the partial discharge detection result may specifically include: Firstly, according to the local discharge characteristic frequency energy of the selected target frequency band, the judgment index is calculated, that is, a larger preset number of local discharge characteristic frequency energies among the local discharge characteristic frequency energies of all frequency bands are used as the local discharge characteristic frequency energies of the target frequency band, and the mean, median and other calculations are performed on the local discharge characteristic frequency energies of these target frequency bands, so as to obtain the local discharge characteristic index.

[0045] Subsequently, partial discharge detection can be performed based on this partial discharge characteristic index, thereby obtaining a partial discharge detection result; here, specifically, the partial discharge characteristic index obtained by the previous step can be compared with the threshold value set with reference to the characteristic index of non-partial discharge, thereby determining whether the power equipment to be detected has a partial discharge phenomenon. Specifically, when the partial discharge characteristic index is greater than or equal to the threshold value, it is judged as a partial discharge phenomenon; conversely, when the partial discharge characteristic index is less than the threshold value, it is judged as no partial discharge phenomenon, that is, normal, and accurate partial discharge detection of power equipment can be achieved by comparing the threshold values.

[0046] Based on the above embodiment, in step 130, partial discharge detection is performed based on the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain a partial discharge detection result, and then the following is further included: When the partial discharge detection result indicates that the power equipment to be detected has a partial discharge phenomenon, normalizing the partial discharge characteristic frequency energy of each frequency band to obtain a signal weight of each frequency band; Based on the signal weight of each frequency band, each discharge sound signal component is reconstructed to obtain a reconstructed sound signal; The partial discharge type is identified based on the reconstructed sound signal to obtain the partial discharge type of the power equipment to be detected.

[0047] Specifically, after performing partial discharge detection based on the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component, if it is determined that the power equipment stores partial discharge, the type of partial discharge can be further identified to achieve accurate detection.

[0048] Specifically, when the partial discharge detection result indicates that the power equipment to be detected has partial discharge phenomenon, the partial discharge type of the power equipment can be identified to obtain the partial discharge type of the power equipment to be detected. However, considering that the feature extraction of partial discharge is susceptible to environmental interference during actual detection and identification, it is difficult to effectively extract the features of partial discharge in a complex environment, and the extracted features will also be seriously "contaminated", resulting in the inability to accurately judge the type of partial discharge.

[0049] In view of this, in an embodiment of the present invention, it is proposed that signal reconstruction can be performed to extract as many components containing local discharge characteristics as possible and suppress interference components, thereby improving signal quality, and then accurately identifying and judging the type of local discharge based on this. Specifically, here, after sliding window filtering and energy index calculation, the local discharge characteristic frequency energy of each frequency band can be normalized to calculate the signal weight of each frequency band; then, according to the signal weight of each frequency band, multiple discharge sound signal components can be reconstructed to extract as many signals reflecting local discharge characteristics as possible and suppress interference signals, thereby obtaining a reconstructed discharge sound signal, that is, a reconstructed sound signal.

[0050] After that, the partial discharge type can be identified based on the reconstructed sound signal, thereby obtaining the partial discharge type of the power equipment to be detected. Here, the partial discharge feature extraction can be performed based on the reconstructed sound signal, which can avoid the interference of environmental factors, obtain accurate, representative and recognizable features related to partial discharge, and the partial discharge type can be identified based on this feature, thereby obtaining the partial discharge type of the power equipment.

[0051] In the embodiment of the present invention, components containing local discharge characteristics are extracted as much as possible through signal reconstruction, thereby effectively suppressing other interference signals in a complex environment. Compared with traditional signal denoising methods, this method is simpler, more efficient, more accurate and reliable.

[0052] Based on the above embodiment, the partial discharge type is identified based on the reconstructed sound signal to obtain the partial discharge type of the power equipment to be detected, including: The reconstructed sound signal is subjected to time domain feature extraction and frequency domain feature extraction respectively to obtain the time domain partial discharge spectrum and the frequency domain Fourier transform feature; Based on the partial discharge spectrum in the time domain and the Fourier transform characteristics in the frequency domain, the partial discharge type is identified to obtain the partial discharge type of the power equipment to be detected.

