GIS disconnecting switch mechanical fault diagnosis equipment and method based on sound waves
By using an acoustic fault diagnosis method, acoustic data of GIS disconnect switches is obtained and compared, overcoming the limitation of existing technologies that require contact-based diagnosis and realizing contactless, high-precision fault diagnosis.
Patent Information
- Application Number
- CN202311792599.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-12-25
- Publication Date
- 2026-02-06
AI Technical Summary
Existing fault diagnosis methods for GIS disconnect switches require physical contact with the equipment, which has certain limitations.
An acoustic fault diagnosis method is adopted. By acquiring acoustic data from GIS disconnect switches, standard and real-time acoustic signature features are extracted and compared using a voting algorithm to output fault diagnosis results.
It enables contactless fault diagnosis, improving the accuracy and reliability of diagnosis.
Smart Images
Figure CN121476913A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of equipment fault diagnosis technology, and in particular to a mechanical fault diagnosis device and method for GIS disconnect switches based on sound waves. Background Technology
[0002] GIS (Gas Insulated Switchgear) is the abbreviation for gas-insulated fully enclosed switchgear. GIS consists of circuit breakers, disconnecting switches, grounding switches, instrument transformers, surge arresters, busbars, connectors, and outgoing terminals. All these devices or components are enclosed in a grounded metal casing filled with pressurized SF6 insulating gas, hence it is also called SF6 fully enclosed switchgear.
[0003] The GIS disconnect switch is an important component of GIS, and its normal working status directly affects the working status of the entire GIS.
[0004] In existing technologies, fault diagnosis of GIS disconnect switches generally employs vibration analysis, temperature monitoring, and pressure sensing. However, these methods all require contact with the GIS disconnect switch to perform monitoring, which has certain limitations. Summary of the Invention
[0005] This application addresses the limitation of existing methods that require contact with the GIS disconnect switch for monitoring. It provides a mechanical fault diagnosis method for GIS disconnect switches based on acoustic waves, comprising the following steps:
[0006] Acquire acoustic wave data of the GIS disconnect switch during normal mechanical operation, extract standard acoustic wave fingerprint features from the acoustic wave data, and store the standard acoustic wave fingerprint features in the database;
[0007] Acquire real-time acoustic waves during the mechanical operation of the GIS disconnect switch, and extract real-time acoustic wave acoustic signature features from the real-time acoustic waves.
[0008] The real-time acoustic signature features are compared with the standard acoustic signature features in the database, and the fault diagnosis results are output based on the voting algorithm.
[0009] In one feasible implementation, the extraction of standard acoustic signature features from the acoustic data and / or the extraction of real-time acoustic signature features from the real-time acoustic data includes the following steps:
[0010] The collected sound is normalized to eliminate the difference in sound volume and the sound amplitude is reduced proportionally.
[0011] The sound is pre-emphasized using a first-order digital filter with a frequency of 6dB / pass.
[0012] The sound wave is windowed and framed to obtain the sound wave signal;
[0013] The acoustic signal is converted into a spectrum using short-time Fourier transform or linear predictive coding and cepstral method, and standard acoustic signature features or real-time acoustic signature features are extracted from the spectrum.
[0014] In one feasible implementation, comparing the real-time acoustic signature features with standard acoustic signature features in the database includes the following steps:
[0015] Set an Euclidean distance threshold, calculate the Euclidean distance between the extracted real-time acoustic signature features and the standard acoustic signature features, and compare it with the Euclidean distance threshold.
[0016] If the distance is less than the Euclidean distance threshold, the output result is "match"; otherwise, the output result is "no match".
[0017] In one feasible implementation, the frame length of the windowed framing is 10-30ms, the frame shift is half the frame length, or the windowed framing adopts a Hamming window, with 256 points per frame and 128 points of overlapping data between adjacent frames.
[0018] In one feasible implementation, the output of the fault diagnosis result based on the voting algorithm includes the following steps:
[0019] Obtain the Euclidean distances for all outputs that result in "match", denoted as L1, L2, ..., L. n ;
[0020] Calculate L1, L2...L using the following formulas. n Value:
[0021]
[0022] In the formula, k i For coefficients;
[0023] Compare L1, L2...L in sequence n If the value is less than the Euclidean distance threshold, the fault diagnosis result is output as "match"; otherwise, the fault diagnosis result is output as "mismatch".
