Transformer high voltage bushing fault detection method and device

Abstract

The present invention provides a fault detection method and device for a transformer high-voltage bushing, which can be applied to the field of electrical technology. The method comprises: in response to a fault alarm event triggered by the transformer high-voltage bushing, ultrasonic data, infrared image data and electromagnetic wave data of the transformer high-voltage bushing are collected respectively; feature fusion is performed based on the ultrasonic data, infrared image data and electromagnetic wave data to obtain fusion features; the fusion features are input into a fault detection model to obtain output features; and the output features are decoded to obtain fault detection results of the transformer high-voltage bushing.

Description

Technical Field

[0001] The present invention relates to the field of electrical technology, and more specifically, to a fault detection method and device for a transformer high-voltage bushing. Background Art

[0002] As an important accessory of the transformer, the transformer high-voltage bushing is mainly used for outgoing wires, thus leading the high-voltage wires in the transformer to the outside of the oil tank. Therefore, various faults of the transformer high-voltage bushing may cause large-scale power outages and affect the safe and stable operation of the entire power grid and the safety of substation workers. Therefore, it is necessary to strengthen the monitoring of the transformer high-voltage bushing and eliminate various possible unsafe hidden dangers in a timely manner.

[0003] At present, it has been found that many reasons may cause transformer high-voltage bushing failure, such as partial discharge, poor sealing, insulation moisture, etc. Therefore, it is necessary to accurately determine the fault type and location of the transformer high-voltage bushing, so as to find various faults of the transformer high-voltage bushing as early as possible, and formulate corresponding maintenance plans according to different fault characteristics, so as to avoid major accidents.

[0004] In the process of realizing the concept of the present invention, the relevant technology usually analyzes the high-voltage bushing of the transformer after a fault occurs through expert experience, so as to determine the fault situation and perform repairs. This method relies too much on expert experience, has low efficiency, and is difficult to ensure accuracy. Summary of the invention

[0005] In view of this, the present invention provides a method and device for detecting faults of a transformer high-voltage bushing.

[0006] One aspect of the present invention provides a fault detection method for a transformer high-voltage bushing, comprising: in response to a fault alarm event triggered by the transformer high-voltage bushing, respectively collecting ultrasonic data, infrared image data and electromagnetic wave data of the transformer high-voltage bushing; performing feature fusion based on the ultrasonic data, infrared image data and electromagnetic wave data to obtain fused features; inputting the fused features into a fault detection model to obtain output features; and decoding the output features to obtain a fault detection result of the transformer high-voltage bushing.

[0007] According to an embodiment of the present invention, the output features are decoded to obtain a fault detection result of a transformer high-voltage bushing, including: decoding the output features to obtain a first classification result characterizing the fault type, a second classification result characterizing the fault location, and a third classification result characterizing the fault degree; and obtaining the fault detection result of the transformer high-voltage bushing based on the first classification result, the second classification result, and the third classification result.

[0008] According to an embodiment of the present invention, feature fusion is performed based on ultrasound data, infrared image data and electromagnetic wave data to obtain fused features, including: normalizing the ultrasound data, infrared image data and electromagnetic wave data respectively to obtain first ultrasound data, target infrared image data and first electromagnetic wave data; aligning the first ultrasound data and the first electromagnetic wave data based on the wide dimension and high dimension of the target infrared image data to obtain second ultrasound data and second electromagnetic wave data; and splicing the second ultrasound data, the target infrared image data and the second electromagnetic wave data according to the channel dimension to obtain fused features.

[0009] According to an embodiment of the present invention, feature fusion is performed based on ultrasonic data, infrared image data and electromagnetic wave data to obtain fused features, including: normalizing the ultrasonic data, infrared image data and electromagnetic wave data respectively to obtain first ultrasonic data, target infrared image data and first electromagnetic wave data; aligning the first ultrasonic data and the first electromagnetic wave data respectively based on the first target dimension of the target infrared image data to obtain third ultrasonic data and third electromagnetic wave data; and concatenating the third ultrasonic data, target infrared image data and the third electromagnetic wave data according to the second target dimension to obtain fused features; wherein, when the first target dimension is a high dimension, the second target dimension is a wide dimension; when the first target dimension is a wide dimension, the second target dimension is a high dimension.

[0010] According to an embodiment of the present invention, the fault detection method for a transformer high-voltage bushing further includes: acquiring real-time ultrasonic data, real-time infrared image data, and real-time electromagnetic wave data of the transformer high-voltage bushing acquired in real time; performing a fault assessment based on the real-time ultrasonic data, real-time infrared image data, and real-time electromagnetic wave data to obtain an assessment result; and triggering a fault alarm event when the assessment result indicates that a fault exists.

[0011] According to an embodiment of the present invention, fault assessment is performed based on real-time ultrasonic data, real-time infrared image data and real-time electromagnetic wave data to obtain an assessment result, including: when the number of abnormal values ​​existing in the real-time ultrasonic data is greater than a first threshold, an assessment result indicating the existence of a fault is obtained; and / or when the number of abnormal pixel points existing in the real-time infrared image data is greater than a second threshold, an assessment result indicating the existence of a fault is obtained; and / or when the number of abnormal values ​​existing in the real-time electromagnetic wave data is greater than a third threshold, an assessment result indicating the existence of a fault is obtained.

[0012] According to an embodiment of the present invention, feature fusion is performed based on ultrasonic data, infrared image data and electromagnetic wave data to obtain fused features, including: determining the expected fault type based on the evaluation result; and based on the expected fault type, performing feature fusion on the ultrasonic data, infrared image data and electromagnetic wave data to obtain fused features.

[0013] According to an embodiment of the present invention, the fault detection method for a transformer high-voltage bushing further includes: using historical ultrasonic data, historical infrared image data, historical electromagnetic wave data, historical fault type data, historical fault location data and historical fault degree data collected in multiple historical time periods respectively as training samples and labels, training an initial model to obtain a fault detection model.

[0014] According to an embodiment of the present invention, historical ultrasonic data, historical infrared image data, historical electromagnetic wave data, historical fault type data, historical fault location data and historical fault degree data respectively collected in multiple historical time periods are used as training samples and labels to train an initial model to obtain a fault detection model, including: for each historical time period, feature fusion is performed based on the historical ultrasonic data, historical infrared image data and historical electromagnetic wave data collected in the historical time period to obtain a training sample corresponding to the historical time period; the historical fault type data, historical fault location data and historical fault degree data collected in the historical time period are respectively encoded to obtain a first code, a second code and a third code; the first code, the second code and the third code are fused to obtain a label corresponding to the historical time period; and the initial model is trained using the training samples and labels corresponding to each of the multiple historical time periods to obtain a fault detection model.

