Method and device for monitoring jamming state of bearing, storage medium and computer program product

Abstract

The invention discloses a method and device for monitoring the jamming state of a bearing, a storage medium and a computer program product, and relates to the field of fan system monitoring, and the method for monitoring the jamming state of the bearing comprises the following steps: carrying out signal conversion processing on a received acoustic emission electric signal to obtain an acoustic emission digital signal, the acoustic emission electric signal is obtained by amplifying an original acoustic signal by an amplifier module installed in the second area, and the original acoustic signal is acquired by a sensor module located in the first area in the acquisition device in the operation process of the target bearing part to be detected; the distance between the position of the first area and the position of the second area meets a preset distance condition; and inputting the at least one acoustic emission digital signal into the state prediction model, and generating early warning information matched with the state of the at least one to-be-detected bearing. By adopting the technical scheme, the problem that the wind measurement data of the anemograph is inaccurate due to bearing jamming cannot be found in time in the related technology is solved.

Description

Technical Field

[0001] The present application relates to the field of fan system monitoring, and more specifically, to a method and device for monitoring a bearing jam state, a storage medium, and a computer program product. Background Art

[0002] Wind resources are an important renewable energy source. Wind turbines can convert wind energy into electrical energy. Wind speed is an important indicator, and its size directly affects the efficiency of wind turbines in capturing wind energy. The anemometer is an instrument for measuring wind speed. Its measurement accuracy will affect the economic evaluation results of wind turbines and the economic benefits of wind farms.

[0003] However, the operating environment of wind turbines is harsh, and the anemometer is installed outside the wind turbine, where the environment is even harsher. Under the influence of low temperature and windy sand, the anemometer bearings are easily worn, resulting in jamming, which makes the measured value of the anemometer smaller. The control system of the wind turbine performs incorrect control based on the inaccurate data of the anemometer, which will affect the working efficiency of the wind turbine and even shut it down. There is currently no effective solution to the problem that the relevant technology cannot promptly detect the inaccurate wind measurement data caused by the anemometer bearing jamming.

[0004] Therefore, it is necessary to improve the related technology to overcome the above-mentioned defects in the related technology. Summary of the invention

[0005] The embodiments of the present application provide a method and device for monitoring the stuck state of a bearing, a storage medium, and a computer program product, so as to at least solve the problem in the related art that an anemometer cannot promptly detect inaccurate wind measurement data caused by a stuck bearing.

[0006] According to one aspect of an embodiment of the present application, a method for monitoring a stuck state of a bearing is provided, comprising: performing signal conversion processing on a received acoustic emission electrical signal to obtain an acoustic emission digital signal, wherein the acoustic emission electrical signal is obtained by amplifying an original acoustic signal by an amplifier module installed in a second area, and the original acoustic signal is collected by a sensor module located in a first area in a collection device during operation of a target bearing component to be tested, and the distance between a position of the first area and a position of the second area meets a preset distance condition; at least one acoustic emission digital signal is input into a state prediction model to obtain a bearing state to be tested that matches each of the at least one acoustic emission digital signal; and when at least one bearing state to be tested meets a warning condition, generating warning information matching at least one bearing state to be tested.

[0007] In an exemplary embodiment, at least one acoustic emission digital signal is input into a state prediction model to obtain a bearing state to be tested that matches the at least one acoustic emission digital signal, including: determining a signal feature that matches the acoustic emission digital signal through a feature extraction layer of the state prediction model; performing nonlinear processing on the signal feature according to an activation function to obtain an output result; and performing state mapping processing on the output result to obtain a bearing state to be tested that matches the acoustic emission digital signal.

[0008] In an exemplary embodiment, after determining the signal features that match the acoustic emission digital signal through the feature extraction layer of the state prediction model, it includes: performing an important feature extraction operation on the signal features to obtain local sound features that match the signal features, wherein the local sound features include vibration information of the acoustic emission digital signal; and determining the probability of matching each bearing state to be tested based on the local sound features.

[0009] In an exemplary embodiment, the output result is subjected to state mapping processing to obtain the state of the bearing to be tested that matches the acoustic emission digital signal, and also includes: determining parameter state information that matches the bearing to be tested based on the output result, wherein the parameter state information includes at least one of the following: signal time domain information, signal frequency domain information; determining the bearing state of the bearing to be tested based on the parameter state information.