[0024] In one feasible implementation, the k i The Euclidean distance of the standard acoustic wave acoustic signature is L. i The reciprocal of the geometric distance between the sound acquisition module and the GIS disconnect switch, where i = 1, 2, ..., n.
[0025] Another aspect of this application provides a mechanical fault diagnosis device for GIS disconnect switches based on acoustic waves, which is used to implement any of the above-described mechanical fault diagnosis methods for GIS disconnect switches based on acoustic waves, including:
[0026] Multiple sound acquisition modules are used to collect sound wave signals when the GIS disconnect switch is working;
[0027] The data processing module is used to process the acquired signals;
[0028] Data storage status, used to store data information;
[0029] The sound acquisition module is a microphone, which is fixed on a support structure.
[0030] In one feasible implementation, the support structure includes: a base, a lifting strut, an angle adjustment disc, and a support and walking assembly fixed below the base;
[0031] One end of the lifting support rod is fixedly connected to the base, and the other end is fixedly connected to the angle adjustment plate, which is used to fix the microphone.
[0032] The angle adjustment disc includes a fixed disc, a boss welded to the axis of the fixed disc, a transition disc rotatably connected to the outside of the boss, and a rotating disc rotatably connected to the outside of the transition disc.
[0033] The top outer wall of the transition plate is welded with a fixing lug, and the inner side of the fixing lug is rotatably connected to a fixing cylinder via a rotating shaft. The inner side of the fixing cylinder is fixed with a plurality of elastic cards arranged in a circular array along the axis of the fixing cylinder via a connecting rod.
[0034] In one feasible implementation, the inner wall of the transition plate is slidably connected with multiple arc-head locking rods;
[0035] The arc head lever is connected to the inner wall of the transition plate by a spring, and a pull rope is fixed to the end face of the arc head lever. The other end of the pull rope is fixed to the inner wall of the rotating plate. The side wall of the boss is provided with a slot that matches the arc head lever.
[0036] The side wall of the fixed ear is provided with a locking assembly for locking the rotating shaft. The locking assembly includes a handle connected to the side wall of the fixed ear via an elastic guide rod and a locking block welded to the side wall of the handle and inserted into the inner wall of the rotating shaft.
[0037] The inner wall of the rotating shaft is provided with a key-shaped groove, and the outer wall of the block is provided with a key-shaped protrusion that fits the key-shaped groove with a clearance.
[0038] The lifting support rod includes a lifting rod welded to the bottom of the fixed plate and a fixed cylinder welded to the top of the base. The lifting rod is movably inserted into the inner wall of the fixed cylinder, and the side wall of the fixed cylinder is provided with a shrinkage groove. The upper side wall of the fixed cylinder is fitted with a clamp.
[0039] In one feasible implementation, the supporting walking assembly includes multiple feet fixed to the outer wall of the bottom of the base and multiple casters that are longitudinally slidably connected to the outer wall of the bottom of the base.
[0040] The base is slidably connected to the inner wall below the lifting rod by a drive rod, the side wall of the drive rod is rotatably connected to a second connecting rod, and the other end of the second connecting rod is rotatably connected to a sliding rod.
[0041] The other end of the slide rod is rotatably connected to a first connecting rod, the other end of the first connecting rod is rotatably connected to the top of the universal wheel, and the slide rod is laterally slidably connected to the inner wall of the base.
