Rail edge acoustic diagnosis method, device and system and computer readable storage medium

Abstract

The invention provides a rail side acoustic diagnosis method, device and system and a computer readable storage medium, is applied to a rail side acoustic diagnosis system, and relates to the technical field of acoustic diagnosis. The method comprises the following steps: acquiring sound information and space information of a target sound source; performing short-time Fourier transform on the sound information; inputting the transformed sound information and spatial information into a pre-trained sound field reconstruction model, and outputting a fused three-dimensional feature tensor; performing interference suppression processing on the fused three-dimensional feature tensor to obtain target sound information; inputting the target sound information into a pre-trained fault diagnosis model, and outputting a fault category; wherein the fault diagnosis model is used for performing unsupervised anomaly classification and transfer learning fine-grained classification; firstly, a sound field is reconstructed based on sound information and space information, then interference suppression processing is carried out on the sound field, then a diagnosis model which is trained in advance is input, noise is suppressed in a targeted mode, effective signals are accurately separated, and the accuracy of fault diagnosis is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of acoustic diagnosis, and in particular to a trackside acoustic diagnosis method, device, system and computer-readable storage medium. Background Art

[0002] Railway systems are central to the global movement of goods and passengers. Monitoring vehicle health is crucial to preventing derailments, especially wheels and bogies, which are critical components of vehicle performance and require rigorous monitoring. Trackside condition monitoring technology, a non-contact and cost-effective method for detecting bearing defects, not only eliminates the additional burden on bogies but also reduces costs and simplifies operational procedures, offering advantages over onboard vibration or acoustic emission detection systems.

[0003] However, the existing trackside detection technology does not have ideal control over noise, which affects the accurate separation of effective signals and thus affects the accuracy of subsequent fault diagnosis. Summary of the Invention

[0004] The object of the present invention is to provide a trackside acoustic diagnostic method, device, system and computer-readable storage medium. First, the sound field is reconstructed based on acoustic information and spatial information, and then interference suppression processing is performed on it. Then, a pre-trained diagnostic model is input to specifically suppress noise, accurately separate effective signals, and improve the accuracy of fault diagnosis.

[0005] In a first aspect, the present invention provides a trackside acoustic diagnosis method, which is applied to a trackside acoustic diagnosis system. The method comprises:

[0006] Acquire acoustic information and spatial information of the target sound source; wherein the acoustic information is a time domain signal;

[0007] Performing short-time Fourier transform on the acoustic information; wherein the transformed acoustic information is a time-frequency domain signal;

[0008] The transformed acoustic and spatial information are input into a pre-trained sound field reconstruction model, which outputs a fused three-dimensional feature tensor. The sound field reconstruction model is a deep learning architecture based on a self-attention mechanism.

[0009] The fused three-dimensional feature tensor is subjected to interference suppression processing to obtain target sound information;

[0010] The target acoustic information is input into a pre-trained fault diagnosis model to output the fault category. The fault diagnosis model is used for unsupervised anomaly classification and transfer learning fine-grained classification.

[0011] In some preferred embodiments of the present invention, the trackside acoustic diagnostic system includes: two acoustic arrays and a camera; wherein the acoustic arrays are disposed on opposite sides of the track; and the steps of acquiring acoustic information and spatial information of a target sound source include:

[0012] Collecting acoustic information of target sound source based on acoustic array;

[0013] The image of the target sound source is captured by the camera and used as spatial information.

[0014] In some preferred embodiments of the present invention, the acoustic array includes: 16 micro-electromechanical sensing modules; the micro-electromechanical sensing module includes: 16 micro-electromechanical acoustic sensors.

[0015] In some preferred embodiments of the present invention, the step of performing interference suppression processing on the fused three-dimensional feature tensor to obtain target sound information includes:

[0016] Perform spatial filtering on the fused three-dimensional feature tensor to obtain a filtered three-dimensional feature tensor;

[0017] The filtered three-dimensional feature tensor is subjected to signal enhancement processing to obtain target acoustic information.

[0018] In some preferred embodiments of the present invention, the signal enhancement processing includes: time-frequency masking processing and spatial enhancement processing based on delay characteristics in the spatial domain.