[0010] In an exemplary embodiment, determining the state of a bearing to be tested according to parameter state information also includes: when the parameter state information is signal time domain information, performing statistical value processing on the signal time domain information to determine amplitude distribution information that matches the parameter state information; when the parameter state information is signal frequency domain information, performing spectral processing on the signal frequency domain information to determine energy distribution information that matches the parameter state information.

[0011] In an exemplary embodiment, before the received acoustic emission electrical signal is converted into an acoustic emission digital signal, the method further includes: setting the distance between the position of the first area and the position of the bearing to be measured to be less than the first distance; and setting the distance between the position of the second area and the position of the bearing to be measured to be greater than the second distance.

[0012] In an exemplary embodiment, the received acoustic emission electrical signal is subjected to signal conversion processing to obtain an acoustic emission digital signal, including: a brain-like edge computing device receives the acoustic emission electrical signal, wherein an amplifier module is integrated on the brain-like edge computing device; the acoustic emission electrical signal is amplified by the amplifier module in the brain-like edge computing device; and the amplified acoustic emission electrical signal is filtered and analog-to-digital converted to obtain an acoustic emission digital signal.

[0013] According to another aspect of an embodiment of the present application, a device for monitoring the stuck state of a bearing is also provided, including: a signal processing unit, used to perform signal conversion processing on the received acoustic emission electrical signal to obtain an acoustic emission digital signal, wherein the acoustic emission electrical signal is obtained by amplifying the original acoustic signal by an amplifier module installed in the second area, and the original acoustic signal is collected by a sensor module located in the first area in the acquisition device during the operation of the target bearing component to be tested, and the distance between the position of the first area and the position of the second area meets the preset distance condition; a state prediction unit, used to input at least one acoustic emission digital signal into a state prediction model to obtain a bearing state to be tested that matches each of the at least one acoustic emission digital signal; when at least one bearing state to be tested meets the warning condition, generate warning information matching at least one bearing state to be tested.

[0014] According to another aspect of the embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned method for monitoring the stuck state of a bearing when running.

[0015] According to another aspect of an embodiment of the present application, an electronic device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the method for monitoring the stuck state of a bearing through the computer program.

[0016] According to another aspect of the embodiments of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps of the method described in each embodiment of the present application are implemented.

[0017] Through the present application, the received acoustic emission electrical signal is converted and processed to obtain an acoustic emission digital signal, wherein the acoustic emission electrical signal is obtained by amplifying the original acoustic signal by the amplifier module installed in the second area, and the original acoustic signal is collected by the sensor module located in the first area in the acquisition device during the operation of the target bearing component to be tested, and the distance between the position of the first area and the position of the second area meets the preset distance condition; at least one acoustic emission digital signal is input into the state prediction model to obtain the state of the bearing to be tested that matches each of the at least one acoustic emission digital signal; when at least one bearing state to be tested meets the warning condition, a warning message matching at least one bearing state to be tested is generated. This solves the problem in the related art that the anemometer cannot promptly detect the inaccurate wind measurement data caused by the bearing jam. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0019] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0020] Figure 1 It is a hardware structure block diagram of a computer terminal of a method for monitoring a bearing stuck state according to an embodiment of the present application;

[0021] Figure 2 is a flow chart of a method for monitoring a bearing stuck state according to an embodiment of the present application;

[0022] Figure 3 is a schematic diagram of a method for monitoring a bearing jam state according to an embodiment of the present application;

[0023] Figure 4 is a flow chart of another method for monitoring a bearing stuck state according to an embodiment of the present application;

[0024] Figure 5 It is a structural block diagram of a device for monitoring the stuck state of a bearing according to an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0026] It should be noted that 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 are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] The method embodiments provided in the embodiments of the present application can be executed in a computer terminal or a similar computing device. Taking running on a computer terminal as an example, Figure 1 1 is a hardware structure block diagram of a computer terminal for monitoring a bearing jam state according to an embodiment of the present application. Figure 1 As shown, the computer terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor (Central Processing Unit, MCU) or a programmable logic device (Field Programmable Gate Array, FPGA)) and a memory 104 for storing data, wherein the above-mentioned computer terminal may also include a transmission device 106 and an input and output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above-mentioned computer terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0028] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the method for monitoring the stuck state of a bearing in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0029] A wireless network provided by a communication provider of a computer terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (Radio Frequency, referred to as RF) module, which is used to communicate with the Internet wirelessly.

