A nondestructive testing device and method for diagnosing acoustic and optical imaging of power equipment

Abstract

The present application discloses a nondestructive testing device and method for diagnosing electric power equipment using acoustic and optical imaging. The device structure includes: a voiceprint probe, a stethoscope array, and a sound source positioning processing device. The sound source positioning processing device structure includes: a sound sensor, a wireless transmission device, a display control panel, an acoustic and optical imaging component, and a power supply component. The stethoscope array is densely packed with voiceprint probes. The sound source positioning processing device is connected to a fault location analysis device via a data transmission circuit. The device proposed in the present application, combined with its adapted method, can achieve uninterrupted fault detection of electric power equipment when the electric power equipment is in operation. By means of voiceprint visualization conversion and AI algorithm analysis, the fault location of the electric power equipment can be intuitively and accurately found, with a high detection accuracy. The present application can solve the problem of equipment fault location and difficulty in finding faults in substations.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of power equipment detection, and in particular to a device and method for nondestructive testing of power equipment using acoustic and optical imaging. Background Art

[0002] When operating, electrical equipment emits sounds and vibrations, which contain a wealth of information about its status. These sounds and vibrations, like a human fingerprint, are identifying characteristics of the equipment's operating status. When parts or components, due to wear and aging, cause faults such as partial discharge or abnormal noise, the equipment's operating status changes, causing the normal operating sounds and vibrations to include the sound of the fault. The characteristics of the soundprint signal emitted by the equipment will also change accordingly.

[0003] The advantage of extracting characteristic quantities from the voiceprint signals generated by power equipment to determine the operating status of the equipment and detect faults is that it can realize uninterrupted online monitoring and non-contact detection of power equipment. In addition, during the detection process, since the arrangement of the device for collecting voiceprints is relatively flexible, no electromagnetic signals are generated when collecting signals, which will not interfere with the normal operation of the equipment.

[0004] However, due to the complex internal structure of power equipment, especially large and complex equipment, the noise generation mechanism is diverse, and the power equipment system is a fully enclosed structure; since sound needs to penetrate multiple media, the transmission pattern of sound from the internal structure to the outside is irregular; coupled with the interference of equipment and environmental noise; it makes it difficult for the voiceprint signals collected by traditional detection methods to match one-to-one with power equipment faults, and the robustness and false alarm rate of the detection system cannot meet the requirements of unmanned intelligent monitoring of power equipment. Summary of the Invention

[0005] In order to solve the problems in the existing power equipment fault detection method based on voiceprint extraction that the equipment fault search, positioning, and identification processes cannot eliminate noise, and the sound detection is easily affected by noise interference from different devices and cannot achieve accurate measurement results, the present application provides a power equipment diagnostic acoustic optical imaging non-destructive testing device and method.

[0006] In a first aspect, the present application provides a nondestructive testing device for acoustic and optical imaging of power equipment diagnosis, the structure of which includes:

[0007] Several voiceprint probes are installed on the power equipment to be tested, and are used to collect the voiceprint signals emitted by the power equipment to be tested;

[0008] A stethoscope array wirelessly connected to the voiceprint probe, used to receive the voiceprint signal sent by the voiceprint probe;

[0009] A sound source localization processing device comprises a shell; a power supply assembly and an acoustic-optical imaging assembly are provided inside the shell; a sound sensor and a display control panel are provided on the outer surface of the shell respectively; the sound sensor is electrically connected to the stethoscope array, and is used to process the soundprint signal fed back by the stethoscope array to obtain a de-noised soundprint electrical signal, and send the de-noised soundprint electrical signal to the acoustic-optical imaging assembly; the acoustic-optical imaging assembly is configured to convert the de-noised soundprint electrical signal into a visual image, and send the visual image to the display control panel for display; the power supply assembly is electrically connected to the sound sensor, the acoustic-optical imaging assembly and the display control panel respectively.

[0010] Optionally, the voiceprint probe is provided with a mounting portion for magnetically connecting to the electrical equipment to be tested.

[0011] Optionally, a plurality of voiceprint heads are densely distributed on the stethoscope array, each of the voiceprint heads corresponds to a channel for receiving the voiceprint signal, and the voiceprint heads are arranged in a honeycomb shape on the stethoscope array and spread out in a spiral shape from the center of the stethoscope array to the periphery of the stethoscope array.

