Circuit breaker voiceprint acquisition orientation correction method based on ultrasonic position calibration

Abstract

The invention discloses a circuit breaker voiceprint acquisition orientation correction method based on ultrasonic position calibration, and the method achieves the real-time space positioning of a moving contact through the deployment of an ultrasonic transmitter at the moving contact of a circuit breaker, the installation of a pre-calibration microphone array in a switch cabinet, and the triggering of an ultrasonic pulse through the switching-on and switching-off actions. On the basis of a positioning result, distance attenuation compensation and directional filtering parameters of the voiceprint signal are dynamically adjusted, acoustic characteristic deviation caused by mechanical vibration or environment temperature change is eliminated, finally, the corrected voiceprint signal is input into a fault diagnosis model, and the detection precision is improved. The method is suitable for on-line monitoring and intelligent operation and maintenance of power switch devices such as high-voltage circuit breakers and GIS equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power equipment fault diagnosis, and in particular to a circuit breaker sound print collection azimuth correction method based on ultrasonic position calibration. BACKGROUND

[0002] The circuit breaker sound print diagnosis technology realizes fault detection by analyzing the acoustic signal characteristics in the opening and closing process, but the existing method has the following technical bottlenecks: 1. Azimuth sensitivity defect: The sound print signal amplitude and frequency domain characteristics are easily affected by the relative position change of the sound source-sensor. The mechanical vibration or installation deviation of the circuit breaker moving contact causes the sound source position to be not fixed, and the traditional fixed gain acquisition system cannot adaptively compensate for the distance attenuation and directivity difference, resulting in inconsistent sound print characteristics of similar faults.

[0003] 2. Environmental interference is not corrected: Temperature fluctuations in the switch cabinet cause sound velocity changes (typical fluctuation range ±10 m / s), resulting in sound source positioning errors based on fixed sound velocity, further reducing the azimuth calibration accuracy.

[0004] 3. Array design limitations: Linear microphone array has azimuth blind area, when the sound source deviates from the array axis, the signal signal-to-noise ratio deteriorates sharply, and it is difficult to capture the effective acoustic characteristics of the main direction of contact movement.

[0005] 4. Insufficient feature stability: The sound print signal without azimuth correction is directly input into the diagnosis model, and the amplitude deviation and noise interference caused by the distance and direction difference cause the model misjudgment rate to rise. SUMMARY

[0006] This part aims to summarize some aspects of the embodiments of the present application and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this part and the abstract and title of the specification to avoid obscuring the purpose of this part, the abstract and the title of the specification. Such simplifications or omissions cannot be used to limit the scope of the present application.

[0007] Therefore, in order to solve the above technical problems, the present application provides the following technical scheme: a circuit breaker sound print collection azimuth correction method based on ultrasonic position calibration, comprising the following steps: S1: deploying a micro ultrasonic transmitter on the surface of the circuit breaker moving contact, and installing a microphone array with pre-calibrated spatial coordinates on the inside of the switch cabinet door; S2: triggering ultrasonic pulse emission by the opening and closing action of the circuit breaker, and synchronously collecting pulse signals and sound print signals by the microphone array; S3: calculating the real-time spatial coordinates of the moving contact according to the pulse arrival time difference and the array geometric relationship, combined with dynamic sound velocity compensation; S4: Based on the real-time coordinates and the spatial azimuth angle of the array center, dynamically adjust the distance attenuation compensation and directional filtering parameters of the acoustic signal; S5: Input the compensated voiceprint signal into the fault diagnosis model and output the result.

[0008] As a preferred embodiment of the circuit breaker acoustic signature acquisition orientation correction method based on ultrasonic position calibration described in this invention, the microphone array is distributed in a tetrahedral topology on the cabinet door, and the coordinates of each sensor are factory calibrated and fixed in place (calibration error ≤ 0.2 mm); a temperature and humidity sensor is added to the switch cabinet door, based on real-time temperature... Dynamically corrected speed of sound The calculation formula is: (Unit: m / s); where The reference speed of sound is 331.3. This is the temperature compensation coefficient, with a value of 0.606.