[0019] In some preferred embodiments of the present invention, unsupervised anomaly classification includes:

[0020] Obtain unlabeled acoustic information;

[0021] Construct a highly discriminative feature representation space to extract abnormal features and summarize patterns of unlabeled acoustic information;

[0022] The density-based unsupervised detection algorithm is used to preliminarily screen the unlabeled acoustic information after feature extraction and pattern induction.

[0023] In some preferred embodiments of the present invention, transfer learning fine-grained classification includes:

[0024] Obtain a basic network model; the underlying convolution kernel of the basic network model has universal texture and edge feature extraction capabilities;

[0025] Replace the fully connected layer of the basic network model with a classifier that matches the number of defect categories to obtain an updated basic network model;

[0026] Perform domain alignment on the updated basic network model;

[0027] The basic network model after domain alignment based on the labeled acoustic information training is used as the transfer learning fine-grained classification model.

[0028] In a second aspect, the present invention provides a trackside acoustic diagnostic device, which is applied to a trackside acoustic diagnostic system. The device includes:

[0029] A data acquisition module is used to obtain acoustic information and spatial information of the target sound source; wherein the acoustic information is a time domain signal;

[0030] An acoustic information processing module is used to perform short-time Fourier transform on the acoustic information; wherein the transformed acoustic information is a time-frequency domain signal;

[0031] The sound field reconstruction module is used to input the transformed acoustic and spatial information into a pre-trained sound field reconstruction model and output a fused three-dimensional feature tensor. The sound field reconstruction model is a deep learning architecture based on the self-attention mechanism.

[0032] The target sound information acquisition module is used to suppress interference on the fused three-dimensional feature tensor to obtain target sound information;

[0033] The fault output module is used to input the target acoustic information into a pre-trained fault diagnosis model and output the fault category; the fault diagnosis model is used to perform unsupervised anomaly classification and transfer learning fine-grained classification.

[0034] In a third aspect, the present invention provides a trackside acoustic diagnostic system, comprising a processor and a memory, wherein the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the trackside acoustic diagnostic method provided in the first aspect above.

[0035] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement the trackside acoustic diagnostic method provided in the first aspect.

[0036] The present invention brings the following beneficial effects:

[0037] The present invention provides a trackside acoustic diagnosis method, device, system and computer-readable storage medium, which are applied to a trackside acoustic diagnosis system. The method includes: obtaining acoustic information and spatial information of a target sound source; wherein the acoustic information is a time domain signal; performing short-time Fourier transform on the acoustic information; wherein the transformed acoustic information is a time-frequency domain signal; inputting the transformed acoustic information and spatial information into a pre-trained sound field reconstruction model, and outputting a fused three-dimensional feature tensor; wherein the sound field reconstruction model is a deep learning architecture based on a self-attention mechanism; performing interference suppression processing on the fused three-dimensional feature tensor to obtain target acoustic information; inputting the target acoustic information into a pre-trained fault diagnosis model, and outputting a fault category; wherein the fault diagnosis model is used for unsupervised anomaly classification and transfer learning fine-grained classification; first, reconstructing the sound field based on the acoustic information and spatial information, performing interference suppression processing on it, and then inputting it into the pre-trained diagnosis model, thereby specifically suppressing noise, accurately separating effective signals, and improving the accuracy of fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0039] Figure 1 A flow chart of a trackside acoustic diagnostic method provided in an embodiment of the present invention;

[0040] Figure 2 A schematic diagram of the deployment of a sound transmission array provided in an embodiment of this aspect;

[0041] Figure 3 A logic block diagram of a trackside acoustic diagnostic system provided by an embodiment of the present invention;

[0042] Figure 4 A schematic structural diagram of a sensor subsystem of a trackside acoustic diagnostic system provided by an embodiment of the present invention;

[0043] Figure 5 A schematic structural diagram of a trackside acoustic diagnostic device provided by an embodiment of the present invention;

[0044] Figure 6 A schematic structural diagram of a trackside acoustic diagnostic system provided by an embodiment of the present invention.