[0030] In this embodiment, a method for monitoring the stuck state of a bearing is provided. Figure 2 is a flow chart of a method for monitoring a bearing stuck state according to an embodiment of the present application, such as Figure 2 As shown, the process includes the following steps:

[0031] S202, performing signal conversion processing on the received acoustic emission electrical signal to obtain an acoustic emission digital signal, wherein the acoustic emission electrical signal is obtained by amplifying the original acoustic signal by the amplifier module installed in the second area, and the original acoustic signal is collected by the sensor module located in the first area in the acquisition device during the operation of the target bearing component to be tested, and the distance between the position of the first area and the position of the second area meets the preset distance condition;

[0032] S204, inputting at least one acoustic emission digital signal into a state prediction model to obtain a bearing state to be measured that matches each of the at least one acoustic emission digital signal;

[0033] S206: When at least one bearing state to be measured satisfies the warning condition, generate warning information matching the at least one bearing state to be measured.

[0034] Optionally, in the above step S202, the received acoustic emission electrical signal is subjected to signal conversion processing to obtain an acoustic emission digital signal, wherein the acoustic emission electrical signal is obtained by amplifying the original acoustic signal by an amplifier module installed in the second area, and the original acoustic signal is collected by a sensor module located in the first area in the acquisition device during the operation of the target bearing component to be tested, and the distance between the position of the first area and the position of the second area meets the preset distance condition.

[0035] The above-mentioned acoustic emission digital signal can be a periodic impact signal collected by the sound sensor when a rolling bearing fails. These signals are generated by rotating parts such as rotors and bearings. The sound signal emitted by the bearing also includes a large amount of environmental interference signals such as aerodynamic noise, electromagnetic noise, construction interference noise, and personnel communication noise.

[0036] The above-mentioned preset distance condition is a close or far condition. In this solution, the collection device and the amplification device are separated from each other as an illustration. The specific distance range can be determined according to actual conditions.

[0037] As an optional implementation, the monitoring device implementing the above method is as follows: Figure 3 As shown, for example, it includes an acoustic emission sensor 1 and a brain-like edge computing device 2, and the anemometer bearing 3 is the device to be monitored. The acoustic emission sensor is used to collect the acoustic emission signal during the operation of the anemometer, and transmit the collected signal to the brain-like edge computing device. The device filters, amplifies, A / D converts, etc. the electrical signal, converts the electrical signal into a digital signal, and inputs it into a deep learning convolutional neural network. The algorithm extracts signal features according to the trained model and establishes the relationship between the selected features and the anemometer bearing state.

[0038] It should be noted that acoustic emission monitoring is an effective bearing jam monitoring method that can identify early bearing jams. Due to the influence of environmental noise, the sensor should be as close to the monitored part as possible and away from noise interference. Therefore, the sensor inside the bearing is the most ideal location. The function of the amplifier is to amplify the weak electrical signal of the sensor. The amplified signal is not easily affected by noise and can be transmitted over long distances. Therefore, this method selects Figure 3 The sensor and amplifier are placed separately to reduce the size of the sensor, thereby placing the sensor inside the anemometer close to the bearing position.

[0039] In step S204, at least one acoustic emission digital signal is input into a state prediction model to obtain a bearing state to be measured that matches each of the at least one acoustic emission digital signal;

[0040] Specifically, the collected signals can be transmitted to the brain-like edge computing device, which filters, amplifies, A / D converts, etc. the electrical signals, converts them into digital signals, and inputs them into the deep learning convolutional neural network. The algorithm extracts signal features based on the trained model and establishes the relationship between the selected features and the bearing status of the anemometer.

[0041] According to the algorithm, the edge computing device will analyze the collected data directly in the local device or network close to the data generation, without the need to transmit the data to the cloud data processing center. Brain-like computing aims to imitate the processing method of the human brain. It is designed to simulate the human brain and has greater advantages in power consumption and learning ability compared to traditional chips.

[0042] Traditionally, computer chips are designed according to the von Neumann architecture, with storage and computing separated in space. Every time a computer performs a calculation, it needs to call back and forth between the CPU and memory. Frequent data exchange leads to low efficiency in processing massive amounts of information. The research on brain-like chips is based on the combination of microelectronics technology and new neuromorphic devices, breaking through traditional computing architecture, achieving deep integration of storage and computing, greatly improving computing performance, increasing integration, and reducing energy consumption. Deep learning convolutional neural networks based on brain-like chips can more efficiently extract and identify features of input data.

[0043] This approach, which uses neural networks and machine learning algorithms to process and analyze data, has potential advantages in processing complex and unstructured data. Brain-like edge computing attempts to implement these advanced algorithms on edge devices to improve decision-making speed and efficiency.