[0042] As described above, this application provides a device and method for fault diagnosis of GIS disconnecting switch machinery based on sound waves. The method includes the following steps: acquiring sound wave data of the GIS disconnecting switch machinery during normal operation, extracting standard sound wave acoustic signature features from the sound wave data, and storing the standard sound wave acoustic signature features in a database; acquiring real-time sound waves of the GIS disconnecting switch machinery during operation, and extracting real-time sound wave acoustic signature features from the real-time sound waves; comparing the real-time sound wave acoustic signature features with the standard sound wave acoustic signature features in the database, and outputting fault diagnosis results based on a voting algorithm. This application uses sound waves for fault diagnosis, thereby achieving non-contact fault diagnosis, solving the limitation problem. Furthermore, it employs multi-point sound acquisition during sound acquisition and uses a voting method for result output, then utilizes coefficients based on geometric distance for comprehensive output, thereby increasing the accuracy of the output results. Attached Figure Description
[0043] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the implementation of the invention and, together with the description, serve to explain the principles of the embodiments of the invention. It is obvious that the drawings described below are merely some embodiments of the invention, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.
[0044] Figure 1 This is a schematic flowchart illustrating an exemplary embodiment of the mechanical fault diagnosis method for GIS disconnect switches based on acoustic waves.
[0045] Figure 2 This is a schematic diagram illustrating the structure of a mechanical fault diagnosis device for a GIS disconnector based on acoustic waves, as shown in an exemplary embodiment of this application.
[0046] Figure 3 This is a schematic diagram of an angle adjustment disc structure shown in an exemplary embodiment of this application;
[0047] Figure 4 This is a schematic cross-sectional view of the angle adjustment disc structure, as shown in an exemplary embodiment of this application.
[0048] Figure 5 A schematic cross-sectional view of the locking component shown in an exemplary embodiment of this application. Figure 1 ;
[0049] Figure 6 A schematic cross-sectional view of the locking component shown in an exemplary embodiment of this application. Figure 2 ;
[0050] Figure 7 This is a schematic diagram of a lifting strut structure shown in an exemplary embodiment of this application;
[0051] Figure 8 This is a partial cross-sectional view of a GIS disconnector mechanical fault diagnosis device based on acoustic waves, as shown in an exemplary embodiment of this application.
[0052] Figure 9 This is a schematic diagram of the structure of a support walking component shown in an exemplary embodiment of this application.
[0053] Explanation of icon numbers:
[0054] 1-Base; 2-Lifting support rod; 3-Angle adjustment disc; 4-Microphone; 5-Support walking assembly; 6-Fixed disc; 7-Rotating disc; 8-Transition disc; 9-Locking assembly; 10-Fixed cylinder; 11-Rotating shaft; 12-Boss; 13-Spring; 14-Arched head latch; 15-Slot; 16-Pull rope; 17-Fixed ear; 18-Elastic guide rod; 19-Handle; 20-Clamping block; 21-Key-shaped groove; 22-Key-shaped protrusion; 23-Lifting rod; 24-Clamping clamp; 25-Fixed cylinder; 26-Retraction groove; 27-Elastic clip; 28-Connecting rod; 29-Machine foot; 30-First connecting rod; 31-Universal wheel; 32-Slide rod; 33-Second connecting rod; 34-Drive rod. Detailed Implementation
[0055] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the embodiments of the invention will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, mechanisms, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a full understanding of how embodiments of the invention are carried out.
[0056] GIS (Gas Insulated Switchgear) is the abbreviation for Gas Insulated Switchgear. A GIS consists of circuit breakers, disconnecting switches, grounding switches, instrument transformers, surge arresters, busbars, connectors, and outgoing terminals. All these devices and components are enclosed in a grounded metal casing filled with pressurized SF6 insulating gas; hence, it is also called an SF6 fully enclosed switchgear. The GIS disconnecting switch is a crucial component of the GIS system, and its proper functioning directly affects the overall operation of the GIS. Current technologies for fault diagnosis of GIS disconnecting switches typically employ vibration analysis, temperature monitoring, and pressure sensing. However, these methods all require contact with the GIS disconnecting switch for monitoring, which has limitations.
[0057] To address the aforementioned problems, this application provides, on the one hand, a mechanical fault diagnosis method for GIS disconnect switches based on acoustic waves, referring to... Figure 1 As shown, it includes the following steps:
[0058] S100: Acquire acoustic wave data during normal mechanical operation of the GIS disconnect switch, extract standard acoustic wave fingerprint features from the acoustic wave data, and store the standard acoustic wave fingerprint features in the database.
