Abnormal sound detection method and system for air conditioner production line

Abstract

The invention provides an abnormal sound detection method and system for an air conditioner production line. The abnormal sound detection method comprises the steps that sound signals are obtained in real time through a microphone spherical array; processing the sound signal based on an HOA-SHT domain analysis framework to generate a panoramic sound image; obtaining an air conditioner image through a panoramic camera; locking an air conditioner area according to the air conditioner image; multi-modal positioning is conducted on the air conditioner area through the panoramic sound image, and a target area is obtained; performing sound signal extraction on the target area through a multi-scale Mel spectrum, and inputting the extracted sound signal into a beam former of a neural network to obtain a noise signal; the noise signals are detected and recognized through an anomaly classifier based on the convolutional neural network, and an air conditioner abnormal sound classification result is obtained; according to the invention, efficient and accurate detection of the abnormal sound of the air conditioner hanging unit is realized, the limitation of a traditional manual detection method is effectively overcome, the production efficiency is remarkably improved, the false detection rate and the omission ratio are reduced, and powerful guarantee is provided for the production line quality.

Description

Technical Field

[0001] The present invention relates to the technical field of air conditioner detection, and particularly to a method and system for detecting abnormal noises in an air conditioner production line. Background Art

[0002] As an important part of the air conditioner system, the wall-mounted air conditioner plays a key role in the processes of refrigeration, heating, and air circulation. Among them, the wind wheel blades, as one of the core components of the wall-mounted air conditioner, directly affect the operating efficiency of the air conditioner and the user experience.

[0003] During the production process of the wall-mounted air conditioner, the manufacturing and assembly of the wind wheel blades are crucial. However, in actual production, due to the influence of factors such as raw materials, production environment, and technological processes, foreign matters may adhere to the surface of the wind wheel blades, such as dust, particulate matter, oil stains, or other production residues. The attachment of these foreign matters on the surface of the wind wheel blades may cause abnormal vibration during the operation of the blades, thus leading to noise problems.

[0004] When the air conditioner is operating, the rotation speed of the wind wheel blades is relatively high. Especially in the refrigeration or heating mode, the high-speed rotation of the blades easily brings the attached foreign matters into the air flow, resulting in friction or collision between the blades and the foreign matters, generating additional noise. This noise may be masked by environmental noise during the day, but at night, when the environmental noise is low, the weak noise may become extremely obvious, seriously affecting the user's sleep quality and living experience.

[0005] Existing methods mainly rely on manual listening and screening. Workers can quickly conduct a preliminary screening of the wall-mounted air conditioners on the production line through auditory perception of abnormal noises and promptly detect abnormalities. This method has the characteristics of low cost, simple operation, and quick judgment. However, this method has significant limitations: firstly, the judgment basis mainly relies on the auditory perception and experience of workers, lacking a unified quantitative standard, and the results are easily affected by individuals and are highly subjective; secondly, it is impossible to accurately measure and record the type, intensity, or position of abnormal noises, making it difficult to form objective data support; in addition, workers are prone to fatigue during long-term high-intensity work, resulting in a decrease in the judgment accuracy rate and a high false detection rate; finally, manual listening and screening requires workers to check one by one, with low efficiency and difficulty in meeting the high-tempo requirements of large-scale production. Summary of the Invention

[0006] In view of this, the purpose of the present invention is to provide a method and system for detecting abnormal noises in an air conditioner production line, realizing efficient and accurate detection of abnormal noises in wall-mounted air conditioners, effectively overcoming the limitations of traditional manual detection methods, significantly improving production efficiency, reducing the false detection rate and missed detection rate, and providing a strong guarantee for the quality of the production line.

