Fault monitoring method and device, equipment, storage medium and program product

Abstract

The invention discloses a fault monitoring method and device, equipment, a storage medium and a program product, and relates to the technical field of equipment monitoring, and the method comprises the steps: collecting audio signals of each monitoring point of a belt conveyor through an optical fiber arranged in the belt motion direction of the belt conveyor, carrying out the coding of the audio signals of each monitoring point based on a channel attention mechanism, and obtaining the audio signals of each monitoring point; obtaining coding characteristics of each monitoring point; according to each monitoring point, fault identification is carried out on the monitoring point based on the coding feature of the monitoring point, the identification result of the monitoring point is obtained, the identification result represents whether the belt conveyor has a fault near the monitoring point or not and the fault type when the fault exists, and the purpose of automatically monitoring the fault of the belt conveyor is achieved. While the safety of workers is ensured, the monitoring cost is reduced, and the monitoring efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of equipment monitoring, and in particular to a fault monitoring method, apparatus, device, storage medium and program product. Background Art

[0002] At present, in order to ensure the smooth progress of industrial production, it is necessary to monitor the faults of industrial production equipment to avoid production accidents caused by equipment failures. The current method of monitoring equipment faults is mainly manual regular inspections. Inspectors use visual inspections, listening and experience to determine whether the equipment is abnormal. The accuracy of this manual inspection method depends largely on the professional knowledge of the staff and the frequency of inspections, and the cost is relatively high. In addition, some equipment is sometimes installed in more dangerous locations, and the safety of the staff cannot be fully guaranteed during inspections. Manual inspections have certain limitations.

[0003] Therefore, it is necessary to provide an automatic fault monitoring solution for industrial equipment. Summary of the invention

[0004] In view of the above problems, the present application provides a fault monitoring method, device, equipment, storage medium and program product to realize automatic fault monitoring of industrial equipment. The specific scheme is as follows: A first aspect of the present application provides a fault monitoring method, the method comprising: The audio signals of each monitoring point of the belt conveyor are collected through optical fibers arranged along the belt movement direction of the belt conveyor; Encode the audio signal of each monitoring point based on the channel attention mechanism to obtain the encoding features of each monitoring point; Corresponding to each monitoring point, fault identification is performed on the monitoring point based on the coding characteristics of the monitoring point to obtain an identification result of the monitoring point, which characterizes whether the belt conveyor has a fault near the monitoring point and the type of fault when a fault exists.

[0005] In a possible implementation, the optical fiber is arranged on a bracket on the belt conveyor that is parallel to the moving direction of the belt, and the optical fiber is located below or to the side of a roller that supports the belt.

[0006] In a possible implementation, collecting audio signals of various monitoring points of the belt conveyor through optical fibers arranged along the belt movement direction of the belt conveyor includes: The audio signals of each monitoring point of the belt conveyor are collected through two optical fibers arranged in parallel on both sides of the belt; The monitoring points arranged on both sides of the belt are located at different positions in the moving direction of the belt.

[0007] In a possible implementation, encoding the audio signal of each monitoring point based on the channel attention mechanism to obtain the encoding feature of each monitoring point includes: Corresponding to each monitoring point, the audio signal of the monitoring point is adaptively filtered to obtain a filtered audio signal; the background noise of the filtered audio signal is less than the background noise of the audio signal of the monitoring point; The filtered audio signal is encoded based on the channel attention mechanism to obtain the encoding features of the monitoring point.

[0008] In a possible implementation, the audio signal of each monitoring point is encoded based on the channel attention mechanism, and corresponding to each monitoring point, the process of fault identification of the monitoring point based on the encoding feature of the monitoring point includes: The fault recognition model encodes the audio signal of each monitoring point based on the channel attention mechanism, and identifies the fault of each monitoring point based on the encoding features of the monitoring point to obtain the identification result of each monitoring point. The fault recognition model is obtained by pre-training with the first audio data set, and fine-tuning the pre-trained fault recognition model with the second audio data set; The audio data in the first audio data set is audio data in a non-belt conveyor scenario, and the labels of the audio data in the first audio data set represent whether the audio data has noise or faults, and the noise category when noise exists or the fault category when faults exist; The audio data in the second audio data set is audio data in a belt conveyor scenario, and the label of the audio data in the second audio data set represents whether there is a fault in the audio data and the fault type when there is a fault.