[0053] Specifically, the process of identifying the partial discharge type according to the reconstructed sound signal to obtain the partial discharge type of the power equipment to be detected specifically includes the following steps: First, feature extraction can be performed on the reconstructed sound signal to extract features related to partial discharge. Specifically, feature extraction can be performed from two dimensions: time domain and frequency domain. That is, time domain feature extraction and frequency domain feature extraction are performed on the reconstructed sound signal respectively to extract phase resolved partial discharge PRPD (Phase Resolved PartialDischarge) features in the time domain and short-time Fourier transform STFT (Short-Time Fourier Transform) features in the frequency domain, thereby obtaining a time domain partial discharge spectrum and a frequency domain Fourier transform feature.

[0054] Next, the partial discharge type can be identified based on the time-domain partial discharge spectrum and the frequency-domain Fourier transform characteristics, thereby obtaining the partial discharge type of the power equipment; here, specifically, the time-domain partial discharge spectrum and the frequency-domain Fourier transform characteristics can be deeply analyzed based on the phase distribution unique to different partial discharge signals, so as to comprehensively analyze and determine the partial discharge type of the power equipment to be detected, thereby achieving accurate partial discharge type identification.

[0055] Here, the process of partial discharge type recognition can be implemented by a discharge type recognition model, that is, the time domain partial discharge spectrum and the frequency domain Fourier transform features can be input into the discharge type recognition model, so that the discharge type recognition model can recognize the partial discharge type according to the input features, and output the recognition results to obtain the partial discharge type. Among them, the discharge type recognition model can be constructed on the basis of a convolutional neural network based on multi-classification, and by utilizing the significant advantages of convolutional neural networks in image classification tasks, it can accurately perform multi-classification of partial discharge types and obtain accurate partial discharge types.

[0056] In the embodiment of the present invention, the time domain and the frequency domain are combined to extract the features of the time domain and the frequency domain respectively. The advantages of the time domain partial discharge spectrum, such as simplicity, small amount of data, good stability, and high requirements on monitoring conditions, and the advantages of the frequency domain Fourier transform features in time-frequency analysis are utilized. The partial discharge type is identified through the discharge type identification model, and the detailed classification of the partial discharge type can be identified with greater robustness and higher accuracy.

[0057] Based on the above embodiment, based on the time domain partial discharge spectrum and the frequency domain Fourier transform characteristics, the partial discharge type is identified to obtain the partial discharge type of the power equipment to be detected, including: Perform feature fusion on the time-domain partial discharge spectrum and the frequency-domain Fourier transform features to obtain a fused feature spectrum; Based on the fusion characteristic spectrum, partial discharge type identification is performed to obtain the partial discharge type of the power equipment to be detected.

[0058] Specifically, the process of identifying the partial discharge type based on the partial discharge spectrum in the time domain and the Fourier transform characteristics in the frequency domain to obtain the partial discharge type may specifically include: Since the traditional one-dimensional PRPD features have been converted into two-dimensional PRPD features during time domain feature extraction, compared with one-dimensional features, in the embodiment of the present invention, the time dimension is added to make the physical meaning clearer, the local discharge features are easier to capture, and it is more conducive to time-frequency analysis to obtain accurate and reliable local discharge types. On this basis, in order to make the advantages of local discharge type identification more obvious and more adaptable in complex environments, in the embodiment of the present invention, the time domain local discharge spectrum and the frequency domain Fourier transform features are selected for feature fusion to obtain a fused feature spectrum; the feature fusion here can be addition, splicing, etc. For example, the two-dimensional PRPD features can be spliced ​​with the STFT features to obtain a PRPD-STFT two-dimensional feature spectrum.

[0059] After that, the local discharge type can be identified based on the fused feature map to obtain the local discharge type of the power equipment to be detected, that is, the local discharge type can be identified based on the PRPD-STFT two-dimensional feature map. Specifically, the two-dimensional feature map can be input into the discharge type identification model to obtain the local discharge type of the power equipment output by the model.