[0009] As a preferred embodiment of the circuit breaker acoustic signature acquisition orientation correction method based on ultrasonic position calibration described in this invention, wherein: the real-time spatial coordinate calculation of the moving contact in step S3 includes: Perform matched filtering on the microphone array signal: Extract pulse arrival time ,in It is a 13-bit Barker code template signal (encoded sequence: +1+1+1+1-1-1+1+1-1+1-1+1). Calculate the relative time difference of arrival based on a reference sensor. ; Will Array coordinates and sound speed Substitute into the hyperbola positioning equations: ; in, For microphone indexing, For pre-calibrated coordinates; The coordinates of the moving contact are solved using a nonlinear optimization algorithm. This includes iterative solutions using the Newton-Raphson method or the least squares method, with the array's geometric center as the initial point.

[0010] As a preferred embodiment of the circuit breaker acoustic signature acquisition orientation correction method based on ultrasonic position calibration described in this invention, wherein: the nonlinear optimization algorithm sets a convergence threshold. The iterative formula is: ,in, For the residual vector, Set the Jacobian matrix; set a convergence threshold to ensure positioning accuracy. .

[0011] In a preferred embodiment of the method described in this invention, the dynamic adjustment in step S4 includes: Distance attenuation compensation, based on real-time distance Normalized according to the inverse square law of sound pressure: ;in, For reference distance; Directional filtering, based on azimuth angle Calculate the directional gain weights: ;in, , This is the gain coefficient, used to enhance the frequency band in the main direction of motion. signal when hour, .

[0012] In a preferred embodiment of the method described in this invention, the fault diagnosis in step S5 includes: Voiceprint extraction after compensation Mel frequency cepstral coefficient characteristics in the (2~8kHz) band; the 2~8kHz band covers the characteristic frequency components of the circuit breaker's mechanical operation (such as high-frequency harmonics of contact collision), and avoids 50 Hz power frequency interference inside the cabinet; The fault type and confidence level are classified and output using a convolutional neural network with 3 convolutional layers and 2 fully connected layers.

[0013] As a preferred embodiment of the method described in this invention, the extraction of Mel frequency cepstral coefficient features includes: a) Pre-emphasize and frame window the acoustic signature signal in the 2~8kHz frequency band; b) Calculate the logarithmic energy using a Mel-scale filter bank; c) The cepstral coefficients are obtained through discrete cosine transform.

[0014] This invention addresses the aforementioned technical bottlenecks through ultrasonic position calibration and dynamic signal compensation mechanisms, with the following specific effects: 1. This invention is based on real-time positioning technology of moving contact of ultrasonic pulse, which dynamically calculates the distance and azimuth angle between the sound source and the sensor; through sound pressure square ratio compensation and directional gain weighting, it eliminates the acoustic feature fluctuation caused by mechanical offset, and greatly improves the similarity of acoustic features of the same fault at different locations.

[0015] 2. This invention monitors the temperature inside the cabinet in real time using a temperature and humidity sensor and uses a linear sound velocity model to dynamically compensate for the sound wave propagation speed; it eliminates the sound velocity deviation caused by temperature and greatly reduces positioning errors.

[0016] 3. This invention uses a tetrahedral topology microphone array to overcome the directional blind zone limitation of traditional linear arrays; through multi-angle non-coplanar distribution, it achieves omnidirectional sound field coverage for circuit breaker operation, and greatly improves the signal gain in the main direction (80°-100°).