[0007] In a first aspect, an embodiment of the present invention provides a method for detecting abnormal noises in an air conditioner production line, the method comprising:

[0008] Obtain sound signals in real time through a microphone spherical array;

[0009] Process the sound signals based on the HOA-SHT domain analysis framework to generate a panoramic sound image;

[0010] Obtain an air conditioner image through a panoramic camera;

[0011] Lock the air conditioner area according to the air conditioner image;

[0012] Perform multi-modal positioning on the air conditioner area through the panoramic sound image to obtain a target area;

[0013] Extract sound signals from the target area through multi-scale Mel spectrograms, and input the extracted sound signals into the beamformer of a neural network to obtain noise signals;

[0014] Detect and identify the noise signals based on an anomaly classifier of a convolutional neural network to obtain an air conditioner abnormal sound classification result.

[0015] Further, processing the sound signals based on the HOA-SHT domain analysis framework to generate a panoramic sound image includes:

[0016] Preprocess the sound signals through Hanning window short-time Fourier transform to obtain preprocessed sound signals;

[0017] Perform spherical harmonic domain transformation on the preprocessed sound signals to obtain a 7th-order surround sound field decomposition;

[0018] Perform spherical harmonic coefficient energy compensation on the 7th-order surround sound field decomposition to obtain spherical harmonic domain features;

[0019] Pass the spherical harmonic domain features through a beamformer to obtain a sound field;

[0020] Pass the sound field through a panoramic beam space to obtain respective spatial azimuth features;

[0021] Overlay the respective spatial azimuth features and the air conditioner image to obtain the panoramic sound image.

[0022] Further, locking the air conditioner area according to the air conditioner image includes:

[0023] Input the air conditioner image into a target detection model to obtain the spatial position of the air conditioner;

[0024] Lock the air conditioner area according to the spatial position of the air conditioner.

[0025] Further, the method further includes:

[0026] Monitor the operation status of the production line in real time through the panoramic camera;

[0027] When the operation status of the production line is the shutdown state, control the suspension of audio and video acquisition.

[0028] Furthermore, the method further includes:

[0029] Prompt the classification result of the air conditioner abnormal sound through voice broadcast.

[0030] Furthermore, the classification result of the air conditioner abnormal sound includes foreign object adhesion, blade burr, and sheet metal vibration.

[0031] In a second aspect, an embodiment of the present invention provides an abnormal sound detection system for an air conditioner production line. The system includes a computing server and a spherical panoramic audio-video device. The spherical panoramic audio-video device includes a microphone spherical array and a panoramic camera; the computing server includes a processing module, a locking module, a multi-modal positioning module, an extraction module, a detection and recognition module;

[0032] The microphone spherical array is used to obtain sound signals in real time;

[0033] The processing module is used to process the sound signals based on the HOA-SHT domain analysis framework to generate a panoramic audio image;

[0034] The panoramic camera is used to obtain air conditioner images;

[0035] The locking module is used to lock the air conditioner area according to the air conditioner image;

[0036] The multi-modal positioning module is used to perform multi-modal positioning on the air conditioner area through the panoramic audio image to obtain a target area;

[0037] The extraction module is used to extract sound signals from the target area through multi-scale Mel spectrograms and input the extracted sound signals into the beamformer of the neural network to obtain noise signals;

[0038] The detection and recognition module is used to detect and recognize the noise signals based on the anomaly classifier of the convolutional neural network to obtain the classification result of the air conditioner abnormal sound.

[0039] Furthermore, the processing module is specifically used for:

[0040] Preprocess the sound signals through Hanning window short-time Fourier transform to obtain preprocessed sound signals;

[0041] Perform spherical harmonic domain transformation on the preprocessed sound signals to obtain a 7th-order spherical harmonic field decomposition;

[0042] Decompose the 7th-order surround sound field and perform spherical harmonic coefficient energy compensation to obtain spherical harmonic domain features;

[0043] Pass the spherical harmonic domain features through a beamformer to obtain a sound field;

[0044] Pass the sound field through a panoramic beam space to obtain various spatial orientation features;

[0045] Superimpose the various spatial orientation features and the air conditioner image to obtain the panoramic sound image.

[0046] In a third aspect, an embodiment of the present invention provides an electronic device, including a memory and a processor. A computer program that can run on the processor is stored on the memory. When the processor executes the computer program, the method described above is implemented.