[0060] In the embodiment of the present invention, the time-domain partial discharge spectrum and the frequency-domain Fourier transform features are feature fused to obtain a fused feature spectrum, and the partial discharge type is identified based on this. This not only utilizes the advantage of the PRPD feature's clear physical meaning, but also utilizes the advantage of the STFT feature in time-frequency analysis, so that the short-term characteristics of partial discharge can be accurately captured, and it is more applicable in complex and changeable environments.

[0061] Figure 3 is the overall framework diagram of the partial discharge detection method provided by the present invention, such as Figure 3 As shown, the method includes: First, a discharge sound signal of the electric power equipment to be detected is obtained.

[0062] Then, based on the discharge periodicity characteristics of the discharge sound signal, a local discharge characteristic frequency is constructed, and based on the local discharge characteristic frequency, the discharge sound signal is subjected to sliding window filtering to obtain multiple discharge sound signal components.

[0063] Subsequently, each discharge sound signal component is subjected to a fast Fourier transform to obtain an envelope spectrum of each discharge sound signal component, and based on the envelope spectrum of each discharge sound signal component, the local discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component is determined.

[0064] Afterwards, the local discharge characteristic frequency energy of the target frequency band is selected from the local discharge characteristic frequency energies of each frequency band; the local discharge characteristic frequency energy of the target frequency band is the local discharge characteristic frequency energies of a preset number of frequency bands when the local discharge characteristic frequency energies are arranged in descending order.

[0065] Then, based on the partial discharge characteristic frequency energy of the target frequency band, a partial discharge characteristic index is determined, and partial discharge detection is performed based on the partial discharge characteristic index to obtain a partial discharge detection result.

[0066] Furthermore, when the partial discharge detection result indicates that the electric power equipment to be detected has a partial discharge phenomenon, the partial discharge characteristic frequency energy of each frequency band is normalized to obtain a signal weight of each frequency band.

[0067] Next, based on the signal weight of each frequency band, each discharge sound signal component is reconstructed to obtain a reconstructed sound signal; time domain feature extraction and frequency domain feature extraction are performed on the reconstructed sound signal to obtain a time domain partial discharge spectrum and a frequency domain Fourier transform feature.

[0068] Then, the time-domain partial discharge spectrum and the frequency-domain Fourier transform features are fused to obtain a fused feature spectrum, and the partial discharge type is identified based on the fused feature spectrum to obtain the partial discharge type of the power equipment to be detected.

[0069] In the embodiment of the present invention, the partial discharge monitoring of the power equipment adopts a two-level monitoring mechanism. The first level adopts the mechanism characteristics of partial discharge for binary classification, and the second level performs multi-classification based on the time domain partial discharge spectrum and the frequency domain Fourier transform characteristics. Compared with the traditional scheme, it has the advantages of higher accuracy, lower false alarm rate, and stronger practicality.

[0070] The partial discharge detection device provided by the present invention is described below. The partial discharge detection device described below and the partial discharge detection method described above can be referred to each other.

[0071] Figure 4 Schematic diagram of the structure of the partial discharge detection device provided by the present invention. Figure 4 As shown, the device comprises: The signal acquisition unit 410 is used to acquire the discharge sound signal of the power equipment to be detected; A sliding window filtering unit 420, configured to perform sliding window filtering on the discharge sound signal based on the discharge periodicity characteristics of the discharge sound signal to obtain a plurality of discharge sound signal components; The discharge detection unit 430 is used to perform partial discharge detection based on the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain a partial discharge detection result.

[0072] The local discharge detection device provided by the present invention performs sliding window filtering on the discharge sound signal according to the discharge periodicity characteristics of the discharge sound signal of the power equipment to obtain multiple discharge sound signal components, and performs local discharge detection according to the local discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain a local discharge detection result, thereby overcoming the defect of the traditional scheme that the local discharge detection cannot be accurately and effectively performed due to environmental interference. The local discharge detection is performed by utilizing the mechanism characteristics of the local discharge in combination with the sliding window filtering method, which can suppress most of the interference factors in the environment, accurately judge whether there is local discharge, obtain accurate and reliable local discharge detection results, and improve the accuracy and credibility of the detection.