[0009] In a possible implementation, the process of pre-training the fault recognition model using the first audio data set and fine-tuning the pre-trained fault recognition model using the second audio data set includes: Inputting the acoustic features of the audio data in the first audio data set into the fault recognition model to obtain a first recognition result of each audio data in the first audio data set; with the goal of making the first recognition result of each audio data in the first audio data set close to the label of the audio data in the first audio data set, updating the parameters of the fault recognition model to obtain a pre-trained fault recognition model; Inputting the acoustic features of the audio data in the second audio data set into the pre-trained fault recognition model to obtain a second recognition result of each audio data in the second audio data set; taking the second recognition result of each audio data in the second audio data set as a goal to be close to the label of the audio data in the second audio data set, updating the parameters of the target layer of the encoding module and the parameters of the non-encoding module of the pre-trained fault recognition model to obtain a trained fault recognition model; The output of the non-target layer of the encoding module is connected to the input of the target layer; the input of the non-target layer is the input of the fault identification model, the non-target layer does not include a channel attention layer, and the target layer includes a channel attention layer.

[0010] A second aspect of the present application provides a fault monitoring device, comprising: A collection module, used for collecting audio signals of various monitoring points of the belt conveyor through optical fibers arranged along the belt movement direction of the belt conveyor; The encoding module is used to encode the audio signal of each monitoring point based on the channel attention mechanism to obtain the encoding features of each monitoring point; The identification module is used to identify the fault of each monitoring point based on the coding characteristics of the monitoring point to obtain the identification result of the monitoring point. The identification result of the monitoring point indicates whether there is a fault in the belt conveyor near the monitoring point and the type of fault when there is a fault.

[0011] A third aspect of the present application provides a computer program product, comprising computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements the fault monitoring method of the first aspect or any implementation of the first aspect.

[0012] A fourth aspect of the present application provides an electronic device, comprising at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the electronic device can implement the fault monitoring method of the above-mentioned first aspect or any implementation manner of the first aspect.

[0013] A fifth aspect of the present application provides a computer storage medium, which carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement the fault monitoring method of the above-mentioned first aspect or any implementation of the first aspect.

[0014] By means of the above-mentioned technical scheme, the fault monitoring method, device, equipment, storage medium and program product provided by the present application collect the audio signals of each monitoring point of the belt conveyor through optical fibers arranged along the belt movement direction of the belt conveyor, encode the audio signals of each monitoring point based on the channel attention mechanism, and obtain the encoding features of each monitoring point; corresponding to each monitoring point, the fault of the monitoring point is identified based on the encoding features of the monitoring point to obtain the identification result of the monitoring point, which characterizes whether there is a fault in the belt conveyor near the monitoring point, and the type of fault when there is a fault, thereby achieving the purpose of automatically monitoring the fault of the belt conveyor, while ensuring the safety of the staff, reducing the monitoring cost and improving the monitoring efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0016] Figure 1 A flowchart of an implementation of the fault monitoring method provided in this application; Figure 2a An example diagram of a distribution of monitoring points provided on one side of a belt for this application; Figure 2b An example diagram of the distribution of monitoring points set on both sides of the belt provided in this application; Figure 3 A flowchart for implementing the encoding of the audio signal of each monitoring point based on the channel attention mechanism provided in the present application to obtain the encoding features of each monitoring point; Figure 4 A structural schematic diagram of a fault monitoring device provided in this application; Figure 5 A schematic diagram of the structure of an electronic device provided in this application. DETAILED DESCRIPTION

[0017] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0018] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0019] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0020] Take the belt conveyor (abbreviated as belt conveyor) for transporting coal as an example. Its working environment is relatively harsh, and key equipment is in an unstable state such as collision and wear for a long time, which can easily cause failure and damage of related equipment, thus affecting the efficiency of the entire link transportation, causing many inconveniences and direct losses in production. As an important component of the belt conveyor for producing and transporting coal, rollers play the role of supporting the conveyor belt and transporting goods. If the failure of rollers is not monitored, it may sometimes even cause serious production accidents.

[0021] At present, the monitoring of such faults mainly relies on regular manual inspections. The inspectors visually judge whether the rollers are stuck or broken, and combine the sound of the belt conveyor and previous work experience to judge whether there are abnormalities such as slippage and wear. However, the accuracy of this method depends largely on the professional knowledge of the staff and the frequency of inspections. In addition, some equipment sometimes needs to be carried in more dangerous locations, and the safety of the staff cannot be fully guaranteed during inspections, which has certain limitations. Therefore, it is particularly important to develop an efficient and accurate fault detection system.