[0059] S200: Acquire real-time acoustic waves during the mechanical operation of GIS disconnect switches and extract real-time acoustic wave acoustic signature features.
[0060] Specifically, when extracting standard acoustic signature features from acoustic data and / or extracting real-time acoustic signature features from real-time acoustic data, the following steps are also included:
[0061] S210: Normalizes the collected sound to eliminate differences in sound volume and reduces the sound amplitude proportionally.
[0062] This step eliminates differences in sound volume, ensuring that all sound samples are within the same amplitude range. That is, regardless of the original intensity of the sound, all sounds will be proportionally reduced to the same amplitude, ensuring that all data are at the same level during subsequent processing.
[0063] S220: Pre-emphasizes the sound by passing it through a first-order digital filter with 6dB / pass frequency. The first-order digital filter with 6dB / pass frequency reduces high-frequency noise interference and improves signal purity.
[0064] S230: Window and frame the sound wave to obtain the sound wave signal.
[0065] Windowing is used to smooth the signal during framing and reduce signal interruptions. Framing, on the other hand, divides a continuous audio signal into multiple short frames, each representing the characteristics of the sound within a short period. After windowing and framing, the sound wave signal is obtained, providing a basis for subsequent spectrum analysis.
[0066] Specifically, in some embodiments of this application, the windowed framing frame length is 10-30 ms, and the frame shift is half the frame length. The choice of a frame length of 10-30 ms is based on the short-term stationarity of the speech signal. Within such a short time, the speech signal can be considered approximately invariant. Therefore, dividing the signal into frames of this length allows for better analysis of the signal characteristics.
[0067] Secondly, the design of shifting the frame by half the frame length is to achieve overlap between adjacent frames. When the frame shift is half the frame length, there will be a 50% overlap between every two frames. This overlap design can reduce signal interruptions caused by framing, allowing the signal to maintain a certain degree of continuity and smoothness after framing processing.
[0068] Alternatively, in some embodiments of this application, the windowing and framing uses a Hamming window, with 256 points per frame and 128 points of overlapping data between adjacent frames.
[0069] Using 256 points per frame means each frame contains 256 sampling points, providing higher temporal resolution. For speech signals, higher resolution allows for better capture of signal details and variations. The overlap between adjacent frames is 128 points, meaning there is 128 sampling points overlapping between every two frames, maintaining signal continuity. By overlapping data from adjacent frames, signal interruptions during framing are reduced, resulting in a more continuous and smoother processed signal. Furthermore, the overlapping data provides greater information redundancy, making feature extraction and recognition more stable and reliable. In this application, by averaging or otherwise processing the overlapping data, the effects of noise and interference can be reduced, improving signal quality and processing effectiveness.
[0070] S240: It uses short-time Fourier transform or linear predictive coding and cepstral method to convert the acoustic signal into a spectrum, and extracts standard acoustic wave voiceprint features or real-time acoustic wave voiceprint features from the spectrum.
[0071] Short-time Fourier transform, linear predictive coding, and cepstral methods can convert time-domain signals into frequency-domain signals, revealing the energy distribution of sound at different frequencies. Extracting standard acoustic wave signature features or real-time acoustic wave signature features from the spectrogram is a key step in sound recognition.
[0072] S300: Compares real-time acoustic wave voiceprint features with standard acoustic wave voiceprint features in the database, and outputs fault diagnosis results based on a voting algorithm.
[0073] Specifically, comparing real-time acoustic wave voiceprint features with standard acoustic wave voiceprint features in the database also includes the following steps:
[0074] S310: Set the Euclidean distance threshold, calculate the Euclidean distance between the extracted real-time acoustic signature features and the standard acoustic signature features, and compare it with the Euclidean distance threshold.
[0075] Euclidean distance is a commonly used method to measure the distance between two vectors. In voiceprint recognition, it is used to compare the similarity between the extracted real-time acoustic voiceprint features and the standard acoustic voiceprint features.
[0076] S320: If the distance is less than the Euclidean distance threshold, the output result is "Match"; otherwise, the output result is "Mismatch". If the Euclidean distance is less than the threshold, the real-time acoustic signature features and the standard acoustic signature features can be considered similar, i.e., they match. Therefore, the output result is "Match".