[0015] Another aspect of the present invention provides a fault detection device for a transformer high-voltage bushing, comprising: a data acquisition module, for collecting ultrasonic data, infrared image data and electromagnetic wave data of the transformer high-voltage bushing in response to a fault alarm event triggered by the transformer high-voltage bushing; a feature fusion module, for performing feature fusion based on the ultrasonic data, infrared image data and electromagnetic wave data to obtain fused features; a feature determination module, for inputting the fused features into a fault detection model to obtain output features; and a feature decoding module, for decoding the output features to obtain a fault detection result of the transformer high-voltage bushing.

[0016] Another aspect of the present invention provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method as described above.

[0017] Another aspect of the present invention provides a computer-readable storage medium storing computer-executable instructions, which are used to implement the above method when executed.

[0018] Another aspect of the present invention provides a computer program product, the computer program product comprising computer executable instructions, the instructions being used to implement the method as described above when being executed.

[0019] According to an embodiment of the present invention, in response to a fault alarm event triggered by a transformer high-voltage bushing, feature fusion is performed based on ultrasonic data, infrared image data, and electromagnetic wave data to obtain multi-modal fusion features. Through multi-modal data, a more accurate feature description of the transformer high-voltage bushing can be provided to improve the accuracy of fault detection. The fusion features are processed using a fault detection model to obtain output features, and the output features are decoded to obtain the fault detection results of the transformer high-voltage bushing. Through a single model, multiple fault detection can be performed on the transformer high-voltage bushing, thereby improving the efficiency of fault detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:

[0021] Figure 1 A diagram showing an application scenario of a transformer high voltage bushing fault detection method and device according to an embodiment of the present invention.

[0022] Figure 2 A flow chart of a transformer high voltage bushing fault detection method according to an embodiment of the present invention is shown.

[0023] Figure 3A The electromagnetic wave signal diagram of the transformer high voltage bushing according to the embodiment of the present invention is shown.

[0024] Figure 3B The electromagnetic wave spectrum diagram of the transformer high voltage bushing according to an embodiment of the present invention is shown.

[0025] Figure 4A The feature fusion process according to an embodiment of the present invention is shown.

[0026] Figure 4B FIG. 4 shows a feature fusion process according to another embodiment of the present invention.

[0027] Figure 5 A block diagram of a fault detection device for a transformer high voltage bushing according to an embodiment of the present invention is shown.

[0028] Figure 6 A block diagram of an electronic device suitable for implementing the method described above according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0029] Below, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present invention. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of embodiments of the present invention. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of concepts of the present invention.

[0030] The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0031] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0032] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0033] In the embodiments of the present invention, the collection, updating, analysis, processing, use, transmission, provision, disclosure, storage and other aspects of the data involved (for example, including but not limited to user personal information) are in compliance with the provisions of relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. In particular, necessary measures are taken for user personal information to prevent illegal access to user personal information data and maintain the security of user personal information, network security and national security.

[0034] In the embodiment of the present invention, before obtaining or collecting the user's personal information, the user's authorization or consent is obtained.

[0035] An embodiment of the present invention provides a fault detection method for a transformer high-voltage bushing, comprising: in response to a fault alarm event triggered by the transformer high-voltage bushing, respectively collecting ultrasonic data, infrared image data and electromagnetic wave data of the transformer high-voltage bushing; performing feature fusion based on the ultrasonic data, infrared image data and electromagnetic wave data to obtain fused features; inputting the fused features into a fault detection model to obtain output features; and decoding the output features to obtain a fault detection result of the transformer high-voltage bushing.

[0036] Figure 1 The following diagram shows an application scenario of a transformer high voltage bushing fault detection method and device according to an embodiment of the present invention. It should be noted that: Figure 1 The examples shown are merely examples of system architectures to which the embodiments of the present invention may be applied, to help those skilled in the art understand the technical content of the present invention, but do not mean that the embodiments of the present invention may not be used in other devices, systems, environments or scenarios.

[0037] like Figure 1 As shown, the system architecture 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is used to provide a medium for a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired and / or wireless communication links, etc.

[0038] The user may use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications may be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, and / or social platform software, etc. (only for example).

[0039] The first terminal device 101, the second terminal device 102, and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0040] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides support for websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0041] It should be noted that the fault detection method for the transformer high-voltage bushing provided in the embodiment of the present invention can generally be executed by the server 105. Accordingly, the fault detection device for the transformer high-voltage bushing provided in the embodiment of the present invention can generally be set in the server 105. The fault detection method for the transformer high-voltage bushing provided in the embodiment of the present invention can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Correspondingly, the fault detection device for the transformer high-voltage bushing provided in the embodiment of the present invention can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Alternatively, the fault detection method for the transformer high-voltage bushing provided in the embodiment of the present invention can also be performed by the first terminal device 101, the second terminal device 102 or the third terminal device 103, or by other terminal devices different from the first terminal device 101, the second terminal device 102 or the third terminal device 103. Correspondingly, the transformer high-voltage bushing fault detection device provided in the embodiment of the present invention can also be set in the first terminal device 101, the second terminal device 102 or the third terminal device 103, or in other terminal devices different from the first terminal device 101, the second terminal device 102 or the third terminal device 103.

[0042] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0043] Figure 2 A flow chart of a transformer high voltage bushing fault detection method according to an embodiment of the present invention is shown.

[0044] like Figure 2 As shown, the method includes operations S210 to S240.

[0045] In operation S210 , in response to a fault alarm event being triggered by a high-voltage bushing of the transformer, ultrasonic data, infrared image data, and electromagnetic wave data of the high-voltage bushing of the transformer are respectively acquired.

[0046] In operation S220, feature fusion is performed based on the ultrasound data, the infrared image data, and the electromagnetic wave data to obtain fused features.

[0047] In operation S230 , the fused features are input into a fault detection model to obtain output features.

[0048] In operation S240, the output feature is decoded to obtain a fault detection result of the transformer high voltage bushing.