[0017] 4. In this invention, the MFCC feature variance is greatly reduced by performing distance normalization and directional filtering on the position-corrected voiceprint signal; the classification confidence of the convolutional neural network for the compensated features is improved, and the misclassification rate is reduced. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a schematic diagram of the workflow of the present invention. Detailed Implementation

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0020] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0021] 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 phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0022] Reference Figure 1 This is the first embodiment of the present invention, which provides a method for directional correction of circuit breaker acoustic signature acquisition based on ultrasonic position calibration. Specific application examples are as follows: 1. Hardware deployment and parameter configuration; (1) Miniature ultrasonic transmitter; The mechanical action of the circuit breaker opening and closing is used to synchronously trigger pulses, ensuring the timing consistency between calibration and voiceprint acquisition, and providing a positioning beacon for the microphone array; Installation location: Fixed on the surface of the moving contact of the circuit breaker, with an installation error of ≤ ±0.5mm.

[0023] Technical parameters: Frequency: 20kHz (adjustable range 18–22kHz), preferably 20kHz to balance penetration and electromagnetic interference resistance; Pulse width: 100μs (adjustable 80–120μs) to ensure concentrated pulse energy and stable response of the matched filter.

[0024] Encoding method: 13-bit Barker code (+1+1+1+1-1-1+1+1-1+1-1+1), which can improve the signal-to-noise ratio by ≥15dB through relevant detection.

[0025] (2) MEMS microphone array; It is used to receive ultrasonic pulses (for positioning) and collect circuit breaker acoustic patterns (for diagnosis), with simultaneous acquisition of dual-mode signals and hardware multiplexing to reduce costs; Topology: Tetrahedrals are arranged inside the cabinet door, and through multi-angle spatial distribution (non-coplanar), the directional blind spots of traditional linear arrays are avoided, covering the omnidirectional sound field for circuit breaker operation. Specifically, the coordinates are pre-calibrated (factory calibration error ≤ 0.2mm). Sensor1: (0, 0) mm; Sensor2: (100, 0) mm; Sensor3: (50, 86.6) mm; Sensor4: (50, 28.9) mm; Sampling rate: 100kHz (satisfies the Nyquist criterion, time resolution 10μs, corresponding distance resolution 3.5mm).

[0026] Dynamic range: ≥120dB, ensuring coverage of the opening and closing acoustic intensity of circuit breakers (typical value 60–110dB).

[0027] (3) Temperature and humidity sensor; Used for real-time monitoring of cabinet temperature T, dynamic correction of sound velocity c, elimination of sound wave propagation speed deviation caused by ambient temperature (e.g., sound velocity error reaches 10.2m / s at 30℃, directly affecting positioning accuracy), and ensuring positioning accuracy; Temperature measurement range: 0–50℃, accuracy ±0.5℃, used for real-time correction of sound velocity.

[0028] Humidity effect: In this embodiment, humidity compensation is ignored, and the speed of sound is calculated based solely on temperature.

[0029] The above hardware layout offers the following technical advantages: Automatic triggering: No additional excitation device is required; it is triggered by the opening and closing current signal, reducing system complexity.

[0030] Electromagnetic interference resistance: The 20 kHz ultrasonic frequency band avoids the main electromagnetic noise in the switch cabinet (such as 50 Hz power frequency and its harmonics).

[0031] 2. Azimuth determination and coordinate calculation; (1) Ultrasonic pulse triggering and acquisition; When the circuit breaker trips, the rising edge of the drive coil current (>5A / ms) triggers the ultrasonic transmitter to emit a 1300μs encoded pulse.

[0032] The microphone array synchronously acquires signals, and matched filtering is performed on each channel signal: ; : The signal received by the i-th channel (including noise and pulse); Barker code template signal (known encoded sequence); Function: To determine the peak position of cross-correlation Detect pulse arrival time The time detection error is ≤1μs.

[0033] Application manifestation: Precise time stamping: By suppressing environmental noise (such as mechanical vibration and electromagnetic interference) through matched filtering, the time detection error is reduced from ±5μs in the traditional threshold method to ±1μs.