[0045] Icons: 1-acoustic array; 2-track; 3-micro-electromechanical acoustic sensor; 310-data acquisition module; 320-acoustic information processing module; 330-acoustic field reconstruction module; 340-target acoustic information acquisition module; 350-fault output module; 400-memory; 401-processor; 402-bus; 403-communication interface. DETAILED DESCRIPTION

[0046] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0047] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention as claimed, but rather merely represents selected embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort shall fall within the scope of protection of the present invention.

[0048] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0049] In the description of the present invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer" and the like indicate positions or locations based on the positions shown in the accompanying drawings, or the positions or locations in which the inventive product is typically placed when in use. These terms are intended solely to facilitate the description of the present invention and to simplify the description, and are not intended to indicate or imply that the devices or components referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limitations on the present invention. Furthermore, the terms "first," "second," and "third," etc., are used solely to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0050] Furthermore, terms such as "horizontal," "vertical," and "overhanging" do not necessarily imply that a component must be absolutely horizontal or overhanging, but rather that it can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal than "vertical," and does not mean that the structure must be completely horizontal, but rather that it can be slightly tilted.

[0051] In the description of the present invention, it should also be noted that, unless otherwise expressly specified or limited, the terms "disposed," "installed," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; they may refer to mechanical connections or electrical connections; they may refer to direct connections or indirect connections through an intermediate medium; and they may refer to internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0052] Some existing technologies utilize 12 microphone arrays deployed on both sides of the track to capture bearing noise signals. These are combined with infrared bearing temperature sensors, vehicle identification antenna systems, and speed sensors to detect faults. Faults are identified by analyzing the amplitude and frequency characteristics of the acoustic signals. However, during train travel, the wheelset rolling bearing signals mix with the impact noise from the wheelset tread and track, making it difficult to effectively separate these signals using such a sparse trackside array. This, in turn, affects the accuracy and performance of subsequent fault detection.

[0053] Some existing technologies use trackside microphones to acquire bearing noise signals and utilize neural networks to directly map fault characteristics, vehicle speed, and diagnostic results. This avoids complex Doppler distortion correction modeling and reduces the complexity of trackside acoustic diagnostic systems. However, these still face the problem of complex noise interference during train operation, such as impact noise between the wheelset tread and the track or ambient noise, which can affect the diagnostic accuracy of the target signal. Furthermore, fault data or prior data from the operational phase is difficult to obtain, which poses a significant challenge to subsequent fault diagnosis.

[0054] In summary, addressing the two major technical bottlenecks in online monitoring of rail transit wheelset bearings—difficult signal extraction and scarcity of annotated data—a trackside acoustic diagnosis method has been proposed. This method is applied to a trackside acoustic intelligent diagnosis system based on a large-scale acoustic sensor array. The innovations of this method are primarily reflected in the following three aspects:

[0055] First, in terms of hardware architecture, a dual-plane array collaborative acquisition solution was adopted. Two high-precision microphone arrays with 256 elements were symmetrically deployed on either side of the track to construct a three-dimensional sound field information acquisition system. Compared to traditional single-point sensors, this architecture can simultaneously acquire the time, frequency, and spatial domain characteristics of the sound pressure signal, improving its spatial resolution by two orders of magnitude and laying the data foundation for subsequent signal processing.

[0056] Secondly, at the signal processing level, neural network-based beamforming technology leverages the array's spatial beam-steering properties to accurately extract and reconstruct the acoustic signature of wheelset bearings in motion, significantly improving signal-to-noise ratios compared to traditional methods. Combined with a sound source localization algorithm, the system effectively isolates interference sources such as wheel-rail contact noise and ambient noise, ensuring the integrity of the target signal.

[0057] Finally, in terms of diagnostic algorithms, a phased intelligent detection architecture was designed: in the first phase, a self-supervised contrastive learning method was used to construct a representation space, and unsupervised anomaly detection was used to achieve initial screening of potential faults; in the second phase, a transfer learning strategy was combined with limited labeled data to complete the fine classification of fault types.

[0058] The following embodiments of the present invention are described in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments may be combined with each other.