[0049] According to an embodiment of the present invention, when the freely moving metal particles bounce in the dry bushing device, the ultrasonic sensor will capture the acoustic signal of the particle movement with high sensitivity. Different types of metal particles have different movement characteristics in the device, and the types of ultrasonic signals are also different. Therefore, the material and size of the freely moving metal particles can be judged based on the ultrasonic data, and the fault condition of the transformer high-voltage bushing can be further judged.

[0050] According to an embodiment of the present invention, due to unreasonable design of the main cap of the transformer high-voltage bushing or manufacturing, installation and maintenance that are not strictly carried out according to process requirements, the main cap and the cable head conductive circuit may cause local overheating and failure. Therefore, the fault condition of the transformer high-voltage bushing can be judged based on the temperature data of the transformer high-voltage bushing, and the temperature data can be determined by infrared image data.

[0051] According to an embodiment of the present invention, since each local discharge in the insulating medium will cause neutralization of positive and negative charges, accompanied by a current pulse, and radiate electromagnetic waves to the surroundings, the fault condition of the transformer high-voltage bushing can be further determined by detecting the electromagnetic wave data.

[0052] According to an embodiment of the present invention, when a transformer high-voltage bushing fails, a fault alarm event is triggered. In response to the fault alarm event, the transformer high-voltage bushing that triggers the fault alarm event is determined, and ultrasonic data, infrared image data, and electromagnetic wave data of the high-voltage bushing are collected, wherein the ultrasonic data can be obtained by an ultrasonic sensor, and the ultrasonic sensor can be installed, for example, at a distance of 50 cm from the transformer high-voltage bushing, the infrared image data can be obtained by detecting infrared radiation emitted by the transformer high-voltage bushing through an infrared sensor, and the electromagnetic wave data can be obtained by electromagnetic wave detection equipment such as antennas.

[0053] According to an embodiment of the present invention, feature fusion is performed on ultrasound data, Hongtai image data and electromagnetic wave data to obtain fused features, wherein the means of feature fusion may be, for example, feature splicing, pyramid pooling, etc.

[0054] According to an embodiment of the present invention, the fused features are input into a fault detection model to obtain output features, wherein the fault detection model is used to process data of a faulty transformer high-voltage bushing and determine the fault condition of the transformer high-voltage bushing.

[0055] According to an embodiment of the present invention, the output feature is decoded to determine a plurality of fault conditions of the transformer high voltage bushing, and a fault detection result of the transformer high voltage bushing is determined based on the plurality of fault conditions.

[0056] According to an embodiment of the present invention, in response to a fault alarm event triggered by a transformer high-voltage bushing, feature fusion is performed based on ultrasonic data, infrared image data, and electromagnetic wave data to obtain multi-modal fusion features. Through multi-modal data, a more accurate feature description of the transformer high-voltage bushing can be provided to improve the accuracy of fault detection. The fusion features are processed using a fault detection model to obtain output features, and the output features are decoded to obtain the fault detection results of the transformer high-voltage bushing. Through a single model, multiple fault detection can be performed on the transformer high-voltage bushing, thereby improving the efficiency of fault detection.

[0057] According to an embodiment of the present invention, the output features are decoded to obtain a fault detection result of a transformer high-voltage bushing, including: decoding the output features to obtain a first classification result characterizing the fault type, a second classification result characterizing the fault location, and a third classification result characterizing the fault degree; and obtaining the fault detection result of the transformer high-voltage bushing based on the first classification result, the second classification result, and the third classification result.

[0058] According to an embodiment of the present invention, the output feature may be in the form of a vector, wherein the value of each component of the output feature may represent a classification result, and each classification result is associated with a fault condition of the transformer high-voltage bushing, for example, fault type, fault location, fault degree, etc.

[0059] According to an embodiment of the present invention, the output feature is decoded to determine the component values ​​of each of the multiple components of the output feature, and the first classification result, the second classification result and the third classification result are determined based on the multiple component values, thereby determining the fault detection result of the transformer high-voltage bushing.

[0060] For example, the fault detection result may be [x, y, z], where x∈[1, 2, 3], y∈[1, 2, 3], and z∈[1, 2, 3]. The output features are decoded to determine that the first classification result is x, the second classification result is y, and the third classification result is z. According to the values ​​of x, y, and z, the fault conditions are determined respectively, and the fault detection result of the transformer high-voltage bushing is further determined. In the case of x=1, the fault type may be overheat discharge, in the case of x=2, the fault type may be internal discharge, in the case of x=3, the fault type may be surface discharge, in the case of y=1, the fault location may be the upper part of the transformer high-voltage bushing, in the case of y=2, the fault location may be the middle part of the transformer high-voltage bushing, in the case of y=3, the fault location may be the lower part of the transformer high-voltage bushing, in the case of z=1, the fault degree may be an early fault, in the case of z=2, the fault degree may be a mid-fault, and in the case of z=3, the fault degree may be a late fault.

[0061] According to an embodiment of the present invention, the output features are decoded to obtain a first classification result, a second classification result and a third classification result, which respectively characterize the fault type, the fault location and the fault degree, so as to determine the fault detection result of the transformer high-voltage bushing. It is possible to simultaneously determine multiple fault conditions of the transformer high-voltage bushing through the vector output by the fault detection model, thereby achieving comprehensive detection of the faults of the transformer high-voltage bushing and improving the efficiency of fault detection.

[0062] According to an embodiment of the present invention, feature fusion is performed based on ultrasound data, infrared image data and electromagnetic wave data to obtain fused features, including: normalizing the ultrasound data, infrared image data and electromagnetic wave data respectively to obtain first ultrasound data, target infrared image data and first electromagnetic wave data; aligning the first ultrasound data and the first electromagnetic wave data based on the wide dimension and high dimension of the target infrared image data to obtain second ultrasound data and second electromagnetic wave data; and splicing the second ultrasound data, the target infrared image data and the second electromagnetic wave data according to the channel dimension to obtain fused features.

[0063] Figure 3A The electromagnetic wave signal diagram of the transformer high voltage bushing according to the embodiment of the present invention is shown.

[0064] like Figure 3A As shown, the horizontal axis of the electromagnetic wave signal graph collected from the high-voltage bushing of the transformer is time, in ns, and the vertical axis is the amplitude of the electromagnetic wave.