[0022] In order to realize automatic fault monitoring of a belt conveyor, the present application scheme is proposed.

[0023] like Figure 1 As shown, a flowchart of an implementation of the fault monitoring method provided in an embodiment of the present application may include: Step S101: collecting audio signals from various monitoring points of the belt conveyor through optical fibers arranged along the belt movement direction of the belt conveyor.

[0024] In this application, the optical fiber in the distributed fiber acoustic sensing (DAS) system is extended along the moving direction of the belt of the conveyor belt, wherein at each monitoring point, the optical fiber is coiled at least once and then extended backward. The principle of sound collection by the distributed fiber acoustic sensing system can be referred to the existing solutions and will not be described in detail here.

[0025] Optical fibers can be arranged on one side of the belt or on both sides of the belt. In the case where optical fibers are arranged on both sides of the belt, two distributed optical fiber acoustic wave sensing systems are required, and optical fibers of different distributed optical fiber acoustic wave sensing systems are arranged on both sides of the belt.

[0026] like Figure 2a FIG. 1 is a distribution diagram of monitoring points provided on one side of a belt according to an embodiment of the present application. In the diagram, the arrow indicates the direction of movement of the belt, the dotted line indicates the optical fiber deployed on the bracket below the roller (not shown in the diagram), and each black dot indicates a monitoring point. The roller is located below the belt and in contact with the belt.

[0027] The multiple monitoring points may be evenly arranged or unevenly arranged. For example, more monitoring points may be arranged in a range prone to failure, and fewer monitoring points may be arranged in a range not prone to failure.

[0028] Compared with setting up microphones at monitoring points, the use of optical fiber for sound reception has the advantages of quick deployment (just pull the wires, while microphones need to be deployed individually), low cost, good sound reception, and high accuracy.

[0029] Step S102: Encode the audio signal of each monitoring point based on the channel attention mechanism to obtain the encoding features of each monitoring point.

[0030] For each monitoring point, the acoustic features of the audio signal at the monitoring point can be extracted, and the acoustic features can be encoded based on the channel attention mechanism to obtain the encoded features of the monitoring point.

[0031] The acoustic features may include but are not limited to any of the following: Mel-frequency cepstral coefficients (MFCC), Fbank features, etc.

[0032] Step S103: corresponding to each monitoring point, fault identification is performed on the monitoring point based on the coding feature of the monitoring point to obtain an identification result of the monitoring point, which characterizes whether the belt conveyor has a fault near the monitoring point and the fault type when a fault exists.

[0033] The audio signal of each monitoring point can be encoded based on the channel attention mechanism through a deep neural network based on the channel attention mechanism, and the fault of each monitoring point can be identified based on the encoding features of the monitoring point to obtain the identification result of the monitoring point.

[0034] The fault monitoring method provided in the embodiment of the present application collects the audio signals of each monitoring point of the belt conveyor through optical fibers arranged along the belt movement direction of the belt conveyor, encodes the audio signals of each monitoring point based on the channel attention mechanism, and obtains the encoding features of each monitoring point; corresponding to each monitoring point, fault identification is performed on the monitoring point based on the encoding features of the monitoring point to obtain an identification result of the monitoring point, which characterizes whether there is a fault in the belt conveyor near the monitoring point and the type of fault when there is a fault, thereby achieving the purpose of automatically monitoring the fault of the belt conveyor, reducing the monitoring cost and improving the monitoring efficiency while ensuring the safety of the staff.

[0035] In an optional embodiment, the belt of the belt conveyor is supported by a plurality of rollers distributed along the moving direction of the belt, and the plurality of rollers are fixed on a bracket parallel to the moving direction of the belt. Based on this, each optical fiber is arranged on a bracket parallel to the moving direction of the belt on the belt conveyor, and the optical fiber is located below or to the side of the rollers supporting the belt.

[0036] That is to say, the belt conveyor has a bracket fixed along the moving direction of the belt under or on the side of the roller for deploying the optical fiber.