[0044] In the above step S206, when at least one bearing state to be measured satisfies the warning condition, warning information matching the at least one bearing state to be measured is generated.

[0045] It should be noted that the above-mentioned warning conditions may be, but are not limited to, conditions such as the frequency of the acoustic emission signal being higher than a preset threshold, a sharp increase in the signal amplitude in a short period of time, a significant increase in the fault characteristic frequency, etc. It may also be a combination of various abnormal conditions, etc., and no specific restrictions are made here.

[0046] In an optional implementation, if a burst feature is found in the acoustic emission signal of bearing A, and a significant increase in the characteristic frequency of the bearing fault is detected in the vibration signal, and the temperature of bearing A also increases slightly, the following warning information can be generated: Bearing number: A-Warning level: High-Warning time: 2024-11-08 10:30:00-Warning reason: The acoustic emission signal is abnormal, the characteristic frequency of the fault is detected in the vibration signal, and the temperature increases slightly. -Recommended measures: Check bearing A immediately, there may be early wear or damage, and further detailed inspection and maintenance are recommended.

[0047] It should be noted that different levels of warning information and warning strategies can be issued according to different abnormal signals. For example, an alarm that lasts for 5 seconds is a level 2 alarm, an alarm that lasts for 10 seconds and the bearing stops working is a level 1 alarm, and other different warning methods are not specifically limited here.

[0048] Through the present application, the received acoustic emission electrical signal is converted and processed to obtain an acoustic emission digital signal, wherein the acoustic emission electrical signal is obtained by amplifying the original acoustic signal by the amplifier module installed in the second area, and the original acoustic signal is collected by the sensor module located in the first area in the acquisition device during the operation of the target bearing component to be tested, and the distance between the position of the first area and the position of the second area meets the preset distance condition; at least one acoustic emission digital signal is input into the state prediction model to obtain the state of the bearing to be tested that matches each of the at least one acoustic emission digital signal; when at least one bearing state to be tested meets the warning condition, a warning message matching at least one bearing state to be tested is generated. This solves the problem in the related art that the anemometer cannot promptly detect the inaccurate wind measurement data caused by the bearing jam.

[0049] In an optional implementation, at least one acoustic emission digital signal is input into a state prediction model to obtain a bearing state to be measured that matches each of the at least one acoustic emission digital signal, including:

[0050] S1, determining the signal features matching the acoustic emission digital signal through the feature extraction layer of the state prediction model;

[0051] S2, performs nonlinear processing on the signal features according to the activation function to obtain the output result;

[0052] S3, performing state mapping processing on the output result to obtain the bearing state to be tested that matches the acoustic emission digital signal.

[0053] In the above steps S1-S3, the signal features matching the acoustic emission digital signal are determined through the feature extraction layer of the state prediction model; the signal features are nonlinearly processed according to the activation function to obtain the output result; the output result is state mapped to obtain the bearing state to be tested matching the acoustic emission digital signal.

[0054] As an optional implementation, for example, in S1, the first layer of the model is a convolutional layer for extracting features of the sound signal. This layer specifies the input shape and may include some regularization techniques, such as L2 regularization, to prevent overfitting.

[0055] Then, in step S2, an activation function is used to extract useful features from the input data. During the training process, the neural network adjusts parameters such as weights and biases so that the activation function can extract important features, thereby improving the classification and prediction capabilities of the model. For acoustic signals, this means that the network can identify key features in the sound, such as frequency, rhythm, and tone.

[0056] Further, in step S3, a softmax activation function may be used for multi-classification to determine the stuck state of the bearing, for example, the bearing state to be tested that matches the acoustic emission digital signal is determined based on the maximum value between the probabilities of outputting various stuck states.

[0057] Through the above implementation, the model structure consists of an input layer, a convolution layer, an activation function, a pooling layer and a fully connected layer. After the acoustic emission data is input into the model, the set convolution kernel slides on the data and extracts features, and the ReLU activation function is used to perform a nonlinear transformation on the output of the convolution layer to enhance the expression ability of the model. The pooling layer reduces the dimension of the convolution layer output, reduces the amount of calculation, and improves the robustness and generalization ability of the model. The fully connected layer maps the output of the pooling layer to the output of the model, establishes a mapping from the acoustic emission signal to the bearing state, and then monitors the stuck state of the bearing, solving the problem in the related technology that the anemometer cannot promptly detect the inaccurate wind measurement data caused by the bearing stuck.