[0034] Example data: (Sensor 1); (Sensor 2); (2) Calculation of TDOA time difference; Using Sensor 1 as a reference, calculate the relative time difference: , , ; Example data: , , , ; Corresponding time difference: , , ; (3) Dynamic compensation for sound velocity; Real-time sound speed calculation (linear model): (Unit: m / s); m / s: This value is the theoretical speed of sound of dry air at 0°C (ideal gas model). : First-order coefficients of the Taylor expansion of sound speed as a function of temperature.

[0035] Example: When the temperature inside the cabinet is T=30℃, ; Corresponding path difference: ; This step calculates the accurate speed of sound based on real-time temperature, ensuring that subsequent TDOA and distance calculation errors are minimized.

[0036] (4) Calculation of the coordinates of the moving contact; Establish the system of equations for the hyperbola's position: enter: Microphone coordinates (e.g., Sensor 1: (0, 0), Sensor 2: (100, 0), Sensor 3: (50, 86.6) mm, Sensor 4: (50, 28.9) mm).

[0037] Time difference (example value): , speed of sound ; Equation establishment: ; The Newton-Raphson iterative solution process: Initial guess: (Array geometric center); Residual function vector: : ; Jacobian matrix: (Partial derivative matrix): ; Iteration formula and convergence condition: ; Where k represents the current iteration number; , : The estimated position at the k-th iteration, i.e., the estimated horizontal and vertical coordinates of the current moving contact; , The updated coordinates after the (k+1)th iteration represent a new, more accurate position estimate. The function vector formed by the simultaneous equations of the hyperbola positioning represents the "residual" or "error vector" under the current estimation. The inverse of the Jacobian matrix, used to correct the current estimated position; Convergence condition: Change in position Or the maximum number of iterations reaches 100.

[0038] Example output: Iteration output results: Positioning error ≤ 1.2mm.

[0039] Application manifestation: Orientation calibration: Real-time spatial coordinates It reflects the spatial position of the moving contact and eliminates changes in the relative position of the sensor and the sound source caused by mechanical vibration or installation misalignment; Error control: Tetrahedral array + temperature compensation limits the positioning error to within ±1.5mm (corresponding to an angular error of ±1.8°).

[0040] 3. Voiceprint signal compensation and diagnosis; (1) Azimuth calculation: Array center coordinates ; Azimuth calculation formula: ; Example: , ; In this step, the azimuth angle is determined to characterize the direction of the sound source relative to the array, and is used to dynamically adjust the directional gain and filtering design of the acoustic signal.

[0041] Distance attenuation compensation: Actual distance: 71.1mm (actual distance); Sound pressure level compensation formula: ; in: (Reference distance); →After compensation ; Effect: The 85 dB sound pressure level collected at 71.1 mm was normalized to 43.0 dB at 100 mm, eliminating the influence of distance on the acoustic amplitude; In this step, distance attenuation compensation is used to eliminate the sound pressure attenuation deviation caused by the difference in distance between different sensors and the sound source, thereby enhancing the comparability of feature extraction.

[0042] (3) Directional filtering; Gain function: ; This formula is used to enhance the signal in the main direction (80°-100°) of the head movement and suppress lateral reflection noise; when hour, There is almost no attenuation; when hour, Suppressing non-dominant direction noise; In this step, directional gain weighting is used to dynamically weight the signals of each channel according to the direction of the sound source, highlighting the true direction component of the sound source and suppressing reflection and lateral noise; Technical advantages: Feature stability: The variance of the amplitude frequency domain features (such as MFCC) of the compensated voiceprint is reduced by 60%, improving the robustness of the diagnostic model.

[0043] (4) Fault diagnosis execution; The compensated voiceprint is input into a pre-trained CNN model (structure: 3 convolutional layers + 2 fully connected layers).

[0044] Example of output: Normal tripping probability: 92%; Probability of contact jamming failure: 76%.

[0045] Application manifestation: Location independence: After coordinate calibration and signal compensation, the similarity of voiceprint features of the same fault in different locations increased from 0.35 to 0.82 (cosine similarity).

