Acoustic-thermal integration-based partial discharge detection method and device

Abstract

The invention discloses a partial discharge detection method and device based on sound and heat integration. The method comprises the following steps: acquiring voiceprint time sequence data and an infrared pseudo-color image; performing data preprocessing on the voiceprint time sequence data to obtain a sound wave detection discharge category and sound wave detection position information, and performing logic processing on the infrared pseudo-color image to obtain an infrared detection discharge category and infrared detection position information; coordinate transformation is carried out on the sound wave detection position information to obtain sound wave target position information, and a first target probability graph is generated; performing coordinate transformation on the infrared detection position information to obtain infrared target position information, and generating a second target probability graph; and performing target fusion to obtain a prediction result. According to the method, two different data can be fused subsequently, the detection result of partial discharge is indicated through the prediction result, and the detection accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical equipment detection and insulation diagnosis, and in particular to a local discharge detection method and device based on acoustic and thermal integration. Background Art

[0002] When local discharge occurs, the discharge causes the surrounding air to vibrate, which in turn generates ultrasonic signals. At the same time, local discharge also releases energy, causing local heat accumulation. When the heat accumulates to a level that can be significantly detected, an abnormal infrared signal is generated.

[0003] Compared with infrared signals, acoustic signals can be sent out earlier. At the same time, acoustic signals have the function of distinguishing discharge types, while infrared signals do not have the function of distinguishing categories. This is because acoustic signals are mainly concentrated in the ultrasonic frequency band. Ultrasonic signals attenuate quickly and are difficult to propagate over long distances, while infrared signals can propagate over long distances. When performing local discharge detection, only one of the soundprint and infrared is used for detection, and the detection accuracy is low. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the above existing problems, the present invention is proposed. Therefore, the present invention provides a partial discharge detection method based on acoustic and thermal integration to solve the problem of how to improve the accuracy of detection through a diagnostic method of voiceprint and infrared collaboration.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a partial discharge detection method based on acoustic and thermal integration, comprising: acquiring voiceprint time series data and infrared pseudo-color images;

[0008] Performing data preprocessing on the voiceprint time series data to obtain acoustic wave detection discharge category and acoustic wave detection position information, and performing logic processing on the infrared pseudo-color image to obtain infrared detection discharge category and infrared detection position information;

[0009] Performing coordinate conversion on the acoustic wave detection position information to obtain acoustic wave target position information, and generating a first target probability map according to the acoustic wave target position information;

[0010] Performing coordinate conversion on the infrared detection position information to obtain infrared target position information, and generating a second target probability map according to the infrared target position information;

[0011] Based on the acoustic wave detection discharge category, the infrared detection discharge category, the first target probability map and the second target probability map, target fusion is performed to obtain a prediction result.

[0012] As a preferred solution of the local discharge detection method based on acoustic-thermal integration described in the present invention, obtaining the acoustic wave detection discharge category and acoustic wave detection position information includes:

[0013] Using a preset target classification model to perform target recognition on the voiceprint time series data to obtain the acoustic wave detection discharge category;

[0014] The voiceprint time series data is processed for positioning detection using a preset positioning algorithm to obtain the sound wave detection position information.

[0015] As a preferred solution of the partial discharge detection method based on acoustic-thermal integration described in the present invention, obtaining the infrared detection discharge category and infrared detection position information includes:

[0016] The infrared pseudo-color image is classified and identified and its position is detected by a preset image recognition algorithm to obtain the infrared detection discharge category and the infrared detection position information.

[0017] As a preferred solution of the partial discharge detection method based on acoustic-thermal integration described in the present invention, obtaining the acoustic wave target position information includes:

[0018] The acoustic wave detection position information is represented by preset camera coordinates;

[0019] Performing world coordinate conversion on the sound wave detection position information to obtain the sound wave position information represented by the world coordinate;

[0020] The acoustic wave position information represented by the world coordinates is converted into polar coordinates to obtain the acoustic wave target position information.

[0021] As a preferred solution of the partial discharge detection method based on acoustic-thermal integration described in the present invention, generating a first target probability map according to the acoustic wave target position information includes:

[0022] Generating a first target probability map from the acoustic wave target position information by a preset probability map generation algorithm;

[0023] The probability map generation algorithm is expressed as:

[0024]

[0025]

[0026] Among them, P U (x) represents the first target probability map, x is a one-dimensional vector, N represents the dimension, Indicates the distance resolution of the voiceprint positioning algorithm, is the angular resolution, represents the covariance matrix of all vectors, and Indicates the location point expressed in polar coordinates.

[0027] As a preferred solution of the partial discharge detection method based on acoustic-thermal integration described in the present invention, obtaining the infrared target position information includes:

[0028] The infrared detection position information is represented by preset pixel coordinates;

[0029] Performing image coordinate conversion on the infrared detection position information to obtain infrared position information represented by image coordinates;

[0030] Performing camera coordinate conversion on the infrared position information represented by the image coordinates to obtain the infrared position information represented by the camera coordinates;

[0031] Performing world coordinate conversion on the infrared position information represented by the camera coordinates to obtain the infrared position information represented by the world coordinates;

[0032] The infrared position information represented by the world coordinates is converted into polar coordinates to obtain the infrared target position information.