[0065] Figure 3B The electromagnetic wave spectrum diagram of the transformer high voltage bushing according to an embodiment of the present invention is shown.

[0066] like Figure 3B As shown, the electromagnetic wave signal diagram is converted into the frequency domain to obtain an electromagnetic wave spectrum diagram, wherein the horizontal axis of the electromagnetic wave spectrum diagram is the frequency in GHz, and the vertical axis is the amplitude of the electromagnetic wave.

[0067] According to an embodiment of the present invention, amplitude extraction is performed based on the electromagnetic wave spectrum diagram, and a spectrum vector is constructed to further determine the unilateral spectrum. The unilateral spectrum is scaled to obtain electromagnetic wave data corresponding to the electromagnetic wave spectrum diagram.

[0068] According to an embodiment of the present invention, the ultrasonic signal graph, the ultrasonic spectrum graph, and the process of determining ultrasonic data according to the ultrasonic spectrum graph may be similar to electromagnetic waves.

[0069] According to an embodiment of the present invention, ultrasound data, infrared image data and electromagnetic wave data are normalized respectively to obtain first ultrasound data, target infrared image data and first electromagnetic wave data, wherein the ultrasound data and electromagnetic wave data are one-dimensional matrices, and the infrared image data is obtained by representing the pixel value of each pixel point in the infrared image in a two-dimensional matrix.

[0070] Figure 4A The feature fusion process according to an embodiment of the present invention is shown.

[0071] like Figure 4A As shown, based on the wide dimension a and high dimension b of the target infrared image data, the first ultrasonic data and the first electromagnetic wave data are aligned respectively to obtain the second ultrasonic data and the second electromagnetic wave data, wherein during the alignment process, the high dimension can be aligned by padding with 0, and the wide dimension can be aligned by column replication.

[0072] According to an embodiment of the present invention, the second ultrasonic data, the second electromagnetic wave data and the target infrared image data are two-dimensional matrices of the same dimension. Therefore, the second ultrasonic data, the target infrared image data and the second electromagnetic wave data can be spliced ​​according to the channel dimension to obtain fusion features, wherein the fusion features are three-dimensional matrices with a depth of 3.

[0073] According to an embodiment of the present invention, after the ultrasonic data, infrared image data and electromagnetic wave data are normalized and aligned, they are spliced ​​in the channel dimension to obtain fused features, so that the fused features simultaneously include the features of the ultrasonic data, infrared image data and electromagnetic wave data. By using the fused features for fault detection, multiple fault situations can be detected at the same time, thereby improving the efficiency of fault detection.

[0074] According to an embodiment of the present invention, feature fusion is performed based on ultrasonic data, infrared image data and electromagnetic wave data to obtain fused features, including: normalizing the ultrasonic data, infrared image data and electromagnetic wave data respectively to obtain first ultrasonic data, target infrared image data and first electromagnetic wave data; aligning the first ultrasonic data and the first electromagnetic wave data respectively based on the first target dimension of the target infrared image data to obtain third ultrasonic data and third electromagnetic wave data; and concatenating the third ultrasonic data, target infrared image data and the third electromagnetic wave data according to the second target dimension to obtain fused features; wherein, when the first target dimension is a high dimension, the second target dimension is a wide dimension; when the first target dimension is a wide dimension, the second target dimension is a high dimension.

[0075] According to an embodiment of the present invention, the ultrasound data, the infrared image data and the electromagnetic wave data are normalized to obtain first ultrasound data, target infrared image data and first electromagnetic wave data.

[0076] Figure 4B FIG. 4 shows a feature fusion process according to another embodiment of the present invention.

[0077] like Figure 4B As shown, based on the first target dimension d of the target infrared image data, the first ultrasonic data and the first electromagnetic wave data are aligned to obtain third ultrasonic data and third electromagnetic wave data, wherein the alignment can be completed by filling in zeros.

[0078] According to an embodiment of the present invention, the third ultrasonic data, the target infrared image data and the third electromagnetic wave data are spliced ​​according to the second target dimension c to obtain a fusion feature, wherein the fusion feature is a two-dimensional matrix, the first target dimension of the fusion feature is the same as the target infrared image data, and the second target dimension of the fusion feature is the second target dimension of the target infrared image data plus 2, that is, c+2.

[0079] According to an embodiment of the present invention, the ultrasonic data, infrared image data and electromagnetic wave data are normalized, aligned in the first target dimension, and then spliced ​​in the second target dimension, so that the fusion features include the features of the ultrasonic data, infrared image data and electromagnetic wave data at the same time. The fusion features are used for fault detection, and multiple fault situations can be detected at the same time, thereby improving the efficiency of fault detection. In addition, compared with fusion in the channel dimension, splicing in the second target dimension can effectively reduce the dimension of the fusion data, thereby further improving the efficiency of fault detection.

[0080] According to an embodiment of the present invention, the fault detection method for a transformer high-voltage bushing further includes: acquiring real-time ultrasonic data, real-time infrared image data, and real-time electromagnetic wave data of the transformer high-voltage bushing acquired in real time; performing a fault assessment based on the real-time ultrasonic data, real-time infrared image data, and real-time electromagnetic wave data to obtain an assessment result; and triggering a fault alarm event when the assessment result indicates that a fault exists.

[0081] According to an embodiment of the present invention, during the operation of the transformer high-voltage bushing, straw rope data, infrared image data and electromagnetic wave data of the transformer high-voltage bushing are collected in real time to obtain real-time ultrasonic data, real-time infrared image data and real-time electromagnetic wave data.

[0082] According to an embodiment of the present invention, a fault assessment is performed on a high-voltage bushing of a transformer based on real-time ultrasonic data, real-time infrared image data, and real-time electromagnetic wave data. When there is one or more data anomalies in the real-time ultrasonic data, the real-time infrared image data, and the real-time electromagnetic wave data, it is determined that the assessment result indicates the presence of a fault, and a fault alarm event is triggered.

[0083] According to an embodiment of the present invention, based on real-time ultrasonic data, real-time infrared image data and real-time electromagnetic wave data, an evaluation result of a transformer high-voltage bushing can be determined, thereby determining whether the transformer high-voltage bushing has a fault, so as to determine whether to trigger a fault alarm event and perform fault detection. Whether to perform fault detection can be determined through preliminary judgment, thereby avoiding waste of resources caused by performing fault detection when no fault occurs, and improving resource utilization.