[0073] Based on the above embodiment, the sliding window filtering unit 420 is used for: Based on the discharge periodicity characteristics, constructing a local discharge characteristic frequency; Based on the local discharge characteristic frequency, the discharge sound signal is subjected to sliding window filtering to obtain a plurality of discharge sound signal components.

[0074] Based on the above embodiment, the discharge detection unit 430 is used for: Performing a fast Fourier transform on each of the discharge sound signal components to obtain an envelope spectrum of each of the discharge sound signal components; Determining the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component based on the envelope spectrum of each discharge sound signal component; Selecting the local discharge characteristic frequency energy of the target frequency band from the local discharge characteristic frequency energy of each frequency band; Partial discharge detection is performed based on the partial discharge characteristic frequency energy of the target frequency band to obtain a partial discharge detection result.

[0075] Based on the above embodiment, the discharge detection unit 430 is used for: Determine a partial discharge characteristic index based on the partial discharge characteristic frequency energy of the target frequency band; the partial discharge characteristic frequency energy of the target frequency band is the partial discharge characteristic frequency energy of a preset number of frequency bands when the partial discharge characteristic frequency energy is arranged in descending order; A partial discharge detection is performed based on the partial discharge characteristic index to obtain a partial discharge detection result.

[0076] Based on the above embodiment, the device further includes a discharge type identification unit, which is used to: The partial discharge detection is performed based on the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain the partial discharge detection result, and then the following further comprises: When the partial discharge detection result indicates that the power equipment to be detected has a partial discharge phenomenon, normalizing the partial discharge characteristic frequency energy of each frequency band to obtain a signal weight of each frequency band; Based on the signal weights of the frequency bands, signal reconstruction is performed on the discharge sound signal components to obtain a reconstructed sound signal; The partial discharge type is identified based on the reconstructed sound signal to obtain the partial discharge type of the electric power equipment to be detected.

[0077] Based on the above embodiment, the discharge type identification unit is used for: Respectively extracting time domain features and frequency domain features of the reconstructed sound signal to obtain a time domain partial discharge spectrum and a frequency domain Fourier transform feature; Based on the time-domain partial discharge spectrum and the frequency-domain Fourier transform characteristics, partial discharge type identification is performed to obtain the partial discharge type of the power equipment to be detected.

[0078] Based on the above embodiment, the discharge type identification unit is used for: Performing feature fusion on the time-domain partial discharge spectrum and the frequency-domain Fourier transform feature to obtain a fused feature spectrum; Based on the fused characteristic spectrum, partial discharge type identification is performed to obtain the partial discharge type of the power equipment to be detected.

[0079] Figure 5 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5As shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530 and a communication bus 540, wherein the processor 510, the communication interface 520 and the memory 530 communicate with each other through the communication bus 540. The processor 510 may call the logic instructions in the memory 530 to execute the partial discharge detection method, which includes: obtaining a discharge sound signal of the power equipment to be detected; based on the discharge periodicity characteristics of the discharge sound signal, performing sliding window filtering on the discharge sound signal to obtain multiple discharge sound signal components; performing partial discharge detection based on the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain a partial discharge detection result.

[0080] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0081] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the local discharge detection method provided by the above-mentioned methods, and the method includes: obtaining a discharge sound signal of the power equipment to be detected; based on the discharge periodic characteristics of the discharge sound signal, performing sliding window filtering on the discharge sound signal to obtain multiple discharge sound signal components; performing local discharge detection based on the local discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain a local discharge detection result.

[0082] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it is implemented to execute the local discharge detection method provided by the above-mentioned methods, the method comprising: obtaining a discharge sound signal of the power equipment to be detected; based on the discharge periodic characteristics of the discharge sound signal, performing sliding window filtering on the discharge sound signal to obtain multiple discharge sound signal components; performing local discharge detection based on the local discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain a local discharge detection result.