[0077] Conversely, if the Euclidean distance is greater than or equal to the threshold, then the real-time acoustic signature features and the standard acoustic signature features can be considered dissimilar, i.e., they do not match. Therefore, the output result is "mismatch".
[0078] In some embodiments of this application, outputting fault diagnosis results based on a voting algorithm includes the following steps:
[0079] S330: Obtain the Euclidean distances for all outputs that result in "match", namely L1, L2...L... n ;
[0080] Calculate L1, L2...L using the following formulas. n Value:
[0081]
[0082] In the formula, k i For coefficients;
[0083] S340: Compare L1, L2...L... in sequence. n The value is compared with the Euclidean distance threshold. If it is less than the Euclidean distance threshold, the fault diagnosis result is "match"; otherwise, the fault diagnosis result is "mismatch".
[0084] Where, k i The Euclidean distance of the standard acoustic wave acoustic signature is L. i The reciprocal of the geometric distance between the sound acquisition module and the GIS disconnect switch, where i = 1, 2, ..., n.
[0085] Another aspect of this application provides a mechanical fault diagnosis device for GIS disconnect switches based on acoustic waves, referring to... Figures 2-9 As shown, it includes: multiple sound acquisition modules for acquiring sound wave signals when the GIS disconnect switch is working; a data processing module for processing the acquired signals; a data storage module for storing data information; and a sound acquisition module consisting of a microphone 4, which is fixed on a support structure.
[0086] In some embodiments of this application, the support structure includes: a base 1, a lifting strut 2, an angle adjustment disc 3, and a support and walking assembly 5 fixed below the base 1; one end of the lifting strut 2 is fixedly connected to the base 1, and the other end is fixedly connected to the angle adjustment disc 3, which is used to fix the microphone 4.
[0087] The angle adjustment disc 3 includes a fixed disc 6, a boss 12 welded above the axis of the fixed disc 6, a transition disc 8 rotatably connected to the outside of the boss 12, and a rotating disc 7 rotatably connected to the outside of the transition disc 8; a fixed ear 17 is welded to the top outer wall of the transition disc 8, and a fixed cylinder 10 is rotatably connected to the inner side of the fixed ear 17 through a rotating shaft 11. A plurality of elastic cards 27 arranged in a circular array along the axis of the fixed cylinder 10 are fixed to the inner side of the fixed cylinder 10 through a connecting rod 28.
[0088] By setting the elastic card 27, the microphone 4 can be inserted between the elastic cards 27 inside the fixed tube 10. The elastic card 27 itself clamps the microphone 4. When not in use, the microphone 4 can be pulled out directly, making the installation and disassembly of the microphone 4 relative to the bracket structure more convenient.
[0089] In some embodiments of this application, a plurality of arc-headed locking rods 14 are slidably connected to the inner wall of the transition plate 8; the arc-headed locking rods 14 are connected to the inner wall of the transition plate 8 by springs 13, and a pull rope 16 is fixed to the end face of the arc-headed locking rods 14. The other end of the pull rope 16 is fixed to the inner side wall of the rotating plate 7, and a slot 15 matching the arc-headed locking rods 14 is opened on the side wall of the boss 12.
[0090] When the horizontal angle of microphone 4 needs to be adjusted, first turn the rotating disk 7. At this time, because the transition disk 8 and the boss 12 are mutually limited by the arc head locking rod 14 and the slot 15, they cannot rotate. The rotating disk 7 will first rotate relative to the transition disk 8, which pulls the pull rope 16, causing the arc head locking rod 14 to be pulled out of the slot 15. The transition disk 8 will then rotate until it reaches the desired position. Then, release the rotating disk 7, and the arc head locking rod 14 will be inserted into the slot 15 by the elastic force of the spring 13.