[0033] As a preferred solution of the partial discharge detection method based on acoustic-thermal integration described in the present invention, obtaining the prediction result includes:

[0034] Performing Hadamard product fusion on a first target probability map corresponding to the acoustic wave detection discharge category and a second target probability map corresponding to the infrared detection discharge category to obtain a target fusion result, wherein the acoustic wave detection discharge category corresponds to the infrared detection discharge category;

[0035] According to a preset detection threshold, the target fusion result is extracted to obtain the prediction result.

[0036] In a second aspect, the present invention provides a partial discharge detection device based on acoustic and thermal integration, comprising:

[0037] A data acquisition module is used to acquire voiceprint time series data and infrared pseudo-color images;

[0038] A data processing module, used for performing data preprocessing on the voiceprint time series data to obtain the acoustic wave detection discharge category and acoustic wave detection position information, and performing logic processing on the infrared pseudo-color image to obtain the infrared detection discharge category and infrared detection position information;

[0039] A first probability map generating module is used to perform coordinate conversion on the acoustic wave detection position information to obtain acoustic wave target position information, and generate a first target probability map according to the acoustic wave target position information;

[0040] A second probability map generating module is used to perform coordinate conversion on the infrared detection position information to obtain infrared target position information, and generate a second target probability map according to the infrared target position information;

[0041] The fusion prediction module is used to perform target fusion based on the acoustic wave detection discharge category, the infrared detection discharge category, the first target probability map and the second target probability map to obtain a prediction result.

[0042] In a third aspect, the present invention provides an electronic device, comprising:

[0043] Memory and processor;

[0044] The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the partial discharge detection method based on acoustic and thermal integration are implemented.

[0045] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the local discharge detection method based on acoustic-thermal integration.

[0046] Compared with the prior art, the present invention has the following beneficial effects: the present invention adopts the method of obtaining voiceprint time series data and infrared pseudo-color images to provide a data basis for subsequent acoustic and thermal integration, and then performs data processing, coordinate conversion and probability map generation processing on the voiceprint time series data and infrared pseudo-color images respectively, and obtains the prediction result according to the acoustic wave detection discharge category, the infrared detection discharge category, the first target probability map and the second target probability map. This technical means effectively solves the problem of low detection accuracy in the related technology due to the use of one of the voiceprint and the infrared for local discharge detection. By acquiring voiceprint time series data and infrared pseudo-color images, voiceprint and infrared data are obtained to provide data support for subsequent collaborative diagnosis; data preprocessing is performed on the voiceprint time series data to obtain the acoustic wave detection discharge category and the acoustic wave detection position information, and by processing the voiceprint time series data, it is beneficial to the subsequent sitting The infrared pseudo-color image is converted into a coordinate, and the infrared detection discharge category and infrared detection position information are obtained by logically processing the infrared pseudo-color image. By processing the infrared pseudo-color image, it is convenient for subsequent coordinate conversion; the acoustic wave detection position information is converted into a coordinate to obtain the acoustic wave target position information, and the acoustic wave target position information generates a first target probability map; the infrared detection position information is converted into a coordinate to obtain the infrared target position information, and the infrared target position information generates a second target probability map. By respectively converting the acoustic wave detection position information and the infrared detection position information into a coordinate and generating a probability map, the voiceprint can be combined with the infrared information, thereby realizing accurate detection of partial discharge; according to the acoustic wave detection discharge category, the infrared detection discharge category, the first target probability map and the second target probability map, the prediction result is obtained, and the detection result of partial discharge is indicated by the prediction result, thereby improving the detection accuracy. At the same time, the voiceprint time series data and the infrared pseudo-color image are converted into polar coordinate form, which is convenient for the subsequent fusion of the two unused data, thereby improving the detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0048] Figure 1 It is a schematic diagram of the overall process of a partial discharge detection method based on acoustic and thermal integration according to an embodiment of the present invention;

[0049] Figure 2 A schematic diagram of a process of generating a first target probability map of a partial discharge detection method based on acoustic-thermal integration according to an embodiment of the present invention;

[0050] Figure 3A schematic diagram of a flow chart of generating a second target probability map of a partial discharge detection method based on acoustic-thermal integration according to an embodiment of the present invention;

[0051] Figure 4 A schematic diagram of the structure of a partial discharge detection device based on acoustic and thermal integration according to an embodiment of the present invention;

[0052] Figure 5 A schematic diagram of the structure of an electronic device for a partial discharge detection method based on acoustic and thermal integration according to an embodiment of the present invention;

[0053] Figure 6 A schematic diagram of a probability generation diagram of a partial discharge detection method based on acoustic-thermal integration according to an embodiment of the present invention.