[0012] Optionally, the sound source positioning processing device also includes: a drone mounting interface and a robot mounting interface, the drone mounting interface and the robot mounting interface are arranged on the outside of the shell and are connected to the control end of the display control panel inside the shell.

[0013] Optionally, the display end of the display control panel includes: a vehicle control interface, which is configured to control the drone matching the drone mounting interface through the drone mounting interface, and to control the robot matching the robot mounting interface through the robot mounting interface.

[0014] Optionally, the device also includes: a data transmission circuit connected to the sound source localization processing device, and a fault location analysis device connected to the sound source localization processing device through the data transmission circuit; the data transmission circuit is configured to transmit the visual image to the fault location analysis device, and the fault location analysis device is configured to perform AI intelligent algorithm analysis on the received visual image to obtain AI intelligent algorithm analysis results, and perform image combination processing on the visual image and the photos of the power equipment collected by the fault location analysis device to obtain voiceprint display photos of the power equipment.

[0015] Optionally, the fault location analysis device further includes: a data integration development interface, which is provided on the outer surface of the fault location analysis device and is used to connect to an external integration device.

[0016] Optionally, the sound source localization processing device also includes: a wireless transmission device, which is located outside the shell and connected to the control end of the display control panel inside the shell, and the wireless transmission device is configured to transmit the AI ​​intelligent algorithm analysis results transmitted by the fault location analysis device to the display control panel control end to an external wireless connection device.

[0017] In a second aspect, the present application provides a method for nondestructive testing of power equipment using acoustic and optical imaging, comprising the following steps:

[0018] Obtain voiceprint signals when power equipment is running;

[0019] Converting the voiceprint signal into an acoustic-electrical signal to obtain the voiceprint electrical signal;

[0020] De-noising the voiceprint electrical signal to obtain the de-noised voiceprint electrical signal;

[0021] Performing visualization processing on the de-noised voiceprint electrical signal to obtain the voiceprint visualization image;

[0022] Performing preliminary manual analysis based on the voiceprint visualization image to obtain preliminary analysis results;

[0023] If the preliminary analysis result indicates that further analysis is required, an AI intelligent algorithm analysis is performed on the voiceprint visualization image to obtain an AI intelligent algorithm analysis result, and the AI ​​intelligent algorithm analysis result is displayed to the operation and maintenance personnel, and the AI ​​intelligent algorithm analysis result is sent to the fault result database.

[0024] Optionally, if the preliminary analysis result indicates that further analysis is required, then performing AI intelligent algorithm analysis on the voiceprint visualization image, and obtaining the AI ​​intelligent algorithm analysis result, further comprising:

[0025] Collect photos of the power equipment to obtain photos of the power equipment;

[0026] Performing image combining processing on the electric power equipment photo and the voiceprint visualization image corresponding to the electric power equipment photo to form the electric power equipment voiceprint display photo;

[0027] The voiceprint display photo of the power equipment is shown to the operation and maintenance personnel.

[0028] The device in this application features an equiaxial composite spiral stethoscope array, enabling 24-hour, non-contact fault detection for power equipment. This application utilizes visual acousto-optical imaging technology to convert the acoustic signal emitted by power equipment into an electrical signal, which is then transformed, through visual imaging, into a visual image containing the acoustic signal information. This can intuitively indicate the fault location and sound field of the power equipment. This application can solve the problem of locating and troubleshooting equipment faults within substations, helping operators to quickly operate and maintain power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solution of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0030] Figure 1 This is a schematic diagram of an acoustic optical imaging nondestructive testing device for diagnosing power equipment in this application;

[0031] Figure 2 This is a schematic diagram of a stethoscope array for an acoustic optical imaging nondestructive testing device for diagnosing power equipment in this application;

[0032] Figure 3 This is a schematic diagram of a voiceprint probe of an acoustic wave optical imaging nondestructive testing device for diagnosing power equipment in this application;

[0033] Figure 4 This application provides a flow chart of a nondestructive testing method for acoustic and optical imaging of power equipment diagnosis;

[0034] Figure 5 A photograph showing the voiceprint of electric equipment used in the application of a nondestructive testing method for diagnosing electric equipment using acoustic optical imaging;

[0035] Figure 6 This is a schematic diagram of the interface of a display control panel of a nondestructive testing method for diagnosing electric power equipment using acoustic and optical imaging. DETAILED DESCRIPTION

[0036] To illustrate the technical solutions of the present application, some embodiments will now be described in detail. The embodiments described in the following examples do not represent all embodiments consistent with the disclosure of this application. Instead, they are merely examples of methods and systems consistent with certain aspects of the disclosure of this application as detailed in the appended claims.