[0058] In an optional implementation, after determining the signal features matching the acoustic emission digital signal through the feature extraction layer of the state prediction model, the method includes:

[0059] S1, performing an important feature extraction operation on the signal feature to obtain a local sound feature that matches the signal feature, wherein the local sound feature includes vibration information of the acoustic emission digital signal;

[0060] S2, determining the probability of matching each bearing state to be tested according to the local characteristics of the sound.

[0061] In the above steps S1-S2, an important feature extraction operation is performed on the signal features to obtain local sound features that match the signal features, wherein the local sound features include vibration information of the acoustic emission digital signal; and the probability of matching each bearing state to be tested is determined based on the local sound features.

[0062] The above steps S1-S2 are described below in a specific implementation manner:

[0063] Assume that the classification task of the bearing jam state has three categories: normal (0), slight jam (1), and severe jam (2). First, the vibration signal data of the bearing is input into the neural network model, and then the signal features of the sound signal are extracted. The extracted signal features are then processed by important feature extraction. For example, the neural network can adjust parameters such as weights and biases so that the activation function can extract important features to obtain local sound features and pass them to the output layer of the neural network model. The output layer has three neurons corresponding to the three categories.

[0064] Apply the softmax function to the raw scores of the output layer to get the probability distribution of each category. Select the category with the highest probability as the prediction result. For example, if the output of softmax is [0.1, 0.3, 0.6], the prediction result is severely stuck (i.e. category 2).

[0065] In an optional implementation, the output result is subjected to state mapping processing to obtain a bearing state to be tested that matches the acoustic emission digital signal, and further includes:

[0066] S1, determining parameter state information matching the bearing to be tested according to the output result, wherein the parameter state information includes at least one of the following: signal time domain information and signal frequency domain information;

[0067] S2, determining the bearing state of the bearing to be tested according to the parameter state information.

[0068] In the above implementation S1-S2, the bearing state can be determined according to the time-frequency domain statistical characteristics of the bearing data by the following steps:

[0069] First, it can be time domain analysis:

[0070] Statistical indicators: In the time domain, the statistical indicators of healthy bearings and faulty bearings can be compared, such as the root mean square (RMS). When a bearing fails, the RMS will increase significantly, which can reflect the operating status of the bearing in real time. Probability density function: Analyze the probability density function of the bearing vibration signal to observe the distribution difference between healthy bearings and faulty bearings.

[0071] Secondly, frequency domain analysis: Power spectral density function: In the frequency domain, the power spectral density function (PSD) of the faulty bearing is calculated to identify the frequency components of the bearing fault. Envelope spectrum analysis: Envelope spectrum analysis is performed to extract the frequency characteristics of the bearing fault.

[0072] Further, time-frequency domain analysis: Short-time Fourier transform (STFT): Use STFT to detect fault information from the vibration signal. STFT can provide the time-frequency representation of the signal. Wavelet transform (WT): Use wavelet transform to extract the time-frequency characteristics of the signal. Wavelet transform is suitable for non-stationary signal analysis. Wavelet packet transform (WPT): The time-frequency analysis of the signal is further refined through wavelet packet transform to detect more subtle fault characteristics.

[0073] And it can also include, signal decomposition methods: EMD and VMD: use signal decomposition methods such as empirical mode decomposition (EMD) and variational mode decomposition (VMD) combined with envelope analysis to extract bearing fault frequency. Feature extraction and condition monitoring: Time domain features: extract time domain features, such as root mean square, peak value, signal kurtosis, etc., for bearing condition monitoring and fault diagnosis. Frequency domain features: extract frequency domain features, such as peak frequency, average frequency, etc., to identify bearing fault frequency.

[0074] In an optional implementation manner, determining the bearing state of the bearing to be tested according to the parameter state information further includes:

[0075] S1, when the parameter state information is signal time domain information, performing statistical value processing on the signal time domain information to determine amplitude distribution information matching the parameter state information;

[0076] S2: When the parameter state information is signal frequency domain information, perform spectrum processing on the signal frequency domain information to determine energy distribution information matching the parameter state information.

[0077] As an optional implementation, in step S1, when the parameter state information is signal time domain information, statistical value processing is performed on the signal time domain information to determine amplitude distribution information matching the parameter state information.

[0078] Optionally, in the time domain analysis, various statistical features of the bearing vibration signal can be extracted, such as: root mean square error (measurement of signal amplitude), peak value (reflecting the amplitude range of the signal), absolute mean, root square amplitude (average level of the signal and square root of the amplitude), kurtosis, skewness (describing the shape of the signal distribution, including sharpness and skewness), form factor, peak factor, pulse factor, margin factor (reflecting the waveform characteristics of the signal).