[0054] Explanation of the accompanying drawings: 601, data acquisition module; 602, data processing module; 603, first probability map generation module; 604, second probability map generation module; 605, fusion prediction module; 701, memory; 702, processor; 703, network interface; 704, user interface; 705, communication bus. DETAILED DESCRIPTION

[0055] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0056] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0057] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0058] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0059] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0060] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0061] Example 1

[0062] Reference Figure 1-Figure 5 , is an embodiment of the present invention, which provides a partial discharge detection method based on acoustic and thermal integration, firstly, obtaining voiceprint time series data and infrared pseudo-color image, obtaining voiceprint and infrared data, and providing data support for subsequent collaborative diagnosis; performing data preprocessing on the voiceprint time series data to obtain acoustic wave detection discharge category and acoustic wave detection position information, and processing the voiceprint time series data is conducive to subsequent coordinate conversion, performing logical processing on the infrared pseudo-color image, obtaining infrared detection discharge category and infrared detection position information, and processing the infrared pseudo-color image is conducive to subsequent coordinate conversion; performing coordinate conversion on the acoustic wave detection position information, Acoustic wave target position information is obtained, and a first target probability map is generated from the acoustic wave target position information; infrared detection position information is coordinate-converted to obtain infrared target position information, and a second target probability map is generated from the infrared target position information. By respectively performing coordinate conversion and probability map generation on the acoustic wave detection position information and the infrared detection position information, the soundprint can be combined with the infrared information, thereby realizing accurate detection of partial discharge; prediction results are obtained according to the acoustic wave detection discharge category, the infrared detection discharge category, the first target probability map and the second target probability map. The detection results of partial discharge are indicated by the prediction results, thereby improving detection accuracy.

[0063] It should be noted that this method is used to detect partial discharge, internal discharge, surface discharge and corona discharge. Internal discharge includes the discharge that occurs in the cavities of solid insulation, or in the bubbles of liquid insulation, and between insulation layers with different dielectric properties; surface discharge occurs when there is a field strength component parallel to the surface of the medium in the electric field, and when the component reaches the breakdown field strength; corona discharge is the local self-sustaining discharge of the gas medium in the non-uniform electric field. By accurately detecting partial discharge, the normal operation of electrical equipment can be monitored and diagnosed to avoid the problem of insulation breakdown.

[0064] In this embodiment, if Figure 1 , the method of the present invention comprises:

[0065] S100, obtaining voiceprint time series data and infrared pseudo-color image.

[0066] Preferably, a microphone array is provided, and the microphone array serves as a voiceprint data receiving device. An infrared sensor is provided to collect infrared images. Since the collected infrared image is a grayscale image, the infrared image is pseudo-color encoded to obtain an infrared pseudo-color image. The voiceprint timing data and the infrared pseudo-color image are obtained through a preset interface to provide data support for subsequent acoustic and thermal integration. The preset interface is preferably a wireless transmission interface, which transmits the data collected by the microphone and the infrared sensor to the processor.

[0067] In another possible implementation, the microphone array may also be replaced by other receiving devices to receive voiceprint timing data.

[0068] It should be noted that the voiceprint time series data is voiceprint time series data of different discharge categories, and the specific categories are determined by partial discharge; the infrared pseudo-color image is also an infrared pseudo-color image of different discharge categories, and the specific categories are determined by partial discharge.

[0069] S200, pre-processing the voiceprint time series data to obtain the acoustic wave detection discharge category and acoustic wave detection position information, and logically processing the infrared pseudo-color image to obtain the infrared detection discharge category and infrared detection position information.

[0070] Preferably, obtaining the acoustic wave detection discharge category and the acoustic wave detection position information includes:

[0071] Use the preset target classification model to identify the target of the voiceprint time series data and obtain the acoustic wave detection discharge category;

[0072] The preset positioning algorithm is used to perform positioning detection processing on the voiceprint time series data to obtain the sound wave detection position information.

[0073] Preferably, obtaining the infrared detection discharge category and the infrared detection position information includes:

[0074] The infrared pseudo-color image is classified and identified and its position is detected through a preset image recognition algorithm to obtain the infrared detection discharge category and infrared detection position information.

[0075] Specifically, first, data preprocessing is performed on the voiceprint time series data, and the preprocessing includes detection of discharge category and detection of discharge position. The acoustic wave detection discharge category and acoustic wave detection position information are obtained through preprocessing, so that the acoustic wave detection discharge category and acoustic wave detection position information can be subsequently used for data fusion to improve detection accuracy; logical processing is also performed on the infrared pseudo-color image, and the logical processing also includes detection of discharge category and detection of discharge position in the infrared pseudo-color image. The infrared detection discharge category and infrared detection position information are obtained through logical processing, so that the infrared detection discharge category and infrared detection position information can be subsequently used for data fusion to improve detection accuracy.

[0076] In another possible implementation, the acoustic wave detection discharge category, acoustic wave detection position information, infrared detection discharge category and infrared detection position information can also be manually screened, and the acoustic wave detection discharge category, acoustic wave detection position information, infrared detection discharge category and infrared detection position information that meet the requirements are selected to ensure the uniformity of the categories.