[0037] In the first aspect, the present application provides a nondestructive testing device for acoustic wave optical imaging of power equipment diagnosis, see Figure 1 , the structure includes:

[0038] Several voiceprint probes 1 arranged on the power equipment to be tested are used to collect voiceprint signals emitted by the power equipment to be tested.

[0039] The stethoscope array 2 wirelessly connected to the voiceprint probe 1 is used to receive the voiceprint signal sent by the voiceprint probe 1 .

[0040] The sound source localization processing device 3 includes a shell 31; a power supply component 32 and an acoustic-optical imaging component 33 are provided inside the shell 31; a sound sensor 34 and a display control panel 35 are respectively provided on the outer surface of the shell 31; the sound sensor 34 is electrically connected to the stethoscope array 2, and is used to process the soundprint signal fed back by the stethoscope array 2 to obtain a de-noised soundprint electrical signal, and send the de-noised soundprint electrical signal to the acoustic-optical imaging component 33; the acoustic-optical imaging component 33 is configured to convert the de-noised soundprint electrical signal into a visual image, and send the visual image to the display control panel 35 for display; the power supply component 32 is electrically connected to the sound sensor 34, the acoustic-optical imaging component 33 and the display control panel 35 respectively.

[0041] Furthermore, the structure of the above-mentioned device can be implemented in many ways, and the components therein can be replaced by components with the same functions, and the components can be combined in many ways.

[0042] In one implementation, the electric power equipment diagnostic acoustic wave optical imaging nondestructive testing device comprises:

[0043] Several voiceprint probes 1 arranged on the power equipment to be tested are used to collect voiceprint signals emitted by the power equipment to be tested.

[0044] The stethoscope array 2 wirelessly connected to the voiceprint probe 1 is used to receive the voiceprint signal sent by the voiceprint probe 1 .

[0045] The sound source localization processing device 3 includes a shell 31; a power supply component 32 and an acoustic-optical imaging component 33 are provided inside the shell 31; a sound sensor 34 and a display control panel 35 are provided on the outer surface of the shell 31 respectively; the sound sensor 34 is electrically connected to the stethoscope array 2, and is used to process the soundprint signal fed back by the stethoscope array 2 to obtain a de-noised soundprint electrical signal, and send the de-noised soundprint electrical signal to the acoustic-optical imaging component 33; the acoustic-optical imaging component 33 is configured to convert the de-noised soundprint electrical signal into a visual image, and send the visual image to the display control panel 35 for display.

[0046] The sound source localization processing device 3 includes a control circuit, which is connected to all other components such as the sound sensor 34, the wireless transmission device 38, the drone mounting interface 36, the robot mounting interface 37, the display control panel 35, the acoustic-optical imaging component 33, etc. After the other components are integrated, the control circuit is connected to the power supply component 32 circuit, and the power supply component 32 supplies power to the other components through the control circuit.

[0047] At the front and rear ends of the sound source localization processing device 3, the control circuit is connected to the stethoscope array 2 and the data transmission circuit 4 respectively to complete the data transmission of this application.

[0048] The data transmission circuit 4 can be hidden in the sound source localization processing device 3 or hidden in the fault localization analysis device 5, or it can be exposed outside the sound source localization processing device 3 or the fault localization analysis device 5. The fault localization analysis device 5 can be integrated with the sound source localization processing device 3, or it can be disconnected from the sound source localization processing device 3 and located in a fixed position.

[0049] In another implementation, the electric power equipment diagnostic acoustic wave optical imaging nondestructive testing device comprises:

[0050] Several voiceprint probes 1 arranged on the power equipment to be tested are used to collect voiceprint signals emitted by the power equipment to be tested.

[0051] The stethoscope array 2 wirelessly connected to the voiceprint probe 1 is used to receive the voiceprint signal sent by the voiceprint probe 1 .