[0037] In an optional embodiment, an implementation method of collecting audio signals of various monitoring points of the belt conveyor through optical fibers arranged along the belt movement direction of the belt conveyor may be: The audio signals of each monitoring point of the belt conveyor are collected through two optical fibers arranged in parallel on both sides of the belt; The monitoring points arranged on both sides of the belt are located at different positions along the moving direction of the belt.

[0038] That is to say, two optical fibers can be arranged on both sides of the belt to collect audio signals at various monitoring points, and at each monitoring point, there is only the monitoring point in the direction perpendicular to the optical fiber.

[0039] like Figure 2b FIG. 1 is a distribution diagram of monitoring points arranged on both sides of a belt provided in an embodiment of the present application. In the diagram, the arrow indicates the direction of movement of the belt, each dotted line represents an optical fiber deployed on a bracket below a roller (not shown in the diagram), and each black dot represents a monitoring point. The roller is located below the belt and in contact with the belt.

[0040] In an optional embodiment, the above-mentioned encoding of the audio signal of each monitoring point based on the channel attention mechanism to obtain the encoding feature of each monitoring point is as follows: Figure 3 As shown, it may include: Step S301: corresponding to each monitoring point, adaptively filter the audio signal of the monitoring point (referred to as the original audio signal for ease of description and distinction) to obtain a filtered audio signal; the background noise of the filtered audio signal is smaller than the background noise of the original audio signal.

[0041] Optionally, the audio signal of each monitoring point may be subjected to a least mean square (LMS) adaptive filtering, or the audio signal of each monitoring point may be subjected to a recursive least squares (RLS) adaptive filtering.

[0042] Through adaptive filtering, the background noise around the belt conveyor can be filtered out and the real signal emitted by the belt conveyor can be well restored.

[0043] Step S302: Encode the filtered audio signal based on the channel attention mechanism to obtain the encoding features of the monitoring point.

[0044] Corresponding to each monitoring point, each audio frame of the filtered audio signal of the monitoring point can be encoded based on the channel attention mechanism.

[0045] Optionally, the acoustic features of each audio frame can be extracted, and the acoustic features of each audio frame can be encoded based on the channel attention mechanism to obtain the encoded features of each audio frame. Specifically, corresponding to the acoustic features of each audio frame, the acoustic features of the audio frame can be encoded at multiple levels, and the multiple levels can include multiple convolutional layers and multiple channel attention layers, and the multiple channel attention layers are cross-distributed with the multiple convolutional layers.

[0046] Optionally, the multiple channel attention layers may be the same or different.

[0047] As an example, each channel attention layer can be any of the following channel attention layers: SE-Nets (Squeeze-and-Excitation Networks) attention layer, ECA (Efficient Channel Attention) attention layer, etc.

[0048] In an optional embodiment, the audio signal of each monitoring point is encoded based on the channel attention mechanism, and corresponding to each monitoring point, a method for implementing fault identification of the monitoring point based on the encoding feature of the monitoring point may be: The audio signal of each monitoring point is encoded based on the channel attention mechanism through the fault recognition model, and the fault of each monitoring point is identified based on the encoding features of the monitoring point to obtain the recognition result of each monitoring point.

[0049] Optionally, the fault identification model may adopt any of the following network structures: a deep convolutional neural network, a time delay neural network (TDNN), an efficient network, etc.

[0050] The fault identification model includes an encoding module.

[0051] As an example, in a deep convolutional neural network, the encoding module may include multiple convolutional layers and multiple channel attention layers; multiple channel attention layers are cross-distributed with multiple convolutional layers. As an example, the multiple layers of the encoding module are: the first 5 layers are convolutional layers, the 6th layer is a channel attention layer, the 7th-8th layers are convolutional layers, the 9th layer is a channel attention layer, the 10th-11th layers are convolutional layers, and the 12th layer is a channel attention layer. Among them, the structures of each convolutional layer can be the same or different.

[0052] As an example, in a time-delay neural network, the encoding module may include multiple feedforward network layers and multiple channel attention layers, and the multiple feedforward network layers and the multiple channel attention layers are cross-distributed.

[0053] As an example, in an Efficient network, the encoding module may include multiple MBConv modules and multiple channel attention modules (i.e., multiple channel attention layers), each MBconv module consists of at least one ordinary convolutional layer and at least one depth-separable convolutional layer, and each channel attention layer may be deployed in one MBConv module, or, deployed between two MBConv modules.