[0077] Specifically, data preprocessing is performed on the voiceprint time series data to obtain the acoustic wave detection discharge category and the acoustic wave detection position information, including but not limited to: using a preset target classification model to perform target recognition on the voiceprint time series data to obtain the acoustic wave detection discharge category; using a preset positioning algorithm to perform positioning detection processing on the voiceprint time series data to obtain the acoustic wave detection position information.

[0078] Preferably, a preset target classification model is specifically used to perform target recognition on the voiceprint time series data. Before performing target recognition, a preset target classification model is first obtained. Different types of voiceprint time series data can be crawled from the Internet through crawler technology. Specifically, the voiceprint time series data includes discharge types such as internal discharge, surface discharge, corona discharge, surface discharge, tip discharge, and suspended discharge of electrical equipment; then the crawled voiceprint time series data is processed, which includes noise reduction, filtering, feature extraction, etc., which helps to reduce noise and interference in the data and extract features useful for identifying discharge categories; the classification model is trained and tested using the processed data to obtain a preset target classification model. Exemplarily, the preset target classification model can be a hidden Markov model, a recurrent neural network, a long short-term memory network, or a combination of a convolutional neural network and a recurrent neural network; specifically, the long short-term memory network is used to perform target recognition on the voiceprint time series data to obtain the acoustic wave detection discharge category. It should be noted that the long short-term memory network can capture the dependency between time series data, thereby extracting a more accurate acoustic wave detection discharge type, which is accurate for subsequent data fusion.

[0079] Preferably, based on the processed voiceprint time series data obtained above, the position where the voiceprint is generated in the voiceprint time series data is detected by a preset positioning algorithm to obtain sound wave detection position information, which is conducive to the subsequent coordinate conversion of the sound wave detection position information for data fusion; exemplarily, the preset positioning algorithm can be a microphone array positioning algorithm, or it can be a method based on beamforming.

[0080] In this embodiment, the infrared pseudo-color image is logically processed to obtain the infrared detection discharge category and infrared detection position information, including but not limited to: using a preset image recognition algorithm to classify and identify the infrared pseudo-color image and perform position detection to obtain the infrared detection discharge category and infrared detection position information.

[0081] Specifically, a preset image recognition algorithm is used to classify and identify the infrared pseudo-color image and detect its position. Before performing image recognition and position, the preset image recognition algorithm is first obtained. Specifically, different types of infrared pseudo-color images are crawled from the Internet through crawler technology. The infrared pseudo-color images include discharge types such as internal discharge of electrical equipment, surface discharge, corona discharge, surface discharge, tip discharge, and suspended discharge. Then, data processing is performed on the crawled infrared pseudo-color images, including image size adjustment, contrast enhancement, etc., which helps to improve the accuracy of image recognition. The processed data is used to train and test the image recognition algorithm to obtain the preset image recognition algorithm.

[0082] Exemplarily, the preset image recognition algorithm can be a convolutional neural network model, a random forest, a convolutional neural network model with an attention mechanism, etc. Specifically, classification recognition and position detection are performed through a convolutional neural network model with an attention mechanism to obtain infrared detection discharge category and infrared detection position information, so as to accurately facilitate subsequent data fusion.

[0083] S300, coordinate conversion is performed on the acoustic wave detection position information to obtain acoustic wave target position information, and a first target probability map is generated according to the acoustic wave target position information.

[0084] It should be noted that in order to enable the fusion of acoustic wave data and infrared data, the coordinates of the acoustic wave detection position information are transformed into position information under a unified coordinate system so that subsequent data fusion can be performed. The converted acoustic wave target position information is then used to generate a heat map, and the target position information is represented on the map through a target probability map. The position information can be clearly displayed, thereby improving the accuracy of detection.

[0085] Preferably, Figure 2 As shown, the acoustic wave detection position information is coordinate-converted to obtain the acoustic wave target position information, and the obtained acoustic wave target position information includes:

[0086] S301, the acoustic wave detection position information is represented by a preset camera coordinate;

[0087] S302, performing world coordinate conversion on the acoustic wave detection position information to obtain the acoustic wave position information represented by the world coordinate;

[0088] Exemplarily, the sound wave detection position information is a preset camera coordinate representation, with the microphone array as the coordinate center. The three-dimensional position in the camera coordinate system needs to be converted to the world coordinate system. The point of the sound wave detection position information is represented by (x, y, z). The extrinsic parameter matrix is ​​required to map the point to the world coordinate system. The camera extrinsic parameter matrix describes the position and direction of the camera in the world coordinate system. It consists of a rotation matrix R and a translation vector t, expressed as [R|t]. The formula for conversion to the world coordinate system is: [X, Y, Z] = R -1 ([x, y, z]-t). [X, Y, Z] represents the position in the world coordinate system. By using the external parameter matrix to transform the acoustic wave detection position information from the camera coordinate system to the world coordinate system, it is beneficial to the subsequent coordinate unification for data fusion.