[0084] According to an embodiment of the present invention, fault assessment is performed based on real-time ultrasonic data, real-time infrared image data and real-time electromagnetic wave data to obtain an assessment result, including: when the number of abnormal values ​​existing in the real-time ultrasonic data is greater than a first threshold, an assessment result indicating the existence of a fault is obtained; and / or when the number of abnormal pixel points existing in the real-time infrared image data is greater than a second threshold, an assessment result indicating the existence of a fault is obtained; and / or when the number of abnormal values ​​existing in the real-time electromagnetic wave data is greater than a third threshold, an assessment result indicating the existence of a fault is obtained.

[0085] According to an embodiment of the present invention, when the number of abnormal values ​​existing in the real-time ultrasonic data is greater than a first threshold value, the real-time ultrasonic data is determined to be abnormal, and an evaluation result indicating the presence of a fault is obtained, wherein, when the frequency of a certain point in the waveform diagram of the real-time ultrasonic data is higher than the ultrasonic maximum value or lower than the ultrasonic minimum value, the data at the point can be determined to be an abnormal value, wherein the ultrasonic maximum value and the ultrasonic minimum value can be determined based on the average value of the real-time ultrasonic data, for example, based on the waveform diagram of the real-time ultrasonic data, it is determined that the mean value of the real-time ultrasonic data is 60000 Hz, the ultrasonic maximum value can be set to, for example, twice the mean value, i.e., 120000 Hz, and the ultrasonic minimum value can be set to, for example, 1 / 3 of the mean value, i.e., 20000 Hz.

[0086] According to an embodiment of the present invention, when the number of abnormal pixel points existing in the real-time infrared image data is greater than a second threshold value, the real-time infrared image data is determined to be abnormal, and an evaluation result indicating the presence of a fault is obtained, wherein, when the pixel value of a pixel point in the real-time infrared image data is outside the normal pixel value range, the pixel point can be determined to be an abnormal pixel point, wherein the normal pixel value range can be determined based on expert experience.

[0087] According to an embodiment of the present invention, when the number of abnormal values ​​existing in the real-time electromagnetic wave data is greater than a third threshold value, the real-time electromagnetic wave data is determined to be abnormal, and an evaluation result indicating the presence of a fault is obtained, wherein, when the frequency of a certain point in the waveform diagram of the real-time electromagnetic wave is higher than the maximum value of the electromagnetic wave or lower than the minimum value of the electromagnetic wave, the data at the point can be determined to be an abnormal value, wherein the maximum value of the electromagnetic wave and the minimum value of the electromagnetic wave can be determined based on the average value of the real-time electromagnetic wave data. For example, based on the waveform diagram of the real-time electromagnetic wave data, it is determined that the mean value of the real-time electromagnetic wave data is 1500MHz, the maximum value of the electromagnetic wave can be set to, for example, twice the mean value, i.e., 3000MHz, and the minimum value of the ultrasonic wave can be set to, for example, 1 / 3 of the mean value, i.e., 300MHz.

[0088] According to an embodiment of the present invention, abnormal pixel points in real-time infrared image data, abnormal values ​​in real-time electromagnetic wave data, and abnormal values ​​in real-time ultrasound data may also be determined by an abnormal value detection algorithm.

[0089] According to the embodiment of the present invention, by performing fault assessment on real-time ultrasonic data, real-time infrared image data and real-time electromagnetic wave data, the assessment result of the transformer high-voltage bushing can be determined, thereby improving the accuracy of the preliminary assessment.

[0090] According to an embodiment of the present invention, feature fusion is performed based on ultrasonic data, infrared image data and electromagnetic wave data to obtain fused features, including: determining the expected fault type based on the evaluation result; and based on the expected fault type, performing feature fusion on the ultrasonic data, infrared image data and electromagnetic wave data to obtain fused features.

[0091] According to an embodiment of the present invention, when the evaluation result indicates that the transformer high-voltage bushing is faulty, the expected fault type can be determined according to the type of data in which the abnormality occurs. For example, when the number of abnormal values ​​in the ultrasonic data is greater than the first threshold value, and / or the number of abnormal values ​​in the electromagnetic wave data is greater than the third threshold value, the expected fault type may be internal discharge or surface discharge. When the number of abnormal pixel points in the infrared image data is greater than the second threshold value, the expected fault type may be overheat discharge.

[0092] According to an embodiment of the present invention, based on the expected fault type, the amount of data that may cause the expected fault type is increased, and features of the three types of data are fused to obtain fused features.

[0093] For example, when the expected fault type is internal discharge or surface discharge, it can be determined that the data affected by the fault includes ultrasonic data and electromagnetic wave data. Therefore, after aligning the ultrasonic data, infrared image data and electromagnetic wave data, the ultrasonic data and electromagnetic wave data are further processed. For example, in the channel dimension, the number of channels of the ultrasonic data and electromagnetic wave data can be doubled, and then the processed ultrasonic data, infrared image data and electromagnetic wave data can be feature fused to obtain fused features, thereby enhancing the feature proportion of the ultrasonic data and electromagnetic wave data in the fused features.

[0094] According to an embodiment of the present invention, based on the evaluation results, the expected fault type is determined, and the data that needs to be highlighted in the feature fusion is further determined to obtain the fusion feature. By increasing the feature proportion of the data corresponding to the expected fault type in the fusion feature, the accuracy of fault detection is improved.

[0095] According to an embodiment of the present invention, the fault detection method for a transformer high-voltage bushing further includes: using historical ultrasonic data, historical infrared image data, historical electromagnetic wave data, historical fault type data, historical fault location data and historical fault degree data collected in multiple historical time periods respectively as training samples and labels, training an initial model to obtain a fault detection model.

[0096] According to an embodiment of the present invention, historical ultrasonic data, historical infrared image data, and historical electromagnetic wave data collected in multiple historical time periods are used as training samples, and historical fault type data, historical fault location data, and historical fault severity data corresponding to the time period where the training samples are located are used as labels to train the initial model and obtain a fault detection model.

[0097] According to the embodiment of the present invention, the initial model is trained using historical data to obtain a fault detection model, thereby ensuring the fitting ability of the fault detection model and improving the accuracy of fault detection.