[0091] In this embodiment, based on the rotating disk 7, transition disk 8, and boss 12, an arc-shaped locking rod 14 and a locking groove 15 are added. On the one hand, the horizontal angle of the microphone 4 can be adjusted in a targeted manner. On the other hand, the transition disk 8 and the boss 12 can be locked in the non-adjustment state to ensure positional accuracy. Moreover, the whole process only requires operation of the rotating disk 7, which is more convenient.
[0092] The side wall of the fixed ear 17 is provided with a locking component 9 for locking the rotating shaft 11. The locking component 9 includes a handle 19 connected to the side wall of the fixed ear 17 via an elastic guide rod 18 and a locking block 20 welded to the side wall of the handle 19 and inserted into the inner wall of the rotating shaft 11. The inner wall of the rotating shaft 11 is provided with a key-shaped groove 21, and the outer wall of the locking block 20 is provided with a key-shaped protrusion 22 that is clearance-fitted with the key-shaped groove 21.
[0093] The key-shaped groove 21 and the key-shaped protrusion 22 can be used to lock the rotating shaft 11 and the locking block 20, thereby locking the position of the rotating shaft 11.
[0094] The lifting support rod 2 includes a lifting rod 23 welded to the bottom of the fixed plate 6 and a fixed cylinder 25 welded to the top of the base 1. The lifting rod 23 is movably inserted into the inner wall of the fixed cylinder 25, and the side wall of the fixed cylinder 25 is provided with a shrinkage groove 26. The upper side wall of the fixed cylinder 25 is fitted with a clamp 24.
[0095] When it is necessary to change the height of microphone 4, the clamp 24 can be loosened so that the lifting rod 23 and the fixed cylinder 25 can slide relative to each other. Then, adjust the position of microphone 4 to the desired position, and tighten microphone 4.
[0096] In some embodiments of this application, the supporting walking assembly 5 includes a plurality of feet 29 fixed to the bottom outer wall of the base 1 and a plurality of casters 31 longitudinally slidably connected to the bottom outer wall of the base 1; a drive rod 34 is slidably connected to the inner wall of the base 1 below the lifting rod 23, a second connecting rod 33 is rotatably connected to the side wall of the drive rod 34, and a slide rod 32 is rotatably connected to the other end of the second connecting rod 33; a first connecting rod 30 is rotatably connected to the other end of the slide rod 32, the other end of the first connecting rod 30 is rotatably connected to the top of the caster 31, and the slide rod 32 is laterally slidably connected to the inner wall of the base 1.
[0097] Specifically, when the equipment needs to be moved, the lifting rod 23 can be adjusted to the lowest retracted position. At this time, the drive rod 34 is pressed by the lifting rod 23 and moves downward, thereby pushing the slide rod 32 outward through the second link 33. Then, the caster wheel 31 is pushed downward through the first link 30, so that the caster wheel 31 contacts the ground and pushes the foot 29 off the ground. At this time, the equipment can be moved by rolling the caster wheel 31. When it is needed, the lifting rod 23 is pulled upward and adjusted to the required height. At this time, since the drive rod 34 loses its longitudinal limit, the entire equipment is lowered by gravity, so that the foot 29 is tightly attached to the ground and fixed.
[0098] When this device is in use, if the device needs to be moved, the lifting rod 23 can be adjusted to the lowest retracted position. At this time, the drive rod 34 is pressed down by the lifting rod 23, causing it to move downwards. This pushes the sliding rod 32 outwards via the second connecting rod 33, which in turn pushes the caster wheel 31 downwards via the first connecting rod 30. As the caster wheel 31 contacts the ground, it lifts the foot 29 off the ground. The device can then be moved by the rolling of the caster wheel 31. When the device is needed again, the lifting rod 23 can be pulled upwards and adjusted to the desired height. At this time, because the drive rod 34 loses its longitudinal limit, the entire device is lowered by gravity, causing the foot 29 to... 29. Fit the microphone 4 tightly to the ground to secure it. Then, insert the microphone 4 between the elastic cards 27 inside 10. Use the elasticity of the elastic cards 27 to hold the microphone 4. Next, turn the rotating disk 7. At this time, because the transition disk 8 and the boss 12 are mutually limited by the arc head lever 14 and the slot 15, they cannot rotate. The rotating disk 7 will first rotate relative to the transition disk 8, which will pull the pull rope 16, causing the arc head lever 14 to be pulled out of the slot 15. The transition disk 8 will then rotate until it reaches the desired position. Then, release the rotating disk 7. The arc head lever 14 will be inserted into the slot 15 by the elastic force of the spring 13 to complete the whole process.