[0059] Example 1

[0060] The embodiment of the present invention provides a trackside acoustic diagnosis method, which is applied to a trackside acoustic diagnosis system. Figure 1 The flowchart of a trackside acoustic diagnostic method provided in an embodiment of the present invention is shown, and the method includes:

[0061] Step S102: Acquire acoustic information and spatial information of the target sound source; wherein the acoustic information is a time domain signal.

[0062] Specifically, the target sound source is generally a train running on the track. The sound information of the train is obtained through the sound collection system, and the spatial information represents the direction of the train's movement. In some preferred embodiments of the present invention, based on the position of the sound collection system, the sound information can also include the position information of some collection equipment.

[0063] Furthermore, in some preferred embodiments of the present invention, the trackside acoustic diagnostic system includes: two acoustic arrays 1 and a camera; wherein the acoustic arrays 1 are relatively arranged on both sides of the track 2; the step of obtaining the acoustic information and spatial information of the target sound source includes: collecting the acoustic information of the target sound source based on the acoustic arrays 1; collecting an image of the target sound source based on the camera, and using the image as spatial information.

[0064] For details, see Figure 2 The diagram shows a deployment diagram of a sound transmission array provided by an embodiment of the present invention, wherein the sound transmission arrays 1 are arranged opposite to each other on both sides of the track 2; Figure 2 The illustrated sound transmission arrays 1 are each provided with a plurality of micro-electromechanical acoustic sensors 3 , through which acoustic signals are collected.

[0065] For further information, see Figure 3The following is a logical block diagram of a trackside acoustic diagnostic system, provided by an embodiment of the present invention. This system uses a visual recognition unit to obtain real-time train positioning information and a radio frequency identification unit (RFID) to automatically read the locomotive's electronic tag number. High-performance data processing units are deployed at edge computing nodes, capable of synchronously storing multimodal sensor data and applying a deep learning-based fault diagnosis algorithm to perform multi-dimensional analysis of bearing soundprint characteristics.

[0066] Furthermore, in some preferred embodiments of the present invention, the acoustic array 1 includes: 16 micro-electromechanical sensing modules; the micro-electromechanical sensing modules include: 16 micro-electromechanical acoustic sensors 3.

[0067] Specifically, each array consists of 256 high-sensitivity micro-electromechanical acoustic sensors 3 in a 16×16 array, which can realize the full-circumferential acoustic wave signal collection of the train wheel set bearing.

[0068] For example, see Figure 4 The embodiment of the present invention shown is a structural diagram of a sensing subsystem of a trackside acoustic diagnostic system. The subsystem includes: 16 micro-electromechanical sensing modules, each of which includes 16 micro-electromechanical acoustic sensors 3, and is connected to the processor 401 signal through a cascaded time-division multiplexing (TDM) data interface. The processor 401 transmits the collected acoustic signal to the user end through the network. At the same time, the system is equipped with a network camera for image capture, wherein the camera 1 is fused with the microphone array, and the acquired image provides target spatial information for the reconstructed sound field, thereby obtaining the target sound source that can be locked through the sound field information; cameras 2 and 3 are responsible for capturing the train's coming and going directions, respectively, to assist in monitoring the vehicle's operating status. Finally, the audio and video signals are converged on a network cable through a router and transmitted to the user end.

[0069] Step S104 , performing short-time Fourier transform on the acoustic information; wherein the transformed acoustic information is a time-frequency domain signal.

[0070] Specifically, the computation serves the data storage unit, utilizing a distributed data storage architecture and a multi-GPU parallel computing platform to construct an acoustic signal processing pipeline. First, a short-time Fourier transform (SFT) is performed on the time-domain signals collected by the 256-channel acoustic sensor array. In some preferred embodiments of the present invention, the SFT includes a 256ms window function, a Hamming window, and a 50% frame shift.

[0071] Step S106 , inputting the transformed acoustic information and spatial information into a pre-trained sound field reconstruction model, and outputting a fused three-dimensional feature tensor; wherein the sound field reconstruction model is a deep learning architecture based on a self-attention mechanism.