[0079] Furthermore, by analyzing these time domain features, the health status of the bearing can be monitored. For example, the kurtosis index of a faulty bearing is greater than 3. It should be noted that different time domain features have different sensitivities to the degree of bearing fault.

[0080] In step S2, when the parameter state information is signal frequency domain information, spectrum processing is performed on the signal frequency domain information to determine energy distribution information matching the parameter state information.

[0081] Optionally, in frequency domain analysis, power spectrum density (PSD) analysis can be used to extract the energy of each frequency component in the signal, frequency domain feature extraction indicators such as centroid frequency (describing the change in the position of the main frequency of the power spectrum), mean square frequency, frequency variance (describing the degree of dispersion of spectral energy), etc., or Hilbert envelope analysis, specifically, a given continuous signal is transformed into an analytical signal, the real part of the analytical signal is the signal itself, the imaginary part is its Hilbert transform, and the amplitude of the analytical signal is the envelope of the signal, that is, the envelope signal. By performing feature extraction on the spectrum analysis of the envelope signal, the low-frequency modulation signal related to the fault can be extracted, and the location and degree of the fault can be determined by observing the spectrum diagram and analyzing the changes in the modulation frequency and its multiple frequency amplitude.

[0082] Through the above method, frequency domain features can be effectively extracted from the bearing vibration signal, and these features can be used to judge the health status of the bearing.

[0083] In an optional implementation, before performing signal conversion processing on the received acoustic emission electrical signal to obtain an acoustic emission digital signal, the method further includes:

[0084] S1, setting the distance between the position of the first area and the position of the bearing to be tested to be less than a first distance;

[0085] S2, setting the distance between the position of the second area and the position of the bearing to be measured to be greater than the second distance.

[0086] In step S1, as an optional implementation, due to the influence of environmental noise, the sensor should be as close to the monitored part as possible and away from noise interference. Therefore, the sensor can be fixed inside the anemometer alone through a coupling agent, close to the bearing, for example, less than the first distance, to reduce the influence of acoustic emission signal attenuation on signal acquisition.

[0087] In step S2, as an optional implementation, the sensor and the amplifier are placed separately. Specifically, the amplifier device with amplification function can be placed at a second distance away from the bearing to be tested according to actual conditions, and no specific limitation is made here.

[0088] Through the above-mentioned implementation mode of the present application, the distance between the position where the first area is located and the position where the bearing to be tested is located is set to be less than the first distance; placing the sensor close to the bearing can reduce the interference of the signal during the transmission process. Since the distance between the sensor and the bearing is short, the attenuation and noise interference during the signal transmission process will be relatively small, thereby improving the quality of the signal and the accuracy of monitoring. The distance between the position where the second area is located and the position where the bearing to be tested is set to be greater than the second distance. The processor and the amplifier will generate heat when working. If they are placed near the bearing, they may affect the operating temperature of the bearing, thereby affecting the accuracy of the monitoring results. Keeping them away from the bearing can reduce this effect, while reducing the power consumption of the overall system, and the weight of the sensor and amplifier may cause additional load on the bearing if they are directly installed on the bearing. Keeping them away from the bearing can reduce this effect, thereby avoiding potential damage to the bearing performance.

[0089] In an optional implementation, the received acoustic emission electrical signal is subjected to signal conversion processing to obtain an acoustic emission digital signal, including:

[0090] S1, the brain-like edge computing device receives the acoustic emission electrical signal, wherein an amplifier module is integrated on the brain-like edge computing device;

[0091] S2, the acoustic emission electrical signal is amplified by the amplifier module in the brain-like edge computing device;

[0092] S3, filtering and analog-to-digital conversion are performed on the amplified acoustic emission electrical signal to obtain an acoustic emission digital signal.

[0093] As an optional implementation, in the above steps S1-S3, the brain-like edge computing device receives the acoustic emission electrical signal, wherein an amplifier module is integrated on the brain-like edge computing device; the acoustic emission electrical signal is amplified by the amplifier module in the brain-like edge computing device; the amplified acoustic emission electrical signal is filtered and analog-to-digital converted to obtain an acoustic emission digital signal.