[0054] The fault recognition model is obtained by pre-training with the first audio data set, and fine-tuning the pre-trained fault recognition model with the second audio data set.

[0055] The audio data in the first audio data set is audio data in a non-belt conveyor scenario, and the labels of the audio data in the first audio data set are whether there is noise or fault, and the noise category when there is noise or the fault category when there is a fault.

[0056] Optionally, the first audio data set includes audio data of general scenes, that is, audio data including environmental noise in general scenes, wherein the environmental noise may include but is not limited to: bird calls, human voices, rain, wind, thunder, etc. For the audio data of general scenes, the corresponding labels represent whether there is noise in the audio data, and the noise category when there is noise. The labels of the audio data may be manually annotated or automatically annotated by an annotation system.

[0057] The first audio data set may also include audio data of other industrial scenes, such as the equipment sounds at wind power generation sites, the equipment sounds at hydropower generation sites, the equipment sounds at production and processing scenes, etc. For audio data of corresponding industrial scenes, the corresponding labels represent whether the equipment has a fault and the fault category when a fault exists. The labels of audio data can be manually annotated or automatically annotated by an annotation system.

[0058] As an example, for each audio data in the first audio data set, each audio frame of the audio data can be labeled separately, and the label of each audio frame represents whether there is noise or fault in the audio frame, and the type of noise when there is noise or the type of fault when there is a fault.

[0059] The audio data in the second audio data set is audio data in a belt conveyor scenario, and the label of the audio data represents whether the belt conveyor has a fault and the fault type when a fault exists.

[0060] The labels of the audio data in the second audio data set may be manually annotated or automatically annotated by a labeling system.

[0061] As an example, for each audio data in the second audio data set, each audio frame of the audio data may be labeled respectively, and the label of each audio frame indicates whether the audio frame has a fault and the fault type when a fault exists.

[0062] Since labeled data is not easy to obtain in the belt conveyor scenario, the number of labeled data in the second audio data set is small, and the fault model obtained by training the fault recognition model using only the second audio data set has poor generalization ability and is prone to overfitting problems. In order to enhance the generalization ability of the fault recognition model and alleviate the overfitting problem, the first audio data set can be used to pre-train the fault recognition model so that the fault recognition model can learn and recognize features related to belt conveyor faults on the first audio data set, such as visual features.

[0063] In an optional embodiment, the process of pre-training the fault recognition model using the first audio data set and fine-tuning the pre-trained fault recognition model using the second audio data set may include: Inputting the acoustic features of the audio data in the first audio data set into the fault recognition model, obtaining recognition results of each audio data in the first audio data set (referred to as first recognition results for ease of description and distinction); taking the first recognition results of each audio data in the first audio data set as the goal of being close to the labels of the audio data in the first audio data set, updating the parameters of the fault recognition model, and obtaining a pre-trained fault recognition model; The first recognition result of each audio data characterizes whether there is noise or fault in the audio data, and the type of noise when there is noise or the type of fault when there is fault.

[0064] For each audio data in the first audio data, the audio data can be framed to obtain multiple audio frames, and the acoustic features of each audio frame can be extracted. The acoustic features of each audio frame are input into a fault recognition model, and the fault recognition model encodes the acoustic features of each audio frame to obtain the encoding features of each audio frame. Based on the encoding features of each audio frame, a first recognition result of each audio frame is determined, and the first recognition result of each audio frame characterizes whether there is noise or fault in the audio frame, and the type of noise or fault when noise or fault exists.

[0065] Calculate the loss of the first recognition result and the label of each audio frame; update the parameters of the fault recognition model with the goal of making the loss of each audio frame smaller and smaller (that is, compared with the first recognition result obtained by processing the audio data using the fault recognition model before the parameter update, the loss of the first recognition result and the label obtained by processing the audio data using the fault recognition model after the parameter update is smaller), and obtain a pre-trained fault recognition model.

[0066] The acoustic features of the audio data in the second audio data set are input into the pre-trained fault recognition model to obtain recognition results of each audio data in the second audio data set (referred to as the second recognition result for ease of description and distinction); with the second recognition results of each audio data in the second audio data set approaching the labels of the audio data in the second audio data set as the goal, the parameters of the target layer of the encoding module and the parameters of the non-coding module of the pre-trained fault recognition model are updated to obtain a trained fault recognition model; the output of the non-target layer of the encoding module is connected to the input of the target layer; the input of the non-target layer is the input of the fault recognition model, the non-target layer does not include the channel attention layer, and the target layer includes the channel attention layer.