[0072] Specifically, through the sound field reconstruction model based on the deep learning (Transformer) architecture, sound field representation learning is carried out in the three-dimensional feature space of time domain, space domain and frequency domain, the correlation between time, space and frequency domain features is obtained, and the acoustic signal characteristics of maneuvering targets are extracted. This can better integrate geometric space features to achieve directional tracking of sound source beams in the wheelset area.

[0073] Step S108: performing interference suppression processing on the fused three-dimensional feature tensor to obtain target sound information.

[0074] Specifically, the fused features are subjected to noise reduction and enhancement processing.

[0075] Furthermore, in some preferred embodiments of the present invention, the step of performing interference suppression processing on the fused three-dimensional feature tensor to obtain target sound information includes: performing spatial domain filtering on the fused three-dimensional feature tensor to obtain a filtered three-dimensional feature tensor; and performing signal enhancement processing on the filtered three-dimensional feature tensor to obtain target sound information.

[0076] Specifically, a neural network beamformer generates spatial filter weights in real time based on the target sound source's orientation, achieving high-precision and high-stability adaptive spatial filtering. This effectively suppresses interfering noise through a deep learning-optimized beamforming algorithm. The neural network first generates spatial filter coefficients in real time based on the data's characteristics. This effectively measures the characteristics of each channel's acoustic signal, background noise, and interfering noise in real time.

[0077] Furthermore, in some preferred embodiments of the present invention, the signal enhancement processing includes: time-frequency masking processing and spatial enhancement processing based on delay characteristics in the spatial domain.

[0078] Specifically, time-frequency masking is performed to remove noise, improving the signal-to-noise ratio of each channel. Spatial enhancement is then performed using spatial delay features to suppress interfering noise. Ultimately, the signal-to-noise ratio of the output wheelset rolling bearing characteristic acoustic signal is improved, providing a high-quality data foundation for subsequent fault diagnosis and condition monitoring.

[0079] In some preferred embodiments of the present invention, since sound sources in different directions have different spatial characteristics, it is necessary to perform delay compensation on microphones deployed in the space to achieve the purpose of increasing the space.

[0080] In step S110 , the target acoustic information is input into a pre-trained fault diagnosis model, and the fault category is output; wherein the fault diagnosis model is used for unsupervised anomaly classification and transfer learning fine-grained classification.

[0081] Specifically, the fault diagnosis model adopts a two-stage intelligent detection architecture, realizing a progressive diagnosis process from initial screening of abnormalities to accurate classification.

[0082] Furthermore, in some preferred embodiments of the present invention, unsupervised anomaly classification includes: obtaining unlabeled acoustic information; constructing a high-discrimination feature representation space to extract anomaly features and summarize patterns of the unlabeled acoustic information; and using a density-based unsupervised detection algorithm to preliminarily screen the unlabeled acoustic information that has completed feature extraction and pattern summarization.

[0083] Specifically, the first phase uses a self-supervised contrastive learning framework to construct a highly discriminative feature representation space, extract abnormal features and summarize patterns from massive unlabeled samples, and use an unsupervised detection algorithm based on density-based spatial clustering of applications with noise (DBSCAN) to perform preliminary screening of potential fault samples.

[0084] Furthermore, in some preferred embodiments of the present invention, transfer learning fine-grained classification includes: obtaining a basic network model; wherein the underlying convolution kernel of the basic network model has universal texture and edge feature extraction capabilities; replacing the fully connected layer of the basic network model with a classifier that matches the number of defect categories to obtain an updated basic network model; performing domain alignment on the updated basic network model; training the domain-aligned basic network model based on the labeled acoustic information, and using the basic network model that meets preset conditions as a transfer learning fine-grained classification model.