[0047] In a fourth aspect, an embodiment of the present invention provides a computer-readable medium having non-volatile program code executable by a processor. The program code causes the processor to execute the method described above.

[0048] An embodiment of the present invention provides a method and system for detecting abnormal sounds in an air conditioner production line, including: obtaining sound signals in real time through a microphone spherical array; processing the sound signals based on the HOA-SHT domain analysis framework to generate a panoramic sound image; obtaining an air conditioner image through a panoramic camera; locking the air conditioner area according to the air conditioner image; performing multi-modal positioning on the air conditioner area through the panoramic sound image to obtain a target area; extracting sound signals from the target area through multi-scale Mel spectrograms and inputting the extracted sound signals into a beamformer of a neural network to obtain noise signals; detecting and identifying the noise signals based on an abnormal classifier of a convolutional neural network to obtain an air conditioner abnormal sound classification result; realizing efficient and accurate detection of abnormal sounds of an air conditioner hanging unit, effectively overcoming the limitations of traditional manual detection methods, significantly improving production efficiency, reducing the false detection rate and the missed detection rate, and providing a strong guarantee for the quality of the production line.

[0049] Other features and advantages of the present invention will be described in the following description, and in part will be obvious from the description, or will be understood by implementing the present invention. The objectives and other advantages of the present invention are achieved and obtained by the structures specifically pointed out in the description, the claims, and the drawings.

[0050] To make the above objectives, features, and advantages of the present invention more obvious and understandable, the following specifically lists preferred embodiments and, in conjunction with the accompanying drawings, makes a detailed description as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0052] Figure 1 Flowchart of the abnormal noise detection method for an air conditioner production line provided in Embodiment 1 of the present invention;

[0053] Figure 2 Schematic diagram of a spherical panoramic sound imager provided in Embodiment 1 of the present invention;

[0054] Figure 3 Schematic diagram of the abnormal noise detection process for an air conditioner production line provided in Embodiment 1 of the present invention;

[0055] Figure 4 Schematic diagram of the abnormal noise detection system for an air conditioner production line provided in Embodiment 2 of the present invention;

[0056] Figure 5 Schematic diagram of another abnormal noise detection system for an air conditioner production line provided in Embodiment 2 of the present invention. Specific embodiments

[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions of the present invention with reference to the accompanying drawings. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0058] In the prior art, foreign objects may adhere to the surface of the wind wheel blades, such as dust, particulate matter, oil stains, or other production residues. The attachment of these foreign objects to the surface of the wind wheel blades may cause abnormal vibration of the blades during operation, thereby triggering noise problems.

[0059] During the operation of the air conditioner, the rotation speed of the wind wheel blades is relatively high. Especially in the cooling or heating mode, the high-speed rotation of the blades easily brings the attached foreign objects into the air flow, resulting in friction or collision between the blades and the foreign objects, generating additional noise. This noise may be masked by environmental noise during the day, but at night, when the environmental noise is low, the weak noise may also become extremely obvious, seriously affecting the user's sleep quality and life experience.

[0060] Therefore, how to effectively avoid foreign objects sticking to the impeller blades during the production process of wall-mounted air conditioners, and how to reduce the blade noise problem caused by foreign objects have become technical problems that air conditioner manufacturing enterprises urgently need to solve. This application provides a method and system for detecting abnormal sounds in an air conditioner production line, which not only helps to improve the quality of air conditioner products, but also significantly improves the user experience, and has important practical significance and market value.

[0061] For the convenience of understanding this embodiment, the embodiments of the present invention will be introduced in detail below.

[0062] Embodiment 1:

[0063] Figure 1 It is a flowchart of a method for detecting abnormal sounds in an air conditioner production line provided in Embodiment 1 of the present invention.