[0067] The second recognition result of each audio data indicates whether there is a fault in the audio data, and the type of the fault when there is a fault.

[0068] For each audio data in the second audio data, the audio data can be framed to obtain multiple audio frames, and the acoustic features of each audio frame can be extracted. The acoustic features of each audio frame are input into a pre-trained fault recognition model, and the pre-trained fault recognition model encodes the acoustic features of each audio frame to obtain the encoding features of each audio frame. Based on the encoding features of each audio frame, the second recognition result of each audio frame is determined, and the second recognition result of each audio frame characterizes whether there is a fault in the audio frame, and the type of fault when a fault exists.

[0069] Calculate the loss of the second recognition result and the label of each audio frame; update some parameters of the pre-trained fault recognition model with the goal of reducing the loss of each audio frame. That is, when fine-tuning the pre-trained fault recognition model, freeze some parameters and only update the parameters that do not need to be frozen.

[0070] Taking the above-mentioned multiple layers of the encoding module as follows: the first 5 layers are convolutional, the 6th layer is a channel attention layer, the 7th-8th layers are convolutional layers, the 9th layer is a channel attention layer, the 10th-11th layers are convolutional layers, and the 12th layer is a channel attention layer as an example, when pre-training the fault recognition model, the parameters of the encoding module and the non-coding module of the fault recognition model are updated; when fine-tuning the pre-trained fault recognition model, the parameters of the first 5 layers of the encoding module are frozen, and only the parameters after the 5th layer (including the parameters of the 6th-12th layers of the encoding module and the parameters of the non-coding module of the fault recognition model) are updated. In this example, the first 5 layers of the encoding module constitute the non-target layer of the encoding module, and the 6th-12th layers of the encoding module constitute the target layer of the encoding module.

[0071] As an example, the fault recognition model may include an encoding module and a prediction module, wherein the encoding module is used to encode each audio frame of the audio data based on the channel attention mechanism to obtain the encoding features of each audio frame, and the prediction module is used to determine the recognition results of each audio frame according to the encoding features of each audio frame. When pre-training the fault recognition model, the parameters of the encoding module and the prediction module are updated; when fine-tuning the pre-trained fault recognition model, the parameters of the non-target layer of the encoding module are frozen, and only the parameters of the target layer of the encoding module and the parameters of the prediction module are updated.

[0072] Corresponding to the method embodiment, the present application also provides a fault monitoring device. A structural schematic diagram of the fault monitoring device provided in the embodiment of the present application is shown in FIG. Figure 4 As shown, it may include: Collection module 401, encoding module 402 and recognition module 403; The acquisition module 401 is used to collect audio signals of various monitoring points of the belt conveyor through optical fibers arranged along the belt movement direction of the belt conveyor; The encoding module 402 is used to encode the audio signal of each monitoring point based on the channel attention mechanism to obtain the encoding features of each monitoring point; The identification module 403 is used to correspond to each monitoring point, identify the fault of the monitoring point based on the coding characteristics of the monitoring point, and obtain the identification result of the monitoring point. The identification result of the monitoring point indicates whether the belt conveyor has a fault near the monitoring point and the type of fault when a fault occurs.

[0073] The fault monitoring device provided in the embodiment of the present application collects audio signals of each monitoring point of the belt conveyor through optical fibers arranged along the belt movement direction of the belt conveyor, encodes the audio signal of each monitoring point based on the channel attention mechanism, and obtains the encoding feature of each monitoring point; corresponding to each monitoring point, fault identification is performed on the monitoring point based on the encoding feature of the monitoring point to obtain an identification result of the monitoring point, which characterizes whether there is a fault in the belt conveyor near the monitoring point and the type of fault when a fault exists, thereby achieving the purpose of automatically monitoring the fault of the belt conveyor, reducing the monitoring cost and improving the monitoring efficiency while ensuring the safety of the staff.

[0074] In an optional embodiment, the optical fiber is arranged on a bracket on the belt conveyor parallel to the moving direction of the belt, and the optical fiber is located below or to the side of the roller supporting the belt.