[0085] Specifically, the second stage introduces a transfer learning strategy. Through knowledge transfer and parameter fine-tuning mechanisms, the pre-trained model (such as ResNet, EfficientNet) is first used as the basic network, and its underlying convolution kernel already has the general texture and edge feature extraction capabilities. Secondly, after feature reuse, the first N layers of the pre-trained model (such as the first three residual blocks of ResNet50) are frozen to retain its general visual feature extraction capabilities; the top network (fully connected layer) is replaced with a classifier that matches the number of defect categories. Then, domain alignment is performed to eliminate the scale difference between the source domain and the target domain through input layer interpolation or adaptive pooling layer, thereby achieving cross-dimensional migration. Finally, the feature extraction capability of the pre-trained model is transferred to the downstream classification task, and a small number of accurately labeled samples are combined to achieve fine-grained recognition of multiple types of defects. This staged processing architecture effectively solves the problem of scarce labeled data in industrial inspection, and significantly improves the practicality and classification accuracy of the model in actual industrial scenarios.

[0086] The trackside acoustic diagnostic method provided by the present invention innovatively integrates soundprint recognition and time-frequency analysis technology. It can effectively identify early cracks, spalling, corrosion and other hidden damages of key components such as the inner ring, outer ring, and rolling elements of rolling bearings, realize intelligent assessment of fault levels and early warning decision support, and significantly improve the status monitoring accuracy and fault prediction capabilities of rail 2 transportation equipment.

[0087] The present invention provides a trackside acoustic diagnosis method, which is applied to a trackside acoustic diagnosis system. The method includes: obtaining acoustic information and spatial information of a target sound source; wherein the acoustic information is a time domain signal; performing short-time Fourier transform on the acoustic information; wherein the transformed acoustic information is a time-frequency domain signal; inputting the transformed acoustic information and spatial information into a pre-trained sound field reconstruction model, and outputting a fused three-dimensional feature tensor; wherein the sound field reconstruction model is a deep learning architecture based on a self-attention mechanism; performing interference suppression processing on the fused three-dimensional feature tensor to obtain target acoustic information; inputting the target acoustic information into a pre-trained fault diagnosis model, and outputting a fault category; wherein the fault diagnosis model is used to perform unsupervised anomaly classification and transfer learning fine-grained classification; first, reconstructing the sound field based on the acoustic information and spatial information, performing interference suppression processing on it, and then inputting it into the pre-trained diagnosis model, thereby specifically suppressing noise, accurately separating effective signals, and improving the accuracy of fault diagnosis.

[0088] Example 2

[0089] Based on the above embodiments, the present invention provides a trackside acoustic diagnostic device, which is applied to a trackside acoustic diagnostic system. Figure 5 The schematic diagram of the structure of a trackside acoustic diagnostic device provided by an embodiment of the present invention is shown, and the device includes:

[0090] The data acquisition module 310 is used to acquire the acoustic information and spatial information of the target sound source; wherein the acoustic information is a time domain signal.

[0091] The acoustic information processing module 320 is used to perform short-time Fourier transform on the acoustic information; wherein the transformed acoustic information is a time-frequency domain signal.

[0092] The sound field reconstruction module 330 is used to input the transformed acoustic information and spatial information into a pre-trained sound field reconstruction model and output a fused three-dimensional feature tensor; wherein the sound field reconstruction model is a deep learning architecture based on the self-attention mechanism.

[0093] The target sound information acquisition module 340 is used to perform interference suppression processing on the fused three-dimensional feature tensor to obtain target sound information.

[0094] The fault output module 350 is used to input the target sound information into a pre-trained fault diagnosis model and output the fault category; wherein the fault diagnosis model is used to perform unsupervised anomaly classification and transfer learning fine-grained classification.

[0095] Furthermore, in some preferred embodiments of the present invention, the trackside acoustic diagnostic system includes: two acoustic arrays 1 and a camera; wherein the acoustic arrays 1 are relatively arranged on both sides of the track 2; a data acquisition module 310, which is used to collect acoustic information of the target sound source based on the acoustic arrays 1; and collect an image of the target sound source based on the camera, and use the image as spatial information.

[0096] Furthermore, in some preferred embodiments of the present invention, the acoustic array 1 includes: 16 micro-electromechanical sensing modules; the micro-electromechanical sensing modules include: 16 micro-electromechanical acoustic sensors 3.

[0097] Furthermore, in some preferred embodiments of the present invention, the target sound information acquisition module 340 is used to perform spatial filtering on the fused three-dimensional feature tensor to obtain a filtered three-dimensional feature tensor; and perform signal enhancement processing on the filtered three-dimensional feature tensor to obtain target sound information.