[0052] The sound source localization processing device 3 includes a shell 31; a power supply component 32 and an acoustic-optical imaging component 33 are provided inside the shell 31; a sound sensor 34 and a display control panel 35 are provided on the outer surface of the shell 31 respectively; the sound sensor 34 is electrically connected to the stethoscope array 2, and is used to process the soundprint signal fed back by the stethoscope array 2 to obtain a de-noised soundprint electrical signal, and send the de-noised soundprint electrical signal to the acoustic-optical imaging component 33; the acoustic-optical imaging component 33 is configured to convert the de-noised soundprint electrical signal into a visual image, and send the visual image to the display control panel 35 for display.

[0053] The sound source localization processing device 3 includes a processor, which can be configured as an all-in-one device combined with the display control panel 35, or can be installed in the control circuit of the sound source localization processing device 3 and divided into two parts with the display control panel 35. The processor is used to control the coordinated operation of various components, act as a controller for the drone corresponding to the drone mounting interface 36 and the robot corresponding to the robot mounting interface 37, and act as a processing and logical operation processing device for the software interface in the display control panel 35.

[0054] The data transmission circuit 4 can be hidden in the sound source localization processing device 3 or the fault localization analysis device 5, or it can be exposed outside the sound source localization processing device 3 or the fault localization analysis device 5. The fault localization analysis device 5 can be integrated with the sound source localization processing device 3, or it can be disconnected from the sound source localization processing device 3 and located in a fixed position.

[0055] It should be noted that the above-mentioned acoustic optical imaging non-destructive testing device for power equipment diagnosis can be implemented in many forms. The technical solution formed by fine-tuning the device structure proposed in this application or replacing some components is also within the scope of protection of this application.

[0056] In some embodiments, see Figure 3 The voiceprint probe 1 is provided with a mounting portion for magnetically connecting to the electrical equipment to be tested.

[0057] In this application, the voiceprint probe 1 is a sound source location instrument and a device equipped with passive voice technology. The voiceprint probe integrates an FPGA processor and advanced sound imaging algorithms. In addition to voice enhancement, it can also provide real-time noise imaging, generate audio and video, visualize sound, and locate noise sources.

[0058] In some embodiments, the voiceprint probe 1 can be installed in a variety of ways. It can be mounted on inspection robots or drones, or it can be fixed around critical facilities, enabling 24-hour uninterrupted voiceprint collection. The voiceprint probe 1 can also analyze the acoustic signals of the facility being tested, evaluate the equipment's operating status based on voiceprint characteristics, conduct early detection of facility failures, and utilize artificial intelligence technology for diagnostic classification. The voiceprint probe 1 also offers ultra-high bandwidth signal frequency coverage from 20Hz to 96kHz, encompassing both audible and ultrasonic signals. This can be used to identify and locate mechanical actuation noise, partial discharge ultrasound, and gas ultrasound. The voiceprint probe 1 can also be integrated with other power plant monitoring or inspection systems. The voiceprint probe 1 also provides a flexible software interface SDK, using the HTTP protocol to perform parameter settings, status queries, and data reading on the voiceprint probe 21. Application software, web applications, mobile apps, and the like can communicate with and control the voiceprint probe 21, enabling real-time data transmission, querying, and setting the sound imaging frequency range.

[0059] Furthermore, the function to be achieved by the voiceprint probe 1 proposed in this application is: to be able to be adsorbed on electrical equipment to obtain voiceprint data without damaging the outer surface or shell of the electrical equipment. Therefore, as long as it can achieve the function of non-destructive adsorption and collection of voiceprint data, the voiceprint probe 1 can be a structure in which the magnetic mounting part is located below the voiceprint probe 1, or a structure in which the voiceprint probe 1 is set inside the magnetic mounting part. The connection method between the voiceprint probe 1 and the electrical equipment can be set to magnetic adsorption, a detachable mechanical combination, or a detachable adhesive bonding.

[0060] In some embodiments, see Figure 2 There are multiple voiceprint heads 21 densely distributed on the stethoscope array 2. Each voiceprint head 21 corresponds to a channel for receiving the voiceprint signal. The voiceprint heads 21 are arranged in a honeycomb shape on the stethoscope array 2 and spread out in a spiral shape from the center of the stethoscope array 2 to the periphery of the stethoscope array 2.