[0064] Referring to Figure 1 , the method includes the following steps:

[0065] Step S101, obtaining sound signals in real time through a microphone spherical array;

[0066] Step S102, processing the sound signals based on the HOA-SHT domain analysis framework to generate a panoramic sound image;

[0067] Step S103, obtaining an air conditioner image through a panoramic camera;

[0068] Step S104, locking the air conditioner area according to the air conditioner image;

[0069] Step S105, performing multi-modal positioning on the air conditioner area through the panoramic sound image to obtain a target area;

[0070] Step S106, extracting sound signals from the target area through multi-scale Mel spectrograms, and inputting the extracted sound signals into the beamformer of the neural network to obtain noise signals;

[0071] Step S107, detecting and identifying the noise signals based on the anomaly classifier of the convolutional neural network to obtain the classification result of the abnormal sound of the air conditioner.

[0072] In this embodiment, the spherical panoramic sound imager includes a microphone spherical array and a panoramic camera; by using the spatially uniform directivity of the microphone spherical array, the interference of background noise is effectively suppressed. By deploying a microphone spherical array on the production line, sound signals are obtained in real time. At the same time, the panoramic camera captures the spatial information of the air conditioner, and combines with the target detection model (YOLOv11) to accurately judge the operation state of the production line and locate the position of the air conditioner. Subsequently, the system extracts sound signals from the target area, and uses the beamformer based on the neural network to accurately extract weak noise signals. Finally, the abnormal classifier is used to detect and identify abnormal sounds, realizing efficient and accurate detection of abnormal sounds of wall-mounted air conditioners.

[0073] Referring to Figure 2 , the spherical panoramic sound imager includes a microphone spherical array and a panoramic camera. The microphone spherical array includes an acoustic acquisition array, a visual assistance unit, and an intelligent processing platform.

[0074] The main technical features are as follows: The sound field acquisition array is a spherical acoustic array based on 64-channel vector microphones. It realizes spatial sound field reconstruction through a precise three-dimensional equiangular distribution algorithm, supports 7th-order Ambisonics high-order surround sound field modeling, and has a spatially uniform directivity. Its unique gradient phase compensation algorithm can effectively suppress multipath reflection and environmental noise. Combined with adaptive beamforming technology, it realizes the directional enhancement extraction of the target sound source of the air conditioner, and the azimuth resolution reaches ±3°.

[0075] The spherical panoramic sound imager also includes a main control unit. The main control unit adopts the FPGA Xilinx ZYNQ UltraScale+ MPSoC platform, integrating real-time signal processing and logic control functions. The audio processing link adopts Cirrus Logic CS5348 multi-channel ADC (24bit / 192kHz), dynamic range expansion processing (DRC 120dB), and digital anti-aliasing filter (FIR 512 taps).

[0076] The spherical panoramic sound imager also includes a panoramic camera. The panoramic camera includes 4 IMX477 cameras (12 million pixels, 60fps) and H.265 hardware encoding (4K@30fps). The time synchronization system supports GPS / Beidou dual-mode time synchronization (±50ns accuracy) and PTPv2 network clock synchronization. The system operates in a 200MHz main clock domain, realizes synchronous acquisition of audio and video through the DMA architecture, the audio stream is transmitted through the JESD204B interface, and the TDM time division multiplexing mode is adopted. The video data is transmitted through the MIPI CSI-2 interface, and YUV422 10bit quantization is used.

[0077] This system can realize the separation of the target sound source through acoustic-optical joint calibration (calibration error <0.1°) and intelligent noise reduction algorithm, and meet the needs of accurate sound field analysis of the air conditioner noise in the production line under complex sound field environments (signal-to-noise ratio ≥ -10dB).

[0078] Further, step S102 includes the following steps:

[0079] Step S201, preprocess the acoustic signal through Hanning window short-time Fourier transform to obtain the preprocessed acoustic signal;

[0080] Step S202, perform spherical harmonic domain transformation on the preprocessed acoustic signal to obtain 7th-order surround sound field decomposition;

[0081] Step S203: Decompose the 7th-order surround sound field and perform spherical harmonic coefficient energy compensation to obtain the spherical harmonic domain features;

[0082] Step S204: Pass the spherical harmonic domain features through a beamformer to obtain the sound field;

[0083] Step S205: Pass the sound field through the panoramic beam space to obtain the spatial orientation features in each direction;

[0084] Step S206: Superimpose the spatial orientation features in each direction and the air conditioner image to obtain the panoramic sound image.