[0089] In another possible implementation, if it is a single device, or the coordinate centers of multiple devices are arranged at the same position, it is also possible to avoid converting the world coordinates, and directly represent the acoustic wave detection position information as acoustic wave target position information for subsequent calculations. This embodiment takes into account the different sensor coordinates and sets a reference coordinate system for coordinate conversion to unify the coordinate system.

[0090] S303, performing polar coordinate conversion on the sound wave position information represented by the world coordinates to obtain the sound wave target position information.

[0091] Exemplarily, after the world coordinate system is obtained according to step S302, it is converted to the polar coordinate system to facilitate unified fusion of coordinates. For the convenience of calculation and display, this embodiment ignores the pitch coordinate Y which has a smaller impact. The world coordinate system is converted to the polar coordinate system through the following formula, which is expressed as:

[0092]

[0093] in, and θ c is the parameter representation in polar coordinates, X0 and Z0 are pre-set reference parameters. By using the above formula to convert the acoustic wave detection position information from the world coordinate system to the polar coordinate system, it is beneficial to coordinate unification for data fusion.

[0094] In another possible implementation, the sound wave position information represented by the world coordinate system can also be represented in a Cartesian coordinate system. Due to the differences in distance and angular resolution between different sensors, the embodiment of the present application uses a polar coordinate system to reflect the discrimination of different sensors and performs a unified coordinate representation for subsequent use in probability map generation.

[0095] Preferably, generating a first target probability map according to the acoustic wave target position information includes:

[0096] Generate a first target probability map from the acoustic wave target position information using a preset probability map generation algorithm;

[0097] The probability map generation algorithm is expressed as:

[0098]

[0099] Among them, P U (x) represents the first target probability map, x is a one-dimensional vector, N represents the dimension, Indicates the distance resolution of the voiceprint positioning algorithm, is the angular resolution, represents the covariance matrix of all vectors, and Indicates the location point expressed in polar coordinates.

[0100] It should be noted that by mapping to Gaussian distribution and generating a probability map, partial discharge can be clearly represented for subsequent fusion.

[0101] S400, performing coordinate conversion on the infrared detection position information to obtain infrared target position information, and generating a second target probability map according to the infrared target position information.

[0102] It should be noted that in order to enable the fusion of acoustic wave data and infrared data, the infrared detection position information is transformed into position information under a unified coordinate system so that subsequent data fusion can be performed; the converted infrared detection position information is then used to generate a heat map, and the position information is represented on the map through a target probability map, which can clearly display the position information and thus improve the accuracy of detection.

[0103] Preferably, Figure 3 As shown, the infrared detection position information is converted into coordinates to obtain the infrared target position information, including:

[0104] S401, infrared detection position information is represented by preset pixel coordinates;

[0105] S402, performing image coordinate conversion on the infrared detection position information to obtain infrared position information represented by image coordinates;

[0106] For example, the infrared detection position information is represented by preset pixel coordinates, which is the initial form of image data. The coordinate origin is usually located in the upper left corner of the image, the x-axis extends to the right, and the y-axis extends downward; the image coordinate system is also called the normalized coordinate system, where the coordinate origin is located at the center of the image, usually used to represent the optical center of the lens. In this coordinate system, coordinates are usually measured according to actual physical dimensions. In practical applications, in order to simplify calculations, these coordinates are usually normalized to the ratio of the focal length. The conversion from pixel coordinates (u, v) to image coordinates (x1, y1) can remove image distortion, and scaling and translation are performed by using the intrinsic parameter matrix; the conversion formula is: x1 = (uc x ) / f x , y1=(vc y ) / f y , where c x and c y are the x1 and y1 coordinates of the image center, respectively, and f x and f y are the focal lengths in the x1 and y1 directions respectively. The above parameters are all part of the intrinsic parameter matrix. By converting the infrared detection position information from pixel coordinates to image coordinates, the infrared position information represented by the image coordinates is obtained for subsequent camera coordinate conversion.

[0107] S403, performing camera coordinate conversion on the infrared position information represented by the image coordinates to obtain the infrared position information represented by the camera coordinates;

[0108] Exemplarily, the origin of the camera coordinate system is located at the optical center of the camera, the z-axis is consistent with the optical axis of the camera lens, the x-axis and the y-axis respectively define planes perpendicular to the z-axis. When converting the image coordinates to the camera coordinate system, the target association algorithm can be used to calculate the depth z of the point in the camera coordinate system, which is replaced by the distance projection measured by the microphone array to convert the normalized image coordinates into actual three-dimensional coordinates. The conversion formula is: x=x1·z, y=y1·z, z=z. By converting the infrared position information represented by the image coordinates into the camera coordinates, the infrared position information represented by the camera coordinates is obtained for subsequent world coordinate conversion.