[0099] As described above, this application provides a device and method for fault diagnosis of GIS disconnecting switch machinery based on sound waves. The method includes the following steps: acquiring sound wave data of the GIS disconnecting switch machinery during normal operation, extracting standard sound wave acoustic signature features from the sound wave data, and storing the standard sound wave acoustic signature features in a database; acquiring real-time sound waves of the GIS disconnecting switch machinery during operation, and extracting real-time sound wave acoustic signature features from the real-time sound waves; comparing the real-time sound wave acoustic signature features with the standard sound wave acoustic signature features in the database, and outputting fault diagnosis results based on a voting algorithm. This application uses sound waves for fault diagnosis, thereby achieving non-contact fault diagnosis, solving the limitation problem. Furthermore, it employs multi-point sound acquisition during sound acquisition and uses a voting method for result output, then utilizes coefficients based on geometric distance for comprehensive output, thereby increasing the accuracy of the output results.
[0100] It should be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an apparatus, article, or device that includes a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an apparatus, article, or device. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of other identical elements in the apparatus, article, or device that includes that element.
Claims
1. A mechanical fault diagnosis method for GIS disconnect switches based on acoustic waves, characterized in that, Includes the following steps: Acquire acoustic wave data of the GIS disconnect switch during normal mechanical operation, extract standard acoustic wave fingerprint features from the acoustic wave data, and store the standard acoustic wave fingerprint features in the database; Acquire real-time acoustic waves during the mechanical operation of the GIS disconnect switch, and extract real-time acoustic wave acoustic signature features from the real-time acoustic waves. The real-time acoustic signature features are compared with the standard acoustic signature features in the database, and the fault diagnosis results are output based on the voting algorithm.
2. The method for mechanical fault diagnosis of GIS disconnect switches based on acoustic waves according to claim 1, characterized in that, The extraction of standard acoustic signature features from the acoustic data and / or extraction of real-time acoustic signature features from the real-time acoustic data includes the following steps: The collected sound is normalized to eliminate the difference in sound volume and the sound amplitude is reduced proportionally. The sound is pre-emphasized using a first-order digital filter with a frequency of 6dB / pass. The sound wave is windowed and framed to obtain the sound wave signal; The acoustic signal is converted into a spectrum using short-time Fourier transform or linear predictive coding and cepstral method, and standard acoustic signature features or real-time acoustic signature features are extracted from the spectrum.
3. The method for mechanical fault diagnosis of GIS disconnect switches based on acoustic waves according to claim 2, characterized in that, The step of comparing the real-time acoustic wave voiceprint features with the standard acoustic wave voiceprint features in the database includes the following steps: Set an Euclidean distance threshold, calculate the Euclidean distance between the extracted real-time acoustic signature features and the standard acoustic signature features, and compare it with the Euclidean distance threshold. If the distance is less than the Euclidean distance threshold, the output result is "match"; otherwise, the output result is "no match".
4. The method for mechanical fault diagnosis of GIS disconnect switches based on acoustic waves according to claim 3, characterized in that, The frame length of the windowed framing is 10-30ms, and the frame shift is half of the frame length. Alternatively, the windowed framing can use a Hamming window with 256 points per frame and 128 points of overlapping data between adjacent frames.
5. The method for mechanical fault diagnosis of GIS disconnect switches based on acoustic waves according to claim 2, characterized in that, The process of outputting fault diagnosis results based on the voting algorithm includes the following steps: Obtain the Euclidean distances for all outputs that result in "match", denoted as L1, L2, ..., L... n ; Calculate L1, L2...L using the following formulas. n Value: In the formula, k i For coefficients; Compare L1, L2...L in sequence n If the value is less than the Euclidean distance threshold, the fault diagnosis result is output as "match"; otherwise, the fault diagnosis result is output as "mismatch".