[0085] Specifically, this application adopts an acoustic-optic fusion detection architecture and implements a multi-modal abnormal acoustic detection method based on the HOA-SHT domain analysis framework. Refer to Figure 3 , this algorithm first performs a joint characterization of the sound field, preprocesses the sound signal through the Hann window short-time Fourier transform (window length 512 points, overlap rate 75%), and obtains the 7th-order surround (Ambisonics) sound field decomposition (64-channel B-format encoding) using the spherical harmonic domain transform (SHT); and decomposes the 7th-order surround sound field and performs spherical harmonic coefficient energy compensation to obtain the spherical harmonic domain features.

[0086] Subsequently, an improved object detection model (YOLOv11) is used for multi-modal localization, and accurate locking of the target area is achieved through panoramic vision guidance. Among them, air conditioner images in multiple directions are collected through a panoramic camera, the air conditioner images in multiple directions can be stitched, and then the stitched image is input into the object detection model.

[0087] On this basis, the system uses a spherical harmonic domain beamformer to dynamically adjust the spatial filter bank for continuous tracking. In the feature extraction stage, the system uses multi-scale Mel spectrograms to extract the frequency domain features of the air conditioner signal, and combines a classifier based on a convolutional neural network and a hybrid attention mechanism to achieve accurate detection and recognition of the types of abnormal sounds of the air conditioner. Finally, the system reports the classification results of the abnormal sounds of the air conditioner to the production line workers in real time, providing them with a timely basis for troubleshooting.

[0088] Further, step S104 includes the following steps:

[0089] Step S301: Input the air conditioner image into the object detection model to obtain the spatial position of the air conditioner;

[0090] Step S302: Lock the air conditioner area according to the spatial position of the air conditioner.

[0091] Further, this method further includes the following steps:

[0092] Step S401: Real-time monitor the operation status of the production line through a panoramic camera;

[0093] Step S402: When the production line is in a stopped state, control the suspension of audio-visual acquisition.

[0094] Furthermore, the method further includes the following steps:

[0095] Step S501: Prompt the classification result of air conditioner abnormal sounds through voice broadcast.

[0096] Furthermore, the classification results of air conditioner abnormal sounds include foreign object adhesion, blade burrs, and sheet metal vibration.

[0097] Specifically, the panoramic camera can not only provide an auxiliary positioning function for the sound field but also be used to monitor the operation status of the production line in real time (such as stopped or running state). When it detects that the production line is in a stopped state, the system will automatically pause the audio-visual acquisition. In addition, the panoramic camera can capture the image of the air conditioner, determine the spatial position of the air conditioner in combination with the target detection algorithm YOLOv11, so as to lock the target area through the panoramic sound imager and extract and enhance the air conditioner impeller noise signal.

[0098] Compared with the traditional manual listening and screening method, this application adopts an innovative panoramic acoustic perception technology, combined with an acoustic-optic fusion detection architecture, which significantly improves the anti-interference performance of the system. At the same time, through the introduction of the AI vision guidance mechanism of the improved YOLOv11 algorithm, the rapid positioning and tracking of the target area are realized, so as to efficiently extract the air conditioner noise signal.

[0099] On this basis, the present invention also adopts an advanced AI abnormal sound detection model to further optimize the performance indicators of abnormal sound detection. Through these technical means, this application effectively overcomes the limitations of the traditional manual detection method, significantly improves the production efficiency, reduces the false detection rate and the missed detection rate, and provides a strong guarantee for the quality of the production line.

[0100] Embodiment 2:

[0101] Figure 4 It is a schematic diagram of the abnormal sound detection system for the air conditioner production line provided by the second embodiment of the present invention.