[0061] Furthermore, in the present application, the stethoscope array 2 adopts an equiaxial composite spiral stethoscope array with 124 signal channels. The signal channels are dense enough and have super strong 124 signal channels. They have symmetrical environmental noise reduction and source filtering functions. The stethoscope array 2 is in a honeycomb shape as a whole, and the signal channels are spirally spread out around the stethoscope array 2. The structural center point of the stethoscope array 2 is gradually concave at the rate of pi, so that the voiceprint signal, sound, and sound source data received by the stethoscope array 2 are more accurate. The voiceprint data can also be amplified and subjected to noise reduction and filtering processing to provide accurate voiceprint data for the sound source positioning processing device 3. A structure for fixing the voiceprint head 21 is provided in the signal channel, so that each signal channel corresponds to a voiceprint head 21, so that 124 voiceprint heads 21 can receive voiceprint signals at the same time.

[0062] Furthermore, the distribution of the stethoscope array 2 in the stethoscope array 2 is based on an equal-area multi-helical structure, which ensures that the acoustic imaging device of the stethoscope array 2 has the advantage of acoustic positioning within the analysis frequency range of 2k to 96k.

[0063] Furthermore, the stethoscope array 2 can be a 124-channel stethoscope array, a 64-channel stethoscope array, or a stethoscope array of other structures. The stethoscope array 2 can be a sound positioning and tracking device, or other devices for determining the location of a sound source, such as a voice positioning and tracking device used in an intelligent remote conference system. During a meeting, the camera will swing in any direction following the sound waves and capture the conference room image from the angle of the sound source, thereby locating the speaker in real time. Another example is a voice positioning and tracking device used in a security monitoring system. The camera can sensitively record all changes in the monitoring environment based on sound wave vibrations, and command a robot through sound guidance to assist it in completing relevant designated actions.

[0064] Furthermore, in the present application, the stethoscope array 2 composed of the voiceprint head 21 is an integrated voiceprint sound-raising device, which contains hundreds of sound-raising sensors and can achieve super-island signal-to-noise ratio. It can collect weak sounds that cannot be distinguished by the human ear and provide a reliable signal source for voiceprint fault diagnosis and recognition algorithms.

[0065] In some embodiments, the sound source positioning processing device 3 also includes: a drone mounting interface 36 and a robot mounting interface 37, the drone mounting interface 36 and the robot mounting interface 37 are arranged on the outside of the shell 31, and are connected to the control end of the display control panel 35 inside the shell 31.

[0066] Furthermore, the sound source localization processing device 3 may also include a fixed tripod, which is located below the sound source localization processing device and is detachably connected to the lower surface of the sound source localization processing device. The fixed tripod serves to support the sound source localization processing device 3 and other devices attached thereto. It may be a loading platform, a tripod, or a support vehicle with wheels, or other objects that can support the sound source localization processing device 3. The fixed tripod and the lower surface of the sound source localization processing device 3 may be detachably connected, separably bonded, fixed by bolts and screw holes, or fixed by grooves and bosses. In some cases where it is not necessary to remove the sound source localization processing device 3 from the supporting device, the fixed tripod and the lower surface of the sound source localization processing device 3 may also be non-detachably connected.

[0067] Furthermore, the acoustic-optical imaging component 33 in the sound source localization processing device 3 can be integrated hardware that converts electrical signals into visual images, or it can be software that has the function of converting electrical signals into visual images and is installed on the mainboard of the sound source localization processing device 3. It can also have some additional functions for converting electrical signals into visual images, such as image retouching and image denoising.

[0068] In some embodiments, the display end of the display control panel 35 includes: a vehicle control interface 351, and the vehicle control interface 351 is configured to control the drone matched with the drone mounting interface 36 through the drone mounting interface 36, and to control the robot matched with the robot mounting interface 37 through the robot mounting interface 37.

[0069] Furthermore, the display control panel 35 can be an integrated display panel including a processor, an integrated circuit, a storage device, etc., or a display-type device that does not include the aforementioned components and serves only as a display. When the display control panel 35 is an integrated display panel including a processor, an integrated circuit, a storage device, etc., the entire display control panel 35 is exposed to the outside of the sound source localization processing device 3, and the display control panel 35 is connected to the circuit inside the sound source localization processing device 3 via a data line. When the display control panel 35 is a display-type device that serves only as a display, the display end of the display control panel 35 is exposed to the outside of the sound source localization processing device 3, and the back-end processing system of the display control panel 35 is located inside the sound source localization processing device 3.