[0109] S404, performing world coordinate conversion on the infrared position information represented by the camera coordinates to obtain the infrared position information represented by the world coordinates;

[0110] Exemplarily, since the coordinate representation is based on the microphone array as the coordinate center, the offset needs to be considered in the conversion process of converting the infrared position information represented by the camera coordinates into the world coordinates, while the offset does not need to be considered in the conversion of the sound wave position information. The three-dimensional position in the camera coordinate system is converted to the world coordinate system, and the point of the infrared position information represented by the camera coordinates is represented by (x, y, z). The extrinsic parameter matrix is ​​also used to map the point to the world coordinate system. The rotation matrix and translation vector in the extrinsic parameter matrix take into account the offset, and the specific parameters can be adjusted to ensure consistency; the extrinsic parameter matrix is ​​also expressed as [R|t], and the formula for conversion to the world coordinate system is: [X, Y, Z] = R-1([x, y, z]-t); wherein, [X, Y, Z] represents the position in the world coordinate system. Except for considering the offset, the parameters need to be modified, and the rest is similar to the conversion of the sound wave position information.

[0111] It should be noted that by using the external parameter matrix to convert the infrared position information represented by the camera coordinates into the world coordinate system, it is beneficial to the subsequent coordinate unification for data fusion.

[0112] S405, performing polar coordinate conversion on the infrared position information represented by the world coordinates to obtain infrared target position information;

[0113] Exemplarily, after the world coordinate system is obtained according to step S404, it is converted to a polar coordinate system to facilitate unified fusion of coordinates, for the sake of computational convenience and display considerations, similar to the conversion of sound wave position information in step S303.

[0114] In another possible implementation, the infrared target position information is also generated into a second target probability map through a preset probability map generation algorithm;

[0115] The probability map generation algorithm corresponding to the infrared target position information is expressed as:

[0116]

[0117] Among them, P I (x) represents the second target probability map, x is a one-dimensional vector, N represents the dimension, d j is the depth information of the target, c j represents the channel, s is a hyperparameter, is the angular resolution, Expressed as the covariance matrix of all vectors, and Indicates the position point represented by the above polar coordinates.

[0118] It should be noted that by mapping to Gaussian distribution and generating a probability map, partial discharge can be clearly represented for subsequent fusion.

[0119] S500, performing target fusion based on the acoustic wave detection discharge category, the infrared detection discharge category, the first target probability map and the second target probability map to obtain a prediction result.

[0120] Preferably, obtaining the prediction result includes:

[0121] The first target probability map corresponding to the acoustic wave detection discharge category and the second target probability map corresponding to the infrared detection discharge category are fused by Hadamard product to obtain a target fusion result, where the acoustic wave detection discharge category corresponds to the infrared detection discharge category;

[0122] Exemplarily, the sound wave detection discharge category corresponds to the infrared detection discharge category, that is, the sound wave detection discharge category detected by voiceprint and the infrared detection discharge category detected by infrared are the same; it is also possible that the sound wave detection discharge category detected by voiceprint is weaker, and the infrared detection discharge category detected by infrared is stronger; it is also possible that the sound wave detection discharge category detected by voiceprint is stronger, and the infrared detection discharge category detected by infrared is weaker, and the above detected categories are all the same.

[0123] The first target probability map corresponding to the acoustic wave detection discharge category and the second target probability map corresponding to the infrared detection discharge category are fused by Hadamard product to achieve fusion at the acoustic and thermal result level. The target fusion result reflects the probability distribution of the target and can clearly show the discharge position and discharge category.

[0124] According to the preset detection threshold, the target fusion result is extracted to obtain the prediction result;

[0125] Specifically, the preset detection thresholds include peak detection, position scores and confidence values, etc., which are set according to application scenarios or needs. Then, based on the obtained target fusion results and combined with the discharge category, each target fusion result is extracted, and the extraction is converted into a score for position information and a calculated category confidence value at the position. The score of the position information is compared with the position score, and the calculated category confidence value at the position is compared with the confidence value. The positions where the position information scores are greater than the position scores and the positions where the calculated category confidence values ​​are greater than the confidence values ​​are screened out to obtain prediction results.

[0126] It should be noted that the prediction result includes the detection position and the category confidence corresponding to the detection position, so that the location of the partial discharge and the category of the partial discharge can be obtained, and the voiceprint and infrared are calculated collaboratively to improve the accuracy of the calculation.

[0127] The above is a schematic scheme of a partial discharge detection method based on acoustic and thermal integration of this embodiment. It should be noted that the technical scheme of the partial discharge detection device based on acoustic and thermal integration and the technical scheme of the partial discharge detection method based on acoustic and thermal integration described above belong to the same concept, and the details of the technical scheme of the partial discharge detection device based on acoustic and thermal integration not described in detail in this embodiment can all be referred to the description of the technical scheme of the partial discharge detection method based on acoustic and thermal integration described above.