[0046] 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 it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0047] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of this application can be implemented in various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0048] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0049] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0050] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0051] 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 it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for directional correction of circuit breaker acoustic signature acquisition based on ultrasonic position calibration, characterized in that: Includes the following steps: S1: Deploy a miniature ultrasonic transmitter on the moving contact of the circuit breaker and install a microphone array with pre-calibrated coordinates on the inside of the switch cabinet door; S2: The opening and closing action of the circuit breaker triggers the emission of ultrasonic pulses, and the microphone array synchronously collects the pulse signals; S3: Calculate the real-time spatial coordinates of the moving contact based on the pulse arrival time difference and array geometry, combined with dynamic sound velocity compensation; S4: Based on real-time spatial coordinates and the spatial azimuth angle of the array center, dynamically adjust the distance attenuation compensation and directional filtering parameters of the acoustic signal; S5: Input the compensated voiceprint signal into the fault diagnosis model and output the result.

2. The method for directional correction of circuit breaker acoustic signature acquisition based on ultrasonic position calibration as described in claim 1, characterized in that: The microphone array is distributed in a tetrahedral topology on the cabinet door, and the coordinates of each sensor are factory calibrated and fixed; temperature and humidity sensors are also arranged on the switch cabinet door to monitor the ambient temperature inside the cabinet in real time and dynamically correct the sound velocity.

3. The method for directional correction of circuit breaker acoustic signature acquisition based on ultrasonic position calibration as described in claim 2, characterized in that: In step S3, calculating the real-time spatial coordinates of the moving contact includes: Perform matched filtering on the pulse signals acquired by the microphone array to extract the pulse arrival time of each channel; Using the reference sensor as a benchmark, calculate the relative arrival time difference of the other sensors; Substituting the time difference, array coordinates, and temperature-compensated sound velocity into the hyperbolic positioning equations, the coordinates of the moving contact are solved using a nonlinear optimization algorithm.

4. The method for directional correction of circuit breaker acoustic signature acquisition based on ultrasonic position calibration as described in claim 3, characterized in that: Nonlinear optimization algorithms include: Using the geometric center of the array as the initial iteration point, the spatial coordinates are calculated by Newton-Raphson method or least squares fusion. Set a convergence threshold of ≤0.01mm and a maximum number of iterations constraint.

5. The method for directional correction of circuit breaker acoustic signature acquisition based on ultrasonic position calibration as described in claim 4, characterized in that: In step S4, dynamically adjusting the voiceprint signal parameters includes: Distance attenuation compensation: Based on the real-time distance between the moving contact and the microphone, the sound pressure level is normalized to the reference distance sound pressure level according to the inverse square law of sound pressure. Directional filtering: Calculate directional gain weights based on azimuth angles to enhance the frequency band signal in the main direction of contact movement.

6. The method for directional correction of circuit breaker acoustic signature acquisition based on ultrasonic position calibration as described in claim 5, characterized in that: In step S5, the dynamic sound velocity compensation specifically involves calculating the sound velocity using a linear model based on the real-time temperature, where the reference sound velocity is... The temperature coefficient is .

7. The method for directional correction of circuit breaker acoustic signature acquisition based on ultrasonic position calibration as described in claim 6, characterized in that: The above directional gain weights are calculated as follows: A sinusoidal function model is used, with a gain of 1 when the azimuth angle is 90°, and the gain decreases according to the function curve when the azimuth angle deviates.

8. The method for directional correction of circuit breaker acoustic signature acquisition based on ultrasonic position calibration as described in claim 7, characterized in that: In step S5, the following processing is performed before the voiceprint signal is input into the fault diagnosis model: Mel frequency cepstral coefficient features in the 2-8 kHz frequency band were extracted from the compensated voiceprint signal; The fault type and confidence level are classified and output using a convolutional neural network.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

Images