[0070] Furthermore, the robot mounting interface 37 and drone mounting interface 36 connected to the display control panel 35 are used to wirelessly connect a robot or drone, enabling the robot or drone to connect to the control circuitry within the sound source localization processing device 3 via the robot mounting interface 37 and drone mounting interface 36. The vehicle interface connected to the display control panel 35 can take various forms, such as a small unmanned vehicle interface or an interface for conveniently carrying equipment. The robot or drone that is compatible with the robot mounting interface 37 and drone mounting interface 36 functions as a tool carrier. The tools carried by the robot or drone can include a voiceprint probe 1, an automatic repair tool for electrical equipment, a camera, a lighting source, and other devices.

[0071] In some embodiments, the device also includes: a data transmission circuit 4 connected to the sound source positioning processing device 3, and a fault location analysis device 5 connected to the sound source positioning processing device 3 through the data transmission circuit 4; the data transmission circuit 4 is configured to transmit the visual image to the fault location analysis device 5, and the fault location analysis device 5 is configured to perform AI intelligent algorithm analysis on the received visual image to obtain AI intelligent algorithm analysis results, and combine the visual image with the photos of the power equipment collected by the fault location analysis device 5 to obtain voiceprint display photos of the power equipment.

[0072] Furthermore, the fault location analysis device 5 can be a computer equipped with AI intelligent algorithm analysis software and image integration software, or a hardware device equipped with AI intelligent algorithm analysis and image integration capabilities. It can be adapted to a photo acquisition device with a camera to capture photos of the power equipment and transmit these photos to the fault location analysis device 5. Alternatively, the fault location analysis device 5 may not be equipped with any photo acquisition device, but rather, maintenance personnel may first capture photos using a camera or other device, and then input the captured photos of the power equipment into the fault location analysis device 5 for image integration processing.

[0073] Furthermore, the data transmission circuit 4 can be a transmission line suitable for transmitting data such as voiceprint visualization images, or it can be a device capable of transmitting data such as voiceprint visualization images and connected to the sound source positioning processing device 3 and the fault positioning analysis device 5 on both sides.

[0074] In some embodiments, the fault location analysis device 5 further includes: a data integration development interface 51 , which is provided on the outer surface of the fault location analysis device 5 and is used to connect to an external integration device.

[0075] Furthermore, in this application, the data integration development interface 51 is an interface reserved for other control systems for secondary development. The data integration development interface 51 can also be a wireless transmission interface that is wirelessly connected to an external control system, or it can be a pluggable data transmission interface.

[0076] In some embodiments, the sound source localization processing device 3 also includes: a wireless transmission device 38, which is located outside the shell 31 and is connected to the control end of the display control panel 35 inside the shell 31. The wireless transmission device 38 is configured to transmit the AI ​​intelligent algorithm analysis results transmitted by the fault location analysis device 5 to the control end of the display control panel 35 to an external wireless connection device.

[0077] Furthermore, the wireless transmission device 38 can be a transmission device with an antenna, whose antenna end is exposed to the outside of the sound source positioning processing device 3, and whose control end is located inside the positioning processing device 3 and connected to the control end of the display control panel 35; or it can be an antenna with only the function of sending signals, whose control end is integrated with the control end of the display control panel 35, and is controlled by the control end of the display control panel 35 to transmit signals.

[0078] In the second aspect, the present application provides a method for nondestructive testing of power equipment using acoustic and optical imaging. Figure 4 , the steps include:

[0079] Acquire the voiceprint signal of power equipment during operation.

[0080] Remove the interference noise signal in the voiceprint data to obtain the noise-removed voiceprint electrical signal.

[0081] The acoustic-electrical signal is converted based on the noise-removed acoustic-electrical signal to obtain the equipment defect acoustic-electrical signal.

[0082] The device defect voiceprint electrical signal is visualized and processed to obtain a voiceprint visualization image.

[0083] Preliminary AI analysis and manual analysis are performed based on the voiceprint visualization image to obtain preliminary analysis results.

[0084] If the preliminary analysis result indicates that further analysis is required, AI intelligent algorithm analysis is performed on the voiceprint visualization image to obtain the AI ​​intelligent algorithm analysis result, and the AI ​​intelligent algorithm analysis result is displayed to the operation and maintenance personnel, and the AI ​​intelligent algorithm analysis result is stored in the fault result database.