[0128] In this embodiment, the local discharge detection device based on acoustic and thermal integration is as follows: Figure 4 As shown, including:

[0129] Data acquisition module 601, used to acquire voiceprint time series data and infrared pseudo-color image;

[0130] The data processing module 602 is used to perform data preprocessing on the voiceprint time series data to obtain the acoustic wave detection discharge category and acoustic wave detection position information, and perform logic processing on the infrared pseudo-color image to obtain the infrared detection discharge category and infrared detection position information;

[0131] A first probability map generating module 603 is used to perform coordinate conversion on the acoustic wave detection position information to obtain acoustic wave target position information, and generate a first target probability map according to the acoustic wave target position information;

[0132] The second probability map generating module 604 is used to perform coordinate conversion on the infrared detection position information to obtain infrared target position information, and generate a second target probability map according to the infrared target position information;

[0133] The fusion prediction module 605 is used to perform target fusion based on the acoustic wave detection discharge category, the infrared detection discharge category, the first target probability map and the second target probability map to obtain a prediction result.

[0134] It should be noted that first, the voiceprint time series data and infrared pseudo-color image are acquired through the data acquisition module 601 to acquire the voiceprint and infrared data to provide data support for subsequent collaborative diagnosis; then the data processing module 602 is used to pre-process the voiceprint time series data to obtain the acoustic wave detection discharge category and the acoustic wave detection position information. By processing the voiceprint time series data, it is beneficial to the subsequent coordinate conversion. The infrared pseudo-color image is logically processed to obtain the infrared detection discharge category and the infrared detection position information. By processing the infrared pseudo-color image, it is beneficial to the subsequent coordinate conversion; the first probability map generation module 603 is used to convert the acoustic wave detection position information to obtain the acoustic wave target position. The first target probability map is generated by the acoustic wave target position information; the second probability map generation module 604 is used to transform the infrared detection position information to obtain the infrared target position information, and the second target probability map is generated by the infrared target position information. By respectively transforming the acoustic wave detection position information and the infrared detection position information and generating a probability map, the voiceprint can be combined with the infrared information, thereby realizing accurate detection of partial discharge; finally, the fusion prediction module 605 is used to obtain the prediction result according to the acoustic wave detection discharge category, the infrared detection discharge category, the first target probability map and the second target probability map. The prediction result indicates the detection result of partial discharge, thereby improving the detection accuracy.

[0135] Specifically, the data acquisition module 601 is connected to the data processing module 602, the data processing module 602 is connected to the first probability map generation module 603, the first probability map generation module 603 is connected to the second probability map generation module 604, and the second probability map generation module 604 is connected to the fusion prediction module 605. The technical voiceprint time series data and infrared pseudo-color image acquired by this device provide a data basis for the subsequent acoustic and thermal integration, and then the voiceprint time series data and infrared pseudo-color image are processed, coordinate converted and probability map generated respectively, and the prediction result is obtained according to the acoustic wave detection discharge category, infrared detection discharge category, first target probability map and second target probability map. By using the voiceprint infrared collaborative diagnosis method, the accuracy of detection can be improved.

[0136] This embodiment also provides an electronic device, such as Figure 5 As shown in the figure, it is applicable to partial discharge detection based on acoustic and thermal integration, including:

[0137] A memory 701 and a processor 702; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the partial discharge detection method based on acoustic and thermal integration as proposed in the above embodiment; and also includes: at least one network interface 703, a user interface 704, and at least one communication bus 705.

[0138] Wherein, the communication bus 705 is used to realize the connection and communication between these components;

[0139] The user interface 704 may include a display screen (Display), a camera (Camera), and the optional user interface 704 may also include a standard wired interface and a wireless interface;

[0140] The network interface 703 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface);

[0141] Among them, the processor 702 may include one or more processing cores. The processor 702 uses various interfaces and lines to connect various parts of the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 701, and calling data stored in the memory 701. Optionally, the processor 702 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 702 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU), and a modem. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; and the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 702, but may be implemented separately through a chip;

[0142] The memory 701 may be used to store instructions, programs, codes, code sets or instruction sets. The memory 701 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 701 may optionally be at least one storage device located away from the aforementioned processor 702. Figure 6 , the memory 701 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application program of a partial discharge detection method based on acoustic-thermal integration.

[0143] exist Figure 5 In the electronic device shown, the user interface 704 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 702 can be used to call the application program stored in the memory 701 for a local discharge detection method based on acoustic and thermal integration. When executed by one or more processors 702, the electronic device executes one or more methods in the above-mentioned embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0144] This embodiment further provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for detecting partial discharge based on acoustic and thermal integration as proposed in the above embodiment is implemented.

[0145] The storage medium proposed in this embodiment and the partial discharge detection method based on acoustic-thermal integration proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0146] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention 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 a computer floppy disk, read-only memory (ReadOnly, Memory, ROM), random access memory (RandomAccess Memory, RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform the methods of various embodiments of the present invention.

[0147] Example 2

[0148] Reference Figure 6 , which is an embodiment of the present invention, provides a partial discharge detection method based on acoustic and thermal integration. In order to verify its beneficial effects, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0149] like Figure 6As shown, the embodiment of the present application provides probability generation diagrams of different discharge states, including a surface discharge probability generation diagram (a), a suspended discharge probability generation diagram (b), and a corona discharge probability generation diagram (c). According to the above figure, the local discharge position can be represented in different generation diagrams or in the same generation diagram. Figure 5 It shows the partial discharge prediction situation of the probability generation diagram generated by infrared detection, voiceprint detection and infrared voiceprint detection. In the three different cases, the detection results of this scheme are more accurate.