6. The method for mechanical fault diagnosis of GIS disconnect switches based on acoustic waves according to claim 5, characterized in that, The k i The Euclidean distance of the standard acoustic wave acoustic signature is L. i The reciprocal of the geometric distance between the sound acquisition module and the GIS disconnect switch, where i = 1, 2, ..., n.
7. A mechanical fault diagnosis device for GIS disconnect switches based on acoustic waves, used to implement the mechanical fault diagnosis method for GIS disconnect switches based on acoustic waves as described in any one of claims 1-6, characterized in that, include: Multiple sound acquisition modules are used to collect sound wave signals when the GIS disconnect switch is working; The data processing module is used to process the acquired signals; Data storage status, used to store data information; The sound acquisition module is a microphone (4), which is fixed on the support structure.
8. The GIS disconnector mechanical fault diagnosis device based on acoustic waves according to claim 7, characterized in that, The support structure includes: a base (1), a lifting strut (2), an angle adjustment disc (3), and a support and walking assembly (5) fixed below the base (1); One end of the lifting support rod (2) is fixedly connected to the base (1), and the other end is fixedly connected to the angle adjustment plate (3). The angle adjustment plate (3) is used to fix the microphone (4). The angle adjustment disk (3) includes a fixed disk (6), a boss (12) welded above the axis of the fixed disk (6), a transition disk (8) rotatably connected to the outside of the boss (12), and a rotating disk (7) rotatably connected to the outside of the transition disk (8). The top outer wall of the transition plate (8) is welded with a fixing ear (17), and the inner side of the fixing ear (17) is rotatably connected to a fixing cylinder (10) via a rotating shaft (11). The inner side of the fixing cylinder (10) is fixed with a plurality of elastic cards (27) arranged in a circular array along the axis of the fixing cylinder (10) via a connecting rod (28).
9. A mechanical fault diagnosis device for GIS disconnect switches based on acoustic waves according to claim 8, characterized in that, The inner wall of the transition plate (8) is slidably connected with multiple arc-head locking rods (14); The arc head clamp (14) is connected to the inner wall of the transition plate (8) through the spring (13), and a pull rope (16) is fixed on the end face of the arc head clamp (14). The other end of the pull rope (16) is fixed to the inner side wall of the rotating plate (7). The side wall of the boss (12) is provided with a groove (15) that matches the arc head clamp (14). The side wall of the fixed ear (17) is provided with a locking assembly (9) for locking the rotating shaft (11). The locking assembly (9) includes a handle (19) connected to the side wall of the fixed ear (17) via an elastic guide rod (18) and a latch (20) welded to the side wall of the handle (19) and inserted into the inner wall of the rotating shaft (11). The inner wall of the rotating shaft (11) is provided with a key-shaped groove (21), and the outer wall of the locking block (20) is provided with a key-shaped protrusion (22) that is in clearance fit with the key-shaped groove (21). The lifting support rod (2) includes a lifting rod (23) welded to the bottom of the fixed plate (6) and a fixed cylinder (25) welded to the top of the base (1). The lifting rod (23) is movably inserted into the inner wall of the fixed cylinder (25), and the side wall of the fixed cylinder (25) is provided with a shrinkage groove (26). The upper side wall of the fixed cylinder (25) is fitted with a clamp (24).
10. A mechanical fault diagnosis device for GIS disconnect switches based on acoustic waves according to claim 9, characterized in that, The supporting walking assembly (5) includes multiple feet (29) fixed to the bottom outer wall of the base (1) and multiple casters (31) that are longitudinally slidably connected to the bottom outer wall of the base (1); The base (1) is slidably connected to the inner wall below the lifting rod (23) by a drive rod (34), and the side wall of the drive rod (34) is rotatably connected to a second connecting rod (33), and the other end of the second connecting rod (33) is rotatably connected to a slide rod (32). The other end of the slide rod (32) is rotatably connected to the first connecting rod (30), the other end of the first connecting rod (30) is rotatably connected to the top of the universal wheel (31), and the slide rod (32) is laterally slidably connected to the inner wall of the base (1).