[0085] In some embodiments, after performing AI intelligent algorithm analysis on the voiceprint visualization image and obtaining the AI ​​intelligent algorithm analysis result, the method further includes:

[0086] Collect photos of the power equipment to obtain photos of the power equipment.

[0087] The power equipment photo and the voiceprint visualization image corresponding to the power equipment photo are combined and processed to form a power equipment voiceprint display photo.

[0088] Show the photos of the power equipment's voiceprint to the operation and maintenance personnel.

[0089] It should be noted that the photos of power equipment voiceprint display are shown in Figure 5 The power equipment photo and the voiceprint visualization image corresponding to the power equipment photo are combined and processed. The resulting power equipment voiceprint display photo can clearly show the fault location of the power equipment and can intuitively reflect the severity of the fault based on the size and shape of the fault sound field. Figure 5 The absence of an acoustic field at the middle bushing indicates that the bushing is fault-free, while a fault acoustic field of a different shape appears at the insulating terminal on the right, indicating that a fault has occurred at the insulating terminal. Based on the size and shape of the fault acoustic field in the photo, it can be analyzed that the insulating terminal at the top of the photo has a more serious fault, while the three insulating terminals at the bottom of the photo have relatively minor faults.

[0090] For example, in some specific embodiments, an operation and maintenance inspection personnel uses the technical solution of the present application to perform fault detection on power equipment at a kilovolt substation.

[0091] Each main substation device is equipped with a voiceprint collection device. Voiceprint probes 1 are installed on various power equipment within the substation, continuously collecting voiceprint signals from the corresponding equipment 24 hours a day. All voiceprint probes 1 are placed near the power equipment and have no electrical connection to the equipment, so they will not affect the normal operation of the equipment.

[0092] However, due to the complex environment within the station, the signal is mixed with noise from corona, fans, switch operation, bird calls, human voices, and other sources, making it difficult to extract effective voiceprint features. Sound sensor 34 uses an intelligent speech recognition algorithm to separate the voiceprint signal from transient and persistent noise, removing all interference from the voiceprint signal. It then converts the processed, de-noised voiceprint signal into an electrical signal, generating the electrical signal of the equipment defect voiceprint.

[0093] The electrical signal of the equipment defect soundprint passes through the acoustic-optical imaging component 33 to form a visual image of the soundprint.

[0094] After collecting voiceprint samples, the operation and maintenance personnel log in to the voiceprint intelligent recognition system through the display control panel 35 in the main control room. The system displays the voiceprint visualization image on the display screen. After the operation and maintenance personnel confirm that further analysis is required, the voiceprint visualization image is transmitted to the fault location analysis device 5.

[0095] The fault location analysis device 5 uses AI algorithm intelligent analysis to confirm that the analysis result is normal, and the obtained voiceprint visualization image confirms that it can reflect the fault status of the power equipment.

[0096] The fault location analysis device 5 combines the collected photos of the power equipment corresponding to the voiceprint visualization image with the voiceprint visualization image to form a voiceprint display photo of the power equipment, which is displayed on the display screen. The voiceprint visualization image and the power equipment voiceprint display photo, along with other data, are then transmitted via wireless transmission device 38 to the voiceprint data center established by the China Electric Power Research Institute Co., Ltd. for storage in the fault result database. The data in the fault result database supports the continuous optimization of the voiceprint intelligent recognition system, promotes the continuous upgrading of voiceprint recognition technology, and enhances the operation and maintenance management and control of power equipment.

[0097] After continuous optimization of the algorithm, the voiceprint intelligent recognition system has achieved real-time monitoring and fault diagnosis of abnormal operating conditions such as winding deformation, loose components, and jamming of operating mechanisms of power equipment such as transformers and circuit breakers, providing strong support for online monitoring of equipment status and fault diagnosis.

[0098] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments.