[0083] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0084] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0085] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. 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 detection method, characterized in that: include: Acquiring a discharge sound signal of the electric power equipment to be detected; Based on the discharge periodicity characteristics of the discharge sound signal, performing sliding window filtering on the discharge sound signal to obtain multiple discharge sound signal components; Based on the local discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component, local discharge detection is performed to obtain a local discharge detection result.

2. The partial discharge detection method according to claim 1, characterized in that: The step of performing sliding window filtering on the discharge sound signal based on the discharge periodicity characteristics of the discharge sound signal to obtain multiple discharge sound signal components includes: Based on the discharge periodicity characteristics, constructing a local discharge characteristic frequency; Based on the local discharge characteristic frequency, the discharge sound signal is subjected to sliding window filtering to obtain a plurality of discharge sound signal components.

3. The partial discharge detection method according to claim 1, characterized in that: The partial discharge detection is performed based on the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain the partial discharge detection result, including: Performing a fast Fourier transform on each of the discharge sound signal components to obtain an envelope spectrum of each of the discharge sound signal components; Determining the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component based on the envelope spectrum of each discharge sound signal component; Selecting the local discharge characteristic frequency energy of the target frequency band from the local discharge characteristic frequency energy of each frequency band; Partial discharge detection is performed based on the partial discharge characteristic frequency energy of the target frequency band to obtain a partial discharge detection result.

4. The partial discharge detection method according to claim 3, characterized in that: The performing partial discharge detection based on the partial discharge characteristic frequency energy of the target frequency band to obtain the partial discharge detection result includes: Determine a partial discharge characteristic index based on the partial discharge characteristic frequency energy of the target frequency band; the partial discharge characteristic frequency energy of the target frequency band is the partial discharge characteristic frequency energy of a preset number of frequency bands when the partial discharge characteristic frequency energy is arranged in descending order; A partial discharge detection is performed based on the partial discharge characteristic index to obtain a partial discharge detection result.

5. The partial discharge detection method according to any one of claims 1 to 4, characterized in that: The partial discharge detection is performed based on the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain the partial discharge detection result, and then the following further comprises: When the partial discharge detection result indicates that the power equipment to be detected has a partial discharge phenomenon, normalizing the partial discharge characteristic frequency energy of each frequency band to obtain a signal weight of each frequency band; Based on the signal weights of the frequency bands, signal reconstruction is performed on the discharge sound signal components to obtain a reconstructed sound signal; The partial discharge type is identified based on the reconstructed sound signal to obtain the partial discharge type of the electric power equipment to be detected.

6. The partial discharge detection method according to claim 5, characterized in that: The performing of partial discharge type identification based on the reconstructed sound signal to obtain the partial discharge type of the power equipment to be detected includes: Respectively extracting time domain features and frequency domain features of the reconstructed sound signal to obtain a time domain partial discharge spectrum and a frequency domain Fourier transform feature; Based on the time-domain partial discharge spectrum and the frequency-domain Fourier transform characteristics, partial discharge type identification is performed to obtain the partial discharge type of the power equipment to be detected.

7. The partial discharge detection method according to claim 6, characterized in that: The method of performing partial discharge type identification based on the time domain partial discharge spectrum and the frequency domain Fourier transform feature to obtain the partial discharge type of the power equipment to be detected includes: Performing feature fusion on the time-domain partial discharge spectrum and the frequency-domain Fourier transform feature to obtain a fused feature spectrum; Based on the fused characteristic spectrum, partial discharge type identification is performed to obtain the partial discharge type of the power equipment to be detected.

8. A partial discharge detection device, characterized in that: include: A signal acquisition unit, used to acquire a discharge sound signal of the electric power equipment to be detected; A sliding window filtering unit, configured to perform sliding window filtering on the discharge sound signal based on the discharge periodicity characteristics of the discharge sound signal to obtain a plurality of discharge sound signal components; The discharge detection unit is used to perform partial discharge detection based on the partial discharge characteristic frequency energy of the frequency band corresponding to each discharge sound signal component to obtain a partial discharge detection result.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the partial discharge detection method according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the partial discharge detection method according to any one of claims 1 to 7 is implemented.

Images