[0094] As an optional implementation, in step S1, the brain-like edge computing device first collects vibration signals of the bearing through sensors, and these signals contain operating status information of the bearing;

[0095] Then, the collected original vibration signal is preprocessed. It should be noted that steps S2 and S3 can be performed simultaneously, that is, filtering, denoising, normalization, and conversion into a signal format recognized by a computer are performed simultaneously during the amplification process to improve signal quality and reduce the complexity of subsequent processing. Then, the deep feature extraction neural network model can be used to extract fault feature information from the preprocessed signal. This step can be automatically completed through multi-layer convolution and long short-term memory networks to effectively extract deep time series features in bearing vibration signals.

[0096] Obviously, the embodiments described above are only some embodiments of the present application, not all embodiments. In order to better understand the above method for monitoring the stuck state of a bearing, the above process is described below in conjunction with an embodiment, but is not intended to limit the technical solution of the embodiments of the present application, specifically:

[0097] In an optional embodiment, the monitoring device is shown in the schematic diagram as follows: Figure 3 As shown:

[0098] The monitoring device mainly consists of an acoustic emission sensor 1 and a brain-like edge computing device 2. The anemometer bearing 3 is the device to be monitored. The acoustic emission sensor is used to collect acoustic emission signals during the operation of the anemometer and transmit the collected signals to the brain-like edge computing device. The device filters, amplifies, and A / D converts the electrical signals, converts the electrical signals into digital signals, and inputs them into the deep learning convolutional neural network.

[0099] The algorithm extracts signal features based on the trained model and establishes the relationship between the selected features and the anemometer bearing status. The algorithm performs corresponding processing. The edge computing device analyzes the collected data directly in the local device or network close to the data generation, without transmitting the data to the cloud data processing center.

[0100] Brain-like computing aims to mimic the way the human brain processes things. It is designed to simulate the human brain and has greater advantages in power consumption and learning capabilities than traditional chips. Traditionally, computer chips are designed in accordance with the von Neumann architecture, with storage and computing separated in space. Every time a computer performs a calculation, it needs to call back and forth between the CPU and memory. Frequent data exchange results in very low efficiency in processing massive amounts of information. The research on brain-like chips is based on the combination of microelectronics technology and new neuromorphic devices, hoping to break through traditional computing architectures, achieve deep integration of storage and computing, and significantly improve computing performance, increase integration, and reduce energy consumption. Deep learning convolutional neural networks based on brain-like chips can more efficiently extract and identify features from input data.

[0101] This approach, which uses neural networks and machine learning algorithms to process and analyze data, has potential advantages in processing complex and unstructured data. Brain-like edge computing attempts to implement these advanced algorithms on edge devices to improve decision-making speed and efficiency.

[0102] The present invention is described in a specific implementation manner as follows. Figure 4 As shown:

[0103] S402, fixing the acoustic emission sensor inside the anemometer through a coupling agent; specifically, near the bearing to reduce the influence of acoustic emission signal attenuation on signal acquisition.

[0104] S404, pre-collecting acoustic emission signals of a normal anemometer and an anemometer with a stuck bearing during operation, and inputting these signals into a brain-like edge computing device as training data, and the edge computing device performs training through a neural network algorithm;

[0105] Specifically, the model structure consists of an input layer, a convolution layer, an activation function, a pooling layer, and a fully connected layer. After the acoustic emission data is input into the model, the set convolution kernel slides on the data and extracts features. The ReLU activation function is used to perform a nonlinear transformation on the output of the convolution layer to enhance the model's expressiveness. The pooling layer reduces the dimensionality of the convolution layer output, reduces the amount of calculation, and improves the robustness and generalization ability of the model.

[0106] The fully connected layer maps the output of the pooling layer to the output of the model, and executes S406 to establish a mapping from the acoustic emission signal to the bearing state.

[0107] The acoustic emission sensor is installed in the anemometer to be monitored to collect the anemometer bearing data in real time and transmit it to the brain-like edge computing device. The device stores the transmitted data, and the neural network algorithm identifies and classifies faults based on the time-frequency domain statistical characteristics of the input bearing data to determine whether the anemometer is stuck.

[0108] Then, S408 is executed to determine that the anemometer has a bearing jam failure, and the edge computing device issues an early warning and sends an early warning message.

[0109] Through the method for monitoring the bearing jam status proposed in the above embodiment, long-term monitoring of the anemometer can be achieved; and the early jam characteristics of the bearing are relatively weak, the acoustic emission sensor is sensitive to defects, and can detect the early jam characteristics of the anemometer bearing, and the brain-like edge computing device can realize rapid data processing, which can be used to monitor the jam status of the anemometer bearing and send early warning information based on the results.

[0110] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods of each embodiment of the present application.