[0098] Furthermore, in some preferred embodiments of the present invention, the signal enhancement processing includes: time-frequency masking processing and spatial enhancement processing based on delay characteristics in the spatial domain.

[0099] Furthermore, in some preferred embodiments of the present invention, unsupervised anomaly classification includes: obtaining unlabeled acoustic information; constructing a high-discrimination feature representation space to extract anomaly features and summarize patterns of the unlabeled acoustic information; and using a density-based unsupervised detection algorithm to preliminarily screen the unlabeled acoustic information that has completed feature extraction and pattern summarization.

[0100] Furthermore, in some preferred embodiments of the present invention, a basic network model is obtained; wherein the underlying convolution kernel of the basic network model has the ability to extract general texture and edge features; the fully connected layer of the basic network model is replaced with a classifier that matches the number of defect categories to obtain an updated basic network model; the updated basic network model is domain aligned; the domain-aligned basic network model is trained based on the labeled acoustic information, and the basic network model that meets the preset conditions is used as a transfer learning fine-grained classification model.

[0101] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working process of the trackside acoustic diagnostic device described above can refer to the corresponding process in the aforementioned embodiment of the trackside acoustic diagnostic method, and will not be repeated here.

[0102] Example 3

[0103] The embodiment of the present invention further provides a trackside acoustic diagnosis system for running a trackside acoustic diagnosis method; see Figure 6 The figure shows a schematic structural diagram of a trackside acoustic diagnostic system provided by an embodiment of the present invention. The system includes a memory 400 and a processor 401, wherein the memory 400 is used to store one or more computer instructions, and the one or more computer instructions are executed by the processor 401 to implement the above-mentioned trackside acoustic diagnostic method.

[0104] Further, Figure 6 The trackside acoustic diagnostic system shown further includes a bus 402 and a communication interface 403 , and the processor 401 , the communication interface 403 and the memory 400 are connected via the bus 402 .

[0105] Among them, the memory 400 may include a high-speed random access memory 400 (RAM), and may also include a non-volatile memory 400 (non-volatile memory), such as at least one disk memory 400. The communication connection between the trackside acoustic diagnostic system network element and at least one other network element is realized through at least one communication interface 403 (which can be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. can be used. The bus 402 can be an ISA bus 402, a PCI bus 402 or an EISA bus 402, etc. The bus 402 can be divided into an address bus 402, a data bus 402, a control bus 402, etc. For ease of representation, Figure 6 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus 402 or only one type of bus 402.

[0106] Processor 401 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method may be completed by hardware integrated logic circuits in processor 401 or by software instructions. The above processor 401 may be a general-purpose processor 401, including a central processing unit (CPU) 401, a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It may implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor 401 may be a microprocessor 401 or any conventional processor 401. The steps of the methods disclosed in conjunction with the embodiments of the present invention may be directly implemented and executed by a hardware decoding processor 401, or by a combination of hardware and software modules in the decoding processor 401. The software module can be located in a storage medium well-known in the art, such as random access memory 400, flash memory, read-only memory 400, programmable read-only memory 400, electrically erasable programmable memory 400, registers, etc. The storage medium is located in memory 400, and processor 401 reads information in memory 400 and, in conjunction with its hardware, completes the steps of the method of the aforementioned embodiment.

[0107] An embodiment of the present invention further provides a computer-readable storage medium storing computer-executable instructions. When the computer-executable instructions are called and executed by the processor 401, the computer-executable instructions prompt the processor 401 to implement the above-mentioned trackside acoustic diagnostic method. For specific implementation, please refer to the method embodiment and will not be repeated here.

[0108] The computer program products of the trackside acoustic diagnostic method, apparatus, and system provided in the embodiments of the present invention include a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods in the previous method embodiments. For specific implementation, please refer to the method embodiments and will not be repeated here.