[0102] Referring to Figure 4 , the system includes a computing server and a spherical panoramic sound imager. The spherical panoramic sound imager includes a microphone spherical array and a panoramic camera; referring to Figure 5 , the computing server includes a processing module, a locking module, a multi-modal positioning module, an extraction module, a detection and recognition module;

[0103] The microphone spherical array is used to obtain sound signals in real time;

[0104] The processing module is used to process the sound signal based on the HOA-SHT domain analysis framework to generate a panoramic sound image;

[0105] A panoramic camera for acquiring air conditioner images;

[0106] A locking module for locking the air conditioner area based on the air conditioner image;

[0107] A multi-modal positioning module for performing multi-modal positioning on the air conditioner area through panoramic audio and video to obtain a target area;

[0108] An extraction module for extracting acoustic signals from the target area through multi-scale Mel spectrograms and inputting the extracted acoustic signals into a beamformer of a neural network to obtain noise signals;

[0109] A detection and recognition module for detecting and recognizing the noise signals based on an anomaly classifier of a convolutional neural network to obtain an air conditioner abnormal sound classification result.

[0110] Specifically, in Figure 4 , the system includes a computing server and a spherical panoramic audio and video device. The spherical panoramic audio and video device includes a microphone spherical array and a panoramic camera, which are installed above the production line.

[0111] The system further includes a label scanner for obtaining the air conditioner ID through the label scanner and associating it with the acoustic signal. Finally, the computing server detects and recognizes the noise signal to obtain an air conditioner abnormal sound classification result; and notifies the production line workers of the air conditioner abnormal sound classification result through voice broadcast for timely troubleshooting and handling of faulty air conditioners.

[0112] Furthermore, the processing module is specifically configured to:

[0113] Preprocess the acoustic signal through Hanning window short-time Fourier transform to obtain a preprocessed acoustic signal;

[0114] Perform spherical harmonic domain transformation on the preprocessed acoustic signal to obtain a 7th-order ambisonic field decomposition;

[0115] Perform spherical harmonic coefficient energy compensation on the 7th-order ambisonic field decomposition to obtain spherical harmonic domain features;

[0116] Pass the spherical harmonic domain features through a beamformer to obtain a sound field;

[0117] Pass the sound field through a panoramic beam space to obtain respective spatial azimuth features;

[0118] Overlay the respective spatial azimuth features and the air conditioner image to obtain a panoramic audio and video.

[0119] An embodiment of the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of the abnormal sound detection method for an air conditioner production line provided in the above embodiment are implemented.

[0120] An embodiment of the present invention further provides a computer-readable medium having non-volatile program code executable by a processor. A computer program is stored on the computer-readable medium. When the computer program is run by the processor, it executes the steps of the abnormal sound detection method for the air conditioner production line in the above embodiment.

[0121] The computer program product provided by the embodiment of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the method described in the foregoing method embodiments. For specific implementation, reference can be made to the method embodiments, which will not be elaborated herein.

[0122] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems and devices described above can refer to the corresponding processes in the foregoing method embodiments, which will not be elaborated herein.

[0123] In addition, in the description of the embodiments of the present invention, unless otherwise clearly specified and limited, the terms "installation", "connection", and "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0124] If the above function is implemented in the form of a software function unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program code.

[0125] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation to the present invention. In addition, the terms "first", "second", and "third" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0126] Finally, it should be noted that the above-described embodiments are only specific embodiments of the present invention, used to illustrate the technical solutions of the present invention, rather than limiting it. The protection scope of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that any person skilled in the art within the technical scope disclosed by the present invention can still modify the technical solutions described in the foregoing embodiments or can easily conceive of changes, or make equivalent replacements for some of the technical features; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for detecting abnormal noise in an air conditioner production line, characterized in that, The method includes: Obtaining acoustic signals in real time through a microphone spherical array; Processing the acoustic signals based on the HOA-SHT domain analysis framework to generate a panoramic sound image; Obtaining an air conditioner image through a panoramic camera; Locking the air conditioner area according to the air conditioner image; Performing multi-modal localization on the air conditioner area through the panoramic sound image to obtain a target area; Extracting acoustic signals from the target area through multi-scale Mel spectrograms, and inputting the extracted acoustic signals into the beamformer of a neural network to obtain noise signals; Detecting and identifying the noise signals based on an anomaly classifier of a convolutional neural network to obtain an air conditioner abnormal sound classification result.