[0098] According to an embodiment of the present invention, historical ultrasonic data, historical infrared image data, historical electromagnetic wave data, historical fault type data, historical fault location data and historical fault degree data respectively collected in multiple historical time periods are used as training samples and labels to train an initial model to obtain a fault detection model, including: for each historical time period, feature fusion is performed based on the historical ultrasonic data, historical infrared image data and historical electromagnetic wave data collected in the historical time period to obtain a training sample corresponding to the historical time period; the historical fault type data, historical fault location data and historical fault degree data collected in the historical time period are respectively encoded to obtain a first code, a second code and a third code; the first code, the second code and the third code are fused to obtain a label corresponding to the historical time period; and the initial model is trained using the training samples and labels corresponding to each of the multiple historical time periods to obtain a fault detection model.

[0099] According to an embodiment of the present invention, for each historical period, the historical ultrasonic data, historical infrared image data and historical electromagnetic wave data of the period are feature fused to obtain a training sample corresponding to the historical period. The historical fault type data, historical fault location data and historical fault Chengdu data of the period are encoded to obtain a first code, a second code and a third code respectively, and the first code, the second code and the third code are further fused to obtain a label corresponding to the historical period.

[0100] According to an embodiment of the present invention, data from multiple historical time periods are processed separately to obtain training samples and labels corresponding to the multiple historical time periods, and then the initial model is trained using the training samples and labels to obtain a fault detection model, wherein the initial model may include a convolutional neural network.

[0101] According to an embodiment of the present invention, the training samples and labels of each historical period are determined, wherein the training samples are controlled to be the same as the data format input to the model after preprocessing, and the labels are controlled to be the same as the data format of the output results of the model, so that the trained fault detection model can accurately obtain the output results that conform to the data format according to the input data, so as to further determine the fault detection results. In addition, training the model based on the real data of the historical period can ensure the prediction ability of the model and improve the accuracy of fault detection.

[0102] Figure 5 A block diagram of a fault detection device for a transformer high voltage bushing according to an embodiment of the present invention is shown.

[0103] like Figure 5As shown, the transformer high-voltage bushing fault detection device 500 includes a data acquisition module 510 , a feature fusion module 520 , a feature determination module 530 and a feature decoding module 540 .

[0104] The data acquisition module 510 is used to collect ultrasonic data, infrared image data and electromagnetic wave data of the transformer high-voltage bushing in response to a fault alarm event triggered by the transformer high-voltage bushing.

[0105] The feature fusion module 520 is used to perform feature fusion based on ultrasound data, infrared image data and electromagnetic wave data to obtain fused features.

[0106] The feature determination module 530 is used to input the fused features into the fault detection model to obtain output features.

[0107] The feature decoding module 540 is used to decode the output features to obtain the fault detection result of the transformer high-voltage bushing.

[0108] According to an embodiment of the present invention, the feature decoding module 540 includes a feature decoding submodule and a result determination submodule.

[0109] The feature decoding submodule is used to decode the output features to obtain a first classification result representing the fault type, a second classification result representing the fault location, and a third classification result representing the fault degree.

[0110] The result determination submodule is used to obtain the fault detection result of the transformer high-voltage bushing based on the first classification result, the second classification result and the third classification result.

[0111] According to an embodiment of the present invention, the feature fusion module 520 includes a data normalization submodule, a first data alignment submodule and a first data splicing submodule.

[0112] The data normalization submodule is used to perform normalization processing on the ultrasonic data, the infrared image data and the electromagnetic wave data respectively to obtain the first ultrasonic data, the target infrared image data and the first electromagnetic wave data.

[0113] The first data alignment submodule is used to align the first ultrasonic data and the first electromagnetic wave data based on the wide dimension and the high dimension of the target infrared image data, respectively, to obtain the second ultrasonic data and the second electromagnetic wave data.

[0114] The first data splicing submodule is used to splice the second ultrasonic data, the target infrared image data and the second electromagnetic wave data according to the channel dimension to obtain a fusion feature.

[0115] According to an embodiment of the present invention, the feature fusion module 520 includes a second data alignment submodule and a second data splicing submodule.

[0116] The second data alignment submodule is used to align the first ultrasonic data and the first electromagnetic wave data based on the first target dimension of the target infrared image data to obtain third ultrasonic data and third electromagnetic wave data.

[0117] The second data splicing submodule is used to splice the third ultrasonic data, the target infrared image data and the third electromagnetic wave data according to the second target dimension to obtain a fusion feature.

[0118] According to an embodiment of the present invention, the transformer high voltage bushing fault detection device 500 further includes a data acquisition module, a fault assessment module and an alarm triggering module.

[0119] The data acquisition module is used to acquire real-time ultrasonic data, real-time infrared image data and real-time electromagnetic wave data of the transformer high-voltage bushing.

[0120] The fault assessment module is used to perform fault assessment based on real-time ultrasonic data, real-time infrared image data and real-time electromagnetic wave data to obtain assessment results.

[0121] The alarm triggering module is used to trigger a fault alarm event when the evaluation result indicates that a fault exists.

[0122] According to an embodiment of the present invention, the fault assessment module includes a first fault assessment submodule, a second fault assessment submodule and a third fault assessment submodule.

[0123] The first fault assessment submodule is used to obtain an assessment result indicating the presence of a fault when the number of abnormal values ​​existing in the real-time ultrasonic data is greater than a first threshold.

[0124] The second fault assessment submodule is used to obtain an assessment result indicating the presence of a fault when the number of abnormal pixel points existing in the real-time infrared image data is greater than a second threshold.

[0125] The third fault assessment submodule is used to obtain an assessment result indicating the presence of a fault when the number of abnormal values ​​existing in the real-time electromagnetic wave data is greater than a third threshold.

[0126] According to an embodiment of the present invention, the feature fusion module 520 includes a type determination submodule and a feature fusion submodule.

[0127] The type determination submodule is used to determine the expected fault type based on the evaluation result.

[0128] The feature fusion submodule is used to fuse the features of ultrasonic data, infrared image data and electromagnetic wave data based on the expected fault type to obtain fused features.

[0129] According to an embodiment of the present invention, the transformer high voltage bushing fault detection device 500 further includes a model training module.

[0130] The model training module is used to train the initial model using historical ultrasonic data, historical infrared image data, historical electromagnetic wave data, historical fault type data, historical fault location data and historical fault degree data collected in multiple historical time periods as training samples and labels to obtain a fault detection model.