Claims

1. A nondestructive testing device for acoustic and optical imaging of power equipment diagnosis, characterized in that: include: A plurality of voiceprint probes (1) are arranged on the power equipment to be tested, and are used to collect voiceprint signals emitted by the power equipment to be tested; a stethoscope array (2) wirelessly connected to the voiceprint probe (1), for receiving the voiceprint signal sent by the voiceprint probe (1); A plurality of voiceprint heads (21) are densely distributed on the stethoscope array (2), each of the voiceprint heads (21) corresponds to a channel for receiving the voiceprint signal, and the voiceprint heads (21) are arranged in a honeycomb shape on the stethoscope array (2) and spread out in a spiral shape from the center of the stethoscope array (2) to the periphery of the stethoscope array (2); A sound source localization processing device (3) comprises a shell (31); a power supply component (32) and an acoustic-optical imaging component (33) are provided inside the shell (31); a sound sensor (34) and a display control panel (35) are provided on the outer surface of the shell (31); the sound sensor (34) is electrically connected to the stethoscope array (2) and is used to process the soundprint signal fed back by the stethoscope array (2) to obtain a de-noised soundprint electrical signal, and send the de-noised soundprint electrical signal to the acoustic-optical imaging component (33); the acoustic-optical imaging component (33) is configured to convert the de-noised soundprint electrical signal into a visual image, and send the visual image to the display control panel (35) for display; the power supply component (32) is electrically connected to the sound sensor (34), the acoustic-optical imaging component (33) and the display control panel (35) respectively; A data transmission circuit (4) connected to the sound source positioning processing device (3), and a fault location analysis device (5) connected to the sound source positioning processing device (3) through the data transmission circuit (4); the data transmission circuit (4) is configured to transmit the visual image to the fault location analysis device (5), and the fault location analysis device (5) is configured to perform AI intelligent algorithm analysis on the received visual image to obtain AI intelligent algorithm analysis results, and perform image combination processing on the visual image and the electric power equipment photo collected by the fault location analysis device (5) to obtain the electric power equipment voiceprint display photo.

2. The device according to claim 1, characterized in that The voiceprint probe (1) is provided with a mounting portion for magnetically connecting to the electrical equipment to be tested.

3. The device according to claim 1, characterized in that The sound source localization processing device (3) further comprises: a drone mounting interface (36) and a robot mounting interface (37); the drone mounting interface (36) and the robot mounting interface (37) are arranged outside the housing (31) and connected to the control end of the display control panel (35) inside the housing (31).

4. The device according to claim 3, characterized in that The display end of the display control panel (35) includes a vehicle control interface (351), wherein the vehicle control interface (351) is configured to control a drone matched with the drone mounting interface (36) through the drone mounting interface (36), and to control a robot matched with the robot mounting interface (37) through the robot mounting interface (37).

5. The device according to claim 1, characterized in that The fault location analysis device (5) further comprises: a data integration development interface (51), wherein the data integration development interface (51) is arranged on the outer surface of the fault location analysis device (5), and the data integration development interface (51) is used to connect with an external integration device.

6. The device according to claim 1, characterized in that The sound source localization processing device (3) further includes: a wireless transmission device (38), the wireless transmission device (38) being located outside the housing (31) and connected to the control end of the display control panel (35) inside the housing (31), the wireless transmission device (38) being configured to transmit the AI ​​intelligent algorithm analysis result transmitted by the fault location analysis device (5) to the control end of the display control panel (35) to an external wireless connection device.

7. A method for nondestructive testing of power equipment using acoustic and optical imaging, applied to the nondestructive testing device for power equipment diagnosis using acoustic and optical imaging according to claim 1, characterized in that the steps include: Obtain voiceprint signals when power equipment is running; Converting the voiceprint signal into an acoustic-electrical signal to obtain a voiceprint electrical signal; De-noising the voiceprint electrical signal to obtain a de-noised voiceprint electrical signal; Performing visualization processing on the de-noised voiceprint electrical signal to obtain a voiceprint visualization image; Performing preliminary manual analysis based on the voiceprint visualization image to obtain preliminary analysis results; If the preliminary analysis result indicates that further analysis is required, an AI intelligent algorithm analysis is performed on the voiceprint visualization image to obtain an AI intelligent algorithm analysis result, and the AI ​​intelligent algorithm analysis result is displayed to the operation and maintenance personnel, and the AI ​​intelligent algorithm analysis result is sent to the fault result database.

8. The method according to claim 7, characterized in that If the preliminary analysis result indicates that further analysis is required, then performing AI intelligent algorithm analysis on the voiceprint visualization image, and obtaining the AI ​​intelligent algorithm analysis result, further comprising: Collect photos of the power equipment to obtain photos of the power equipment; Performing image combining processing on the electric power equipment photo and the voiceprint visualization image corresponding to the electric power equipment photo to form the electric power equipment voiceprint display photo; The voiceprint display photo of the power equipment is shown to the operation and maintenance personnel.

Images