[0150] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A partial discharge detection method based on acoustic and thermal integration, characterized in that: include: Obtain voiceprint time series data and infrared pseudo-color images; Performing data preprocessing on the voiceprint time series data to obtain acoustic wave detection discharge category and acoustic wave detection position information, and performing logic processing on the infrared pseudo-color image to obtain infrared detection discharge category and infrared detection position information; Performing coordinate conversion on the acoustic wave detection position information to obtain acoustic wave target position information, and generating a first target probability map according to the acoustic wave target position information; Performing coordinate conversion on the infrared detection position information to obtain infrared target position information, and generating a second target probability map according to the infrared target position information; Based on the acoustic wave detection discharge category, the infrared detection discharge category, the first target probability map and the second target probability map, target fusion is performed to obtain a prediction result.

2. The method for detecting partial discharge based on acoustic-thermal integration according to claim 1, characterized in that: The acoustic wave detection discharge category and acoustic wave detection location information include: Using a preset target classification model to perform target recognition on the voiceprint time series data to obtain the acoustic wave detection discharge category; The voiceprint time series data is processed for positioning detection using a preset positioning algorithm to obtain the sound wave detection position information.

3. The method for detecting partial discharge based on acoustic-thermal integration according to claim 2, characterized in that: The infrared detection discharge category and infrared detection position information include: The infrared pseudo-color image is classified and identified and its position is detected by a preset image recognition algorithm to obtain the infrared detection discharge category and the infrared detection position information.

4. The method for detecting partial discharge based on acoustic-thermal integration according to claim 1 or 3, characterized in that: Obtaining the acoustic wave target position information includes: The acoustic wave detection position information is represented by preset camera coordinates; Performing world coordinate conversion on the sound wave detection position information to obtain the sound wave position information represented by the world coordinate; The acoustic wave position information represented by the world coordinates is converted into polar coordinates to obtain the acoustic wave target position information.

5. The method for detecting partial discharge based on acoustic-thermal integration according to claim 4, characterized in that: Generating a first target probability map according to the acoustic wave target position information includes: Generating a first target probability map from the acoustic wave target position information by a preset probability map generation algorithm; The probability map generation algorithm is expressed as: Among them, P U (x) represents the first target probability map, x is a one-dimensional vector, N represents the dimension, Indicates the distance resolution of the voiceprint positioning algorithm, is the angular resolution, represents the covariance matrix of all vectors, and Indicates the location point expressed in polar coordinates.

6. The method for detecting partial discharge based on acoustic-thermal integration according to claim 5, characterized in that: Obtaining the infrared target position information includes: The infrared detection position information is represented by preset pixel coordinates; Performing image coordinate conversion on the infrared detection position information to obtain infrared position information represented by image coordinates; Performing camera coordinate conversion on the infrared position information represented by the image coordinates to obtain the infrared position information represented by the camera coordinates; Performing world coordinate conversion on the infrared position information represented by the camera coordinates to obtain the infrared position information represented by the world coordinates; The infrared position information represented by the world coordinates is converted into polar coordinates to obtain the infrared target position information.

7. The partial discharge detection method based on acoustic-thermal integration according to any one of claim 6, characterized in that: The prediction results include: Performing Hadamard product fusion on a first target probability map corresponding to the acoustic wave detection discharge category and a second target probability map corresponding to the infrared detection discharge category to obtain a target fusion result, wherein the acoustic wave detection discharge category corresponds to the infrared detection discharge category; According to a preset detection threshold, the target fusion result is extracted to obtain the prediction result.

8. A partial discharge detection device based on acoustic and thermal integration, characterized in that: include, A data acquisition module is used to acquire voiceprint time series data and infrared pseudo-color images; A data processing module, used for performing data preprocessing on the voiceprint time series data to obtain the acoustic wave detection discharge category and acoustic wave detection position information, and performing logic processing on the infrared pseudo-color image to obtain the infrared detection discharge category and infrared detection position information; A first probability map generating module is used to perform coordinate conversion on the acoustic wave detection position information to obtain acoustic wave target position information, and generate a first target probability map according to the acoustic wave target position information; A second probability map generating module is used to perform coordinate conversion on the infrared detection position information to obtain infrared target position information, and generate a second target probability map according to the infrared target position information; The fusion prediction module is used to perform target fusion based on the acoustic wave detection discharge category, the infrared detection discharge category, the first target probability map and the second target probability map to obtain a prediction result.

9. An electronic device, characterized in that: include: Memory and processor; The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the partial discharge detection method based on acoustic and thermal integration as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: It stores computer executable instructions, which, when executed by a processor, implement the steps of the partial discharge detection method based on acoustic-thermal integration as described in any one of claims 1 to 7.

Images