[0075] In an optional embodiment, when the acquisition module 401 collects the audio signals of each monitoring point of the belt conveyor through the optical fiber arranged along the belt movement direction of the belt conveyor, it is used to: The audio signals of each monitoring point of the belt conveyor are collected through two optical fibers arranged in parallel on both sides of the belt; The monitoring points arranged on both sides of the belt are located at different positions in the moving direction of the belt.

[0076] In an optional embodiment, the encoding module 402 encodes the audio signal of each monitoring point based on the channel attention mechanism to obtain the encoding feature of each monitoring point, which is used to: Corresponding to each monitoring point, the audio signal of the monitoring point is adaptively filtered to obtain a filtered audio signal; the background noise of the filtered audio signal is less than the background noise of the audio signal of the monitoring point; The filtered audio signal is encoded based on the channel attention mechanism to obtain the encoding features of the monitoring point.

[0077] In an optional embodiment, the encoding module 402 encodes the audio signal of each monitoring point based on the channel attention mechanism, and the identification module 403 corresponds to each monitoring point, and when performing fault identification on the monitoring point based on the encoding feature of the monitoring point, is used to: The fault recognition model encodes the audio signal of each monitoring point based on the channel attention mechanism, and identifies the fault of each monitoring point based on the encoding features of the monitoring point to obtain the identification result of each monitoring point. The fault recognition model is obtained by pre-training with the first audio data set, and fine-tuning the pre-trained fault recognition model with the second audio data set; The audio data in the first audio data set is audio data in a non-belt conveyor scenario, and the labels of the audio data in the first audio data set represent whether the audio data has noise or faults, and the noise category when noise exists or the fault category when faults exist; The audio data in the second audio data set is audio data in a belt conveyor scenario, and the label of the audio data in the second audio data set represents whether there is a fault in the audio data and the fault type when there is a fault.

[0078] In an optional embodiment, the fault monitoring device further includes a training module, which is used to pre-train the fault recognition model using the first audio data set, and fine-tune the pre-trained fault recognition model using the second audio data set, specifically including: Inputting the acoustic features of the audio data in the first audio data set into the fault recognition model to obtain a first recognition result of each audio data in the first audio data set; with the goal of making the first recognition result of each audio data in the first audio data set close to the label of the audio data in the first audio data set, updating the parameters of the fault recognition model to obtain a pre-trained fault recognition model; Inputting the acoustic features of the audio data in the second audio data set into the pre-trained fault recognition model to obtain a second recognition result of each audio data in the second audio data set; taking the second recognition result of each audio data in the second audio data set as a goal to be close to the label of the audio data in the second audio data set, updating the parameters of the target layer of the encoding module and the parameters of the non-encoding module of the pre-trained fault recognition model to obtain a trained fault recognition model; The output of the non-target layer of the encoding module is connected to the input of the target layer; the input of the non-target layer is the input of the fault identification model, the non-target layer does not include a channel attention layer, and the target layer includes a channel attention layer.

[0079] The present application also provides an electronic device in an embodiment. Figure 5 As shown, it shows a structural schematic diagram of an electronic device suitable for implementing the fault monitoring method in the embodiment of the present application. The electronic device in the embodiment of the present application can be a terminal device (for example, a mobile phone, a tablet computer, a laptop computer, a desktop computer, etc.), or a server (can be a single server, can be a server cluster, or can be a cloud server, etc.). Figure 5 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0080] like Figure 5As shown, the electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage device 508 to a random access memory (RAM) 503. When the electronic device is powered on, various programs and data required for the operation of the electronic device are also stored in the RAM 503. The processing device 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0081] Typically, the following devices may be connected to the I / O interface 505: an input device 506 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 507 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 508 including, for example, a memory card, a hard disk, etc.; and a communication device 509. The communication device 509 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although Figure 5 An electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0082] An embodiment of the present application also provides a computer program product including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements any fault monitoring method provided in the embodiment of the present application.

[0083] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any fault monitoring method provided in the embodiment of the present application.

[0084] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0085] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0086] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware, or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product. Professionals and technicians may use different methods to implement the described functions for each specific solution, but such implementation should not be considered beyond the scope of this application.

[0087] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

[0088] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0089] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A fault monitoring method, characterized in that: The method comprises: The audio signals of each monitoring point of the belt conveyor are collected through optical fibers arranged along the belt movement direction of the belt conveyor; Encode the audio signal of each monitoring point based on the channel attention mechanism to obtain the encoding features of each monitoring point; Corresponding to each monitoring point, fault identification is performed on the monitoring point based on the coding characteristics of the monitoring point to obtain an identification result of the monitoring point, which characterizes whether the belt conveyor has a fault near the monitoring point and the type of fault when a fault exists.