[0109] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working process of the system and / or device described above can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0110] In addition, in the description of the embodiments of the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; they may refer to mechanical connections or electrical connections; they may refer to direct connections or indirect connections through an intermediate medium; and they may refer to internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0111] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory 400 (ROM), a random access memory 400 (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0112] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for trackside acoustic diagnosis, characterized in that: Applied to a trackside acoustic diagnostic system, the method comprises: Acquiring acoustic information and spatial information of a target sound source; wherein the acoustic information is a time domain signal; Performing a short-time Fourier transform on the acoustic information; wherein the acoustic information after the transform is a time-frequency domain signal; Inputting the transformed acoustic information and the spatial information into a pre-trained sound field reconstruction model, and outputting a fused three-dimensional feature tensor; wherein the sound field reconstruction model is a deep learning architecture based on a self-attention mechanism; The fused three-dimensional feature tensor is subjected to interference suppression processing to obtain target sound information; The target acoustic information is input into a pre-trained fault diagnosis model to output a fault category; wherein the fault diagnosis model is used to perform unsupervised anomaly classification and transfer learning fine-grained classification.

2. The railside acoustic diagnostic method according to claim 1, characterized in that: The trackside acoustic diagnostic system includes: two acoustic arrays and cameras; wherein the acoustic arrays are arranged on opposite sides of the track; the steps of obtaining acoustic information and spatial information of the target sound source include: collecting acoustic information of a target sound source based on the acoustic array; An image of the target sound source is captured by the camera, and the image is used as the spatial information.

3. The railside acoustic diagnostic method according to claim 2, characterized in that: The acoustic array includes 16 micro-electromechanical sensing modules; the micro-electromechanical sensing modules include 16 micro-electromechanical acoustic sensors.

4. The railside acoustic diagnostic method according to claim 1, characterized in that: The step of performing interference suppression processing on the fused three-dimensional feature tensor to obtain target sound information includes: Performing spatial domain filtering on the fused three-dimensional feature tensor to obtain the filtered three-dimensional feature tensor; Signal enhancement processing is performed on the filtered three-dimensional feature tensor to obtain the target sound information.

5. The railside acoustic diagnostic method according to claim 4, characterized in that: The signal enhancement processing includes: time-frequency masking processing and space enhancement processing based on delay characteristics in the spatial domain.

6. The railside acoustic diagnostic method according to claim 1, characterized in that: The unsupervised anomaly classification includes: Obtain unlabeled acoustic information; Constructing a high-discrimination feature representation space to extract abnormal features and summarize patterns of the unlabeled acoustic information; The unlabeled acoustic information after feature extraction and pattern induction is preliminarily screened using a density-based unsupervised detection algorithm.

7. The railside acoustic diagnostic method according to claim 6, characterized in that: The transfer learning fine-grained classification includes: Obtaining a basic network model; wherein the underlying convolution kernel of the basic network model has universal texture and edge feature extraction capabilities; Replacing the fully connected layer of the basic network model with a classifier that matches the number of defect categories to obtain an updated basic network model; Performing domain alignment on the updated basic network model; The basic network model aligned with the labeled acoustic information training domain is used as a transfer learning fine-grained classification model that meets preset conditions.

8. A trackside acoustic diagnostic device, characterized in that: Applied to a trackside acoustic diagnostic system, the device comprises: A data acquisition module, configured to acquire acoustic information and spatial information of a target sound source; wherein the acoustic information is a time domain signal; An acoustic information processing module, configured to perform a short-time Fourier transform on the acoustic information; wherein the acoustic information after the transform is a time-frequency domain signal; A sound field reconstruction module, configured to input the transformed acoustic information and spatial information into a pre-trained sound field reconstruction model and output a fused three-dimensional feature tensor; wherein the sound field reconstruction model is a deep learning architecture based on a self-attention mechanism; The target sound information acquisition module is used to suppress interference on the fused three-dimensional feature tensor to obtain target sound information; The fault output module is used to input the target sound information into a pre-trained fault diagnosis model and output the fault category; wherein the fault diagnosis model is used to perform unsupervised anomaly classification and transfer learning fine-grained classification.

9. A trackside acoustic diagnostic system, characterized in that: The system comprises a processor and a memory, wherein the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the trackside acoustic diagnosis method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement the trackside acoustic diagnostic method according to any one of claims 1 to 7.

Images