[0131] According to an embodiment of the present invention, the model training module includes a historical feature fusion submodule, a data encoding submodule, a coding fusion submodule and a model training submodule.

[0132] The historical feature fusion submodule is used to perform feature fusion for each historical period based on the historical ultrasonic data, historical infrared image data and historical electromagnetic wave data collected in the historical period to obtain training samples corresponding to the historical period.

[0133] The data encoding submodule is used to encode the historical fault type data, historical fault location data and historical fault degree data collected in the historical period to obtain a first code, a second code and a third code.

[0134] The code fusion submodule is used to fuse the first code, the second code and the third code to obtain a label corresponding to the historical period.

[0135] The model training submodule is used to train the initial model using the training samples and labels corresponding to the multiple historical time periods to obtain a fault detection model.

[0136] According to the embodiments of the present invention, any one or more of the modules, submodules, units, and subunits, or at least part of the functions of any one of them can be implemented in one module. According to the embodiments of the present invention, any one or more of the modules, submodules, units, and subunits can be split into multiple modules for implementation. According to the embodiments of the present invention, any one or more of the modules, submodules, units, and subunits can be at least partially implemented as hardware circuits, such as field programmable gate arrays (FPGAs), programmable logic arrays (PLAs), systems on chips, systems on substrates, systems on packages, application specific integrated circuits (ASICs), or can be implemented by hardware or firmware in any other reasonable way of integrating or packaging the circuit, or by any one of the three implementation methods of software, hardware, and firmware, or by a proper combination of any of them. Alternatively, according to the embodiments of the present invention, one or more of the modules, submodules, units, and subunits can be at least partially implemented as computer program modules, and when the computer program modules are run, the corresponding functions can be executed.

[0137] For example, any of the data acquisition module 510, the feature fusion module 520, the feature determination module 530, and the feature decoding module 540 can be combined in one module / unit / sub-unit for implementation, or any of the modules / units / sub-units can be split into multiple modules / units / sub-units. Alternatively, at least part of the functions of one or more of these modules / units / sub-units can be combined with at least part of the functions of other modules / units / sub-units and implemented in one module / unit / sub-unit. According to an embodiment of the present invention, at least one of the data acquisition module 510, the feature fusion module 520, the feature determination module 530, and the feature decoding module 540 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or by any one of the three implementation methods of software, hardware, and firmware, or by a suitable combination of any of them. Alternatively, at least one of the data acquisition module 510, the feature fusion module 520, the feature determination module 530 and the feature decoding module 540 may be at least partially implemented as a computer program module, and when the computer program module is executed, the corresponding function may be performed.

[0138] It should be noted that the fault detection system part of the transformer high-voltage bushing in the embodiment of the present invention corresponds to the fault detection method part of the transformer high-voltage bushing in the embodiment of the present invention. The description of the fault detection system part of the transformer high-voltage bushing specifically refers to the fault detection method part of the transformer high-voltage bushing, which will not be repeated here.

[0139] Figure 6 A block diagram of an electronic device suitable for implementing the method described above according to an embodiment of the present invention is shown. Figure 6 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0140] like Figure 6As shown, the electronic device 600 according to an embodiment of the present invention includes a processor 601, which can perform various appropriate actions and processes according to the program stored in the read-only memory ROM 602 or the program loaded from the storage part 608 to the random access memory RAM 603. The processor 601 may include, for example, a general-purpose microprocessor (such as a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (for example, an application-specific integrated circuit (ASIC)), etc. The processor 601 may also include an onboard memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0141] In RAM 603, various programs and data required for the operation of electronic device 600 are stored. Processor 601, ROM 602 and RAM 603 are connected to each other via bus 604. Processor 601 performs various operations of the method flow according to the embodiment of the present invention by executing the program in ROM 602 and / or RAM 603. It should be noted that the program can also be stored in one or more memories other than ROM 602 and RAM 603. Processor 601 can also perform various operations of the method flow according to the embodiment of the present invention by executing the program stored in the one or more memories.

[0142] According to an embodiment of the present invention, the electronic device 600 may further include an input / output (I / O) interface 605, which is also connected to the bus 604. The electronic device 600 may further include one or more of the following components connected to the input / output (I / O) interface 605: an input portion 606 including a keyboard, a mouse, etc.; an output portion 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 608 including a hard disk, etc.; and a communication portion 609 including a network interface card such as a LAN card, a modem, etc. The communication portion 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the input / output (I / O) interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage portion 608 as needed.

[0143] According to an embodiment of the present invention, the method flow according to an embodiment of the present invention can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable storage medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the processor 601, the above-mentioned functions defined in the system of the embodiment of the present invention are executed. According to an embodiment of the present invention, the system, equipment, device, module, unit, etc. described above can be implemented by a computer program module.

[0144] The present invention also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiment; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present invention is implemented.

[0145] According to an embodiment of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium. For example, it may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus or a device.

[0146] For example, according to an embodiment of the present invention, the computer-readable storage medium may include the ROM 602 and / or the RAM 603 described above and / or one or more memories other than the ROM 602 and the RAM 603 .

[0147] An embodiment of the present invention also includes a computer program product, which includes a computer program, and the computer program contains program code for executing the method provided by the embodiment of the present invention. When the computer program product runs on an electronic device, the program code is used to enable the electronic device to implement the transformer high-voltage bushing fault detection method provided by the embodiment of the present invention.

[0148] When the computer program is executed by the processor 601, the above functions defined in the system / device of the embodiment of the present invention are executed. According to the embodiment of the present invention, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0149] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 609, and / or installed from a removable medium 611. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0150] According to an embodiment of the present invention, the program code for executing the computer program provided by the embodiment of the present invention can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (e.g., using an Internet service provider to connect through the Internet).

[0151] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions. It can be understood by those skilled in the art that the features recorded in the various embodiments of the present invention can be combined and / or combined in various ways, even if such a combination or combination is not explicitly recorded in the present invention. In particular, without departing from the spirit and teachings of the present invention, the features recorded in the various embodiments of the present invention can be combined and / or combined in various ways. All such combinations and / or combinations fall within the scope of the present invention.