2. The method according to claim 1, characterized in that The optical fiber is arranged on a bracket on the belt conveyor which is parallel to the moving direction of the belt, and the optical fiber is located below or to the side of the roller supporting the belt.

3. The method according to claim 1, characterized in that The method of collecting audio signals of various monitoring points of the belt conveyor through optical fibers arranged along the belt movement direction of the belt conveyor includes: The audio signals of each monitoring point of the belt conveyor are collected through two optical fibers arranged in parallel on both sides of the belt; The monitoring points arranged on both sides of the belt are located at different positions in the moving direction of the belt.

4. The method according to claim 1, characterized in that The audio signal of each monitoring point is encoded based on the channel attention mechanism to obtain the encoding features of each monitoring point, including: Corresponding to each monitoring point, the audio signal of the monitoring point is adaptively filtered to obtain a filtered audio signal; the background noise of the filtered audio signal is less than the background noise of the audio signal of the monitoring point; The filtered audio signal is encoded based on the channel attention mechanism to obtain the encoding features of the monitoring point.

5. The method according to claim 1, characterized in that The audio signal of each monitoring point is encoded based on the channel attention mechanism. For each monitoring point, the process of fault identification for the monitoring point based on the encoding features of the monitoring point includes: The fault recognition model encodes the audio signal of each monitoring point based on the channel attention mechanism, and identifies the fault of each monitoring point based on the encoding features of the monitoring point to obtain the identification result of each monitoring point. The fault recognition model is obtained by pre-training with the first audio data set, and fine-tuning the pre-trained fault recognition model with the second audio data set; The audio data in the first audio data set is audio data in a non-belt conveyor scenario, and the labels of the audio data in the first audio data set represent whether the audio data has noise or faults, and the noise category when noise exists or the fault category when faults exist; The audio data in the second audio data set is audio data in a belt conveyor scenario, and the label of the audio data in the second audio data set represents whether there is a fault in the audio data and the fault type when there is a fault.

6. The method according to claim 5, characterized in that The process of pre-training the fault recognition model by using the first audio data set and fine-tuning the pre-trained fault recognition model by using the second audio data set includes: Inputting the acoustic features of the audio data in the first audio data set into the fault recognition model to obtain a first recognition result of each audio data in the first audio data set; with the goal of making the first recognition result of each audio data in the first audio data set close to the label of the audio data in the first audio data set, updating the parameters of the fault recognition model to obtain a pre-trained fault recognition model; Inputting the acoustic features of the audio data in the second audio data set into the pre-trained fault recognition model to obtain a second recognition result of each audio data in the second audio data set; taking the second recognition result of each audio data in the second audio data set as a goal to be close to the label of the audio data in the second audio data set, updating the parameters of the target layer of the encoding module and the parameters of the non-encoding module of the pre-trained fault recognition model to obtain a trained fault recognition model; The output of the non-target layer of the encoding module is connected to the input of the target layer; the input of the non-target layer is the input of the fault identification model, the non-target layer does not include a channel attention layer, and the target layer includes a channel attention layer.

7. A fault monitoring device, characterized in that: include: A collection module, used for collecting audio signals of various monitoring points of the belt conveyor through optical fibers arranged along the belt movement direction of the belt conveyor; The encoding module is used to encode the audio signal of each monitoring point based on the channel attention mechanism to obtain the encoding features of each monitoring point; The identification module is used to identify the fault of each monitoring point based on the coding characteristics of the monitoring point to obtain the identification result of the monitoring point. The identification result of the monitoring point indicates whether there is a fault in the belt conveyor near the monitoring point and the type of fault when there is a fault.

8. A computer program product, characterized in that It comprises computer-readable instructions, and when the computer-readable instructions are executed on an electronic device, the electronic device implements the fault monitoring method as claimed in any one of claims 1 to 6.

9. An electronic device, characterized in that: The electronic device comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the electronic device can implement the fault monitoring method according to any one of claims 1 to 6.

10. A computer storage medium, characterized in that: The storage medium carries one or more computer programs, and when the one or more computer programs are executed by an electronic device, the electronic device can implement the fault monitoring method as described in any one of claims 1 to 6.

Images