2. The abnormal sound detection method for the air conditioner production line according to claim 1, wherein Processing the acoustic signals based on the HOA-SHT domain analysis framework to generate a panoramic sound image, including: Preprocessing the acoustic signals through Hanning window short-time Fourier transform to obtain preprocessed acoustic signals; Performing spherical harmonic domain transformation on the preprocessed acoustic signals to obtain a 7th-order surround sound field decomposition; Performing spherical harmonic coefficient energy compensation on the 7th-order surround sound field decomposition to obtain spherical harmonic domain features; Passing the spherical harmonic domain features through a beamformer to obtain a sound field; Passing the sound field through a panoramic beam space to obtain respective spatial azimuth features; Superimposing the respective spatial azimuth features and the air conditioner image to obtain the panoramic sound image.

3. The abnormal sound detection method for an air conditioner production line according to claim 1, characterized in that, Locking the air conditioner area according to the air conditioner image, including: Inputting the air conditioner image into a target detection model to obtain the spatial position of the air conditioner; Locking the air conditioner area according to the spatial position of the air conditioner.

4. The abnormal sound detection method for the air conditioner production line according to claim 1, wherein The method further includes: Monitoring the operating state of the production line in real time through the panoramic camera; When the operating state of the production line is a shutdown state, controlling the suspension of audio and video acquisition.

5. The abnormal sound detection method for the air conditioner production line according to claim 1, characterized in that, The method further includes: Prompting the air conditioner abnormal sound classification result through voice broadcast.

6. The abnormal noise detection method for the air conditioner production line according to claim 1, wherein, The air conditioner abnormal sound classification result includes foreign object adhesion, blade burrs, and sheet metal vibration.

7. An abnormal noise detection system for an air conditioner production line, characterized in that, The system includes a computing server and a spherical panoramic sound imager. The spherical panoramic sound imager includes a microphone spherical array and a panoramic camera; the computing server includes a processing module, a locking module, a multi-modal localization module, an extraction module, a detection and identification module; The microphone spherical array is used for obtaining acoustic signals in real time; The processing module is used for processing the acoustic signals based on the HOA-SHT domain analysis framework to generate a panoramic sound image; The panoramic camera is used for obtaining an air conditioner image; The locking module is used for locking the air conditioner area according to the air conditioner image; The multi-modal localization module is used for performing multi-modal localization on the air conditioner area through the panoramic sound image to obtain a target area; The extraction module is used for extracting acoustic signals from the target area through multi-scale Mel spectrograms, and inputting the extracted acoustic signals into the beamformer of a neural network to obtain noise signals; The detection and identification module is used for detecting and identifying the noise signals based on an anomaly classifier of a convolutional neural network to obtain an air conditioner abnormal sound classification result.

8. The abnormal sound detection system for the air conditioner production line according to claim 7, characterized in that, The processing module is specifically used for: Preprocessing the acoustic signals through Hanning window short-time Fourier transform to obtain preprocessed acoustic signals; The preprocessed acoustic signal is subjected to spherical harmonic domain transformation to obtain a 7th-order surround sound field decomposition; The 7th-order surround sound field decomposition is subjected to spherical harmonic coefficient energy compensation to obtain spherical harmonic domain features; The spherical harmonic domain features are passed through a beamformer to obtain a sound field; The sound field is passed through a panoramic beam space to obtain respective spatial azimuth features; The respective spatial azimuth features and the air conditioner image are superimposed to obtain the panoramic sound image.

9. An electronic device, comprising a memory and a processor, wherein a computer program capable of running on the processor is stored on the memory, characterized in that, When the processor executes the computer program, the method described in any one of claims 1 to 6 above is implemented.

10. A computer-readable medium having non-volatile program code executable by a processor, characterized in that, The program code causes the processor to execute the method described in any one of claims 1 to 6.

Images