[0152] The embodiments of the present invention are described above. However, these embodiments are only for the purpose of illustration, and are not intended to limit the scope of the present invention. Although each embodiment is described above, it does not mean that the measures in each embodiment cannot be used in combination advantageously. Without departing from the scope of the present invention, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

Claims

1. A fault detection method for a transformer high voltage bushing, characterized in that: The method comprises: In response to a fault alarm event triggered by a transformer high-voltage bushing, ultrasonic data, infrared image data and electromagnetic wave data of the transformer high-voltage bushing are respectively collected; Perform feature fusion based on the ultrasound data, the infrared image data and the electromagnetic wave data to obtain fusion features; Inputting the fused features into a fault detection model to obtain output features, wherein the fault detection model is obtained by training in the following manner: using historical ultrasonic data, historical infrared image data, historical electromagnetic wave data, historical fault type data, historical fault location data, and historical fault degree data collected in multiple historical time periods as training samples and labels, training an initial model to obtain the fault detection model; and Decoding the output feature to obtain a fault detection result of the transformer high-voltage bushing; The method further comprises: Acquire real-time ultrasonic data, real-time infrared image data and real-time electromagnetic wave data of the transformer high-voltage bushing acquired in real time; Performing fault assessment based on the real-time ultrasonic data, the real-time infrared image data, and the real-time electromagnetic wave data to obtain an assessment result; The step of performing feature fusion based on the ultrasound data, the infrared image data and the electromagnetic wave data to obtain fusion features includes: Based on the evaluation results, determining an expected failure type; and Based on the expected fault type, feature fusion is performed on the ultrasonic data, the infrared image data and the electromagnetic wave data to obtain the fusion feature.

2. The method according to claim 1, characterized in that The decoding process of the output feature to obtain the fault detection result of the transformer high-voltage bushing includes: Decoding the output features to obtain a first classification result representing the fault type, a second classification result representing the fault location, and a third classification result representing the fault degree; and Based on the first classification result, the second classification result and the third classification result, a fault detection result of the transformer high-voltage bushing is obtained.

3. The method according to claim 1, characterized in that The step of performing feature fusion based on the ultrasound data, the infrared image data and the electromagnetic wave data to obtain fusion features includes: Normalizing the ultrasonic data, the infrared image data, and the electromagnetic wave data respectively to obtain first ultrasonic data, target infrared image data, and first electromagnetic wave data; Based on the wide dimension and the high dimension of the target infrared image data, aligning the first ultrasonic data and the first electromagnetic wave data respectively to obtain second ultrasonic data and second electromagnetic wave data; and The second ultrasonic data, the target infrared image data and the second electromagnetic wave data are spliced ​​according to the channel dimension to obtain the fusion feature.

4. The method according to claim 1, characterized in that: The step of performing feature fusion based on the ultrasound data, the infrared image data and the electromagnetic wave data to obtain fusion features includes: Normalizing the ultrasonic data, the infrared image data, and the electromagnetic wave data respectively to obtain first ultrasonic data, target infrared image data, and first electromagnetic wave data; Based on the first target dimension of the target infrared image data, aligning the first ultrasonic data and the first electromagnetic wave data respectively to obtain third ultrasonic data and third electromagnetic wave data; and splicing the third ultrasonic data, the target infrared image data and the third electromagnetic wave data according to the second target dimension to obtain the fusion feature; Among them, when the first target dimension is a high dimension, the second target dimension is a wide dimension; when the first target dimension is a wide dimension, the second target dimension is a high dimension.

5. The method according to claim 1, characterized in that The method further comprises: When the evaluation result indicates that a fault exists, the fault alarm event is triggered.

6. The method according to claim 5, characterized in that The performing fault assessment based on the real-time ultrasonic data, the real-time infrared image data and the real-time electromagnetic wave data to obtain an assessment result includes: When the number of abnormal values ​​existing in the real-time ultrasonic data is greater than a first threshold, obtaining the evaluation result indicating the presence of a fault; and / or When the number of abnormal pixels in the real-time infrared image data is greater than a second threshold, obtaining the evaluation result indicating the presence of a fault; and / or In the case where the number of abnormal values ​​present in the real-time electromagnetic wave data is greater than a third threshold value, the evaluation result indicating the presence of a fault is obtained.

7. The method according to claim 1, characterized in that The fault detection model is obtained by training the initial model using the historical ultrasonic data, historical infrared image data, historical electromagnetic wave data, historical fault type data, historical fault location data and historical fault degree data collected in multiple historical time periods as training samples and labels, including: For each of the historical time periods, feature fusion is performed based on the historical ultrasound data, historical infrared image data, and historical electromagnetic wave data collected in the historical time period to obtain a training sample corresponding to the historical time period; respectively encoding the historical fault type data, the historical fault location data and the historical fault degree data collected in the historical period to obtain a first code, a second code and a third code; fusing the first code, the second code, and the third code to obtain a label corresponding to the historical period; and The initial model is trained using training samples and labels corresponding to each of the plurality of historical time periods to obtain the fault detection model.

8. A fault detection device for a transformer high voltage bushing, characterized in that: The device comprises: A data acquisition module, for collecting ultrasonic data, infrared image data and electromagnetic wave data of the transformer high-voltage bushing in response to a fault alarm event triggered by the transformer high-voltage bushing; A feature fusion module, used for performing feature fusion based on the ultrasound data, the infrared image data and the electromagnetic wave data to obtain fusion features; a feature determination module, configured to input the fused feature into a fault detection model to obtain an output feature, wherein the fault detection model is obtained by training an initial model using historical ultrasonic data, historical infrared image data, historical electromagnetic wave data, historical fault type data, historical fault location data, and historical fault degree data collected in multiple historical time periods as training samples and labels to obtain the fault detection model; and A feature decoding module, used to decode the output feature to obtain a fault detection result of the transformer high-voltage bushing; The device also includes: A data acquisition module, used to acquire real-time ultrasonic data, real-time infrared image data and real-time electromagnetic wave data of the transformer high-voltage bushing; A fault assessment module, used to perform fault assessment based on the real-time ultrasonic data, the real-time infrared image data and the real-time electromagnetic wave data to obtain an assessment result; Wherein, the feature fusion module includes: a type determination submodule, configured to determine an expected fault type based on the evaluation result; and The feature fusion submodule is used to perform feature fusion on the ultrasonic data, the infrared image data and the electromagnetic wave data based on the expected fault type to obtain the fusion feature.

Images