[0111] In this embodiment, a device for monitoring the stuck state of a bearing is also provided, and the device is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made are not repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0112] Figure 5 : is a structural block diagram of a device for monitoring a bearing stuck state according to an embodiment of the present application, the device comprising:

[0113] The signal processing unit 52 is used to perform signal conversion processing on the received acoustic emission electrical signal to obtain an acoustic emission digital signal, wherein the acoustic emission electrical signal is obtained by amplifying the original acoustic signal by the amplifier module installed in the second area, and the original acoustic signal is collected by the sensor module located in the first area in the acquisition device during the operation of the target bearing component to be tested, and the distance between the position of the first area and the position of the second area meets the preset distance condition;

[0114] The state prediction unit 54 is used to input at least one acoustic emission digital signal into the state prediction model to obtain the bearing state to be tested that matches the at least one acoustic emission digital signal; when at least one bearing state to be tested meets the warning condition, generate warning information matching the at least one bearing state to be tested.

[0115] Through the above device, the received acoustic emission electrical signal can be converted and processed to obtain an acoustic emission digital signal, wherein the acoustic emission electrical signal is obtained by amplifying the original acoustic signal by the amplifier module installed in the second area, and the original acoustic signal is collected by the sensor module located in the first area in the acquisition device during the operation of the target bearing component to be tested, and the distance between the position of the first area and the position of the second area meets the preset distance condition; at least one acoustic emission digital signal is input into the state prediction model to obtain the state of the bearing to be tested that matches each of the at least one acoustic emission digital signal; when at least one bearing state to be tested meets the warning condition, a warning message matching at least one bearing state to be tested is generated. This solves the problem in the related art that the anemometer cannot promptly detect the inaccurate wind measurement data caused by the bearing jam.

[0116] In an exemplary embodiment, the state prediction unit 54 is also used to determine the signal features that match the acoustic emission digital signal through the feature extraction layer of the state prediction model; perform nonlinear processing on the signal features according to the activation function to obtain the output result; and perform state mapping processing on the output result to obtain the bearing state to be tested that matches the acoustic emission digital signal.

[0117] In an exemplary embodiment, the state prediction unit 54 is also used to determine the signal features that match the acoustic emission digital signal through the feature extraction layer of the state prediction model, and then include: performing important feature extraction operations on the signal features to obtain local sound features that match the signal features, wherein the local sound features include vibration information of the acoustic emission digital signal; and determining the probability of matching each bearing state to be tested based on the local sound features.

[0118] In an exemplary embodiment, the state prediction unit 54 is also used to determine parameter state information matching the bearing to be tested based on the output results, wherein the parameter state information includes at least one of the following: signal time domain information, signal frequency domain information; and determine the bearing state of the bearing to be tested based on the parameter state information.

[0119] In an exemplary embodiment, the above-mentioned signal processing unit 52 is also used to perform statistical value processing on the signal time domain information when the parameter status information is signal time domain information, and determine the amplitude distribution information that matches the parameter status information; when the parameter status information is signal frequency domain information, perform spectrum processing on the signal frequency domain information, and determine the energy distribution information that matches the parameter status information.

[0120] In an exemplary embodiment, the above-mentioned device for monitoring the stuck state of the bearing is also used to set the distance between the position of the first area and the position of the bearing to be tested to be less than the first distance; and set the distance between the position of the second area and the position of the bearing to be tested to be greater than the second distance.

[0121] In an exemplary embodiment, the signal processing unit 52 is also used in a brain-like edge computing device to receive acoustic emission electrical signals, wherein an amplifier module is integrated on the brain-like edge computing device; the acoustic emission electrical signals are amplified by the amplifier module in the brain-like edge computing device; the amplified acoustic emission electrical signals are filtered and analog-to-digital converted to obtain acoustic emission digital signals.

[0122] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0123] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0124] S1, performing signal conversion processing on the received acoustic emission electrical signal to obtain an acoustic emission digital signal, wherein the acoustic emission electrical signal is obtained by amplifying the original acoustic signal by the amplifier module installed in the second area, and the original acoustic signal is collected by the sensor module located in the first area in the acquisition device during the operation of the target bearing component to be tested, and the distance between the position of the first area and the position of the second area meets the preset distance condition;

[0125] S2, inputting at least one acoustic emission digital signal into a state prediction model to obtain a bearing state to be measured that matches each of the at least one acoustic emission digital signal;

[0126] S3, when at least one bearing state to be measured satisfies the warning condition, generating warning information matching the at least one bearing state to be measured.

[0127] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0128] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0129] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0130] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0131] S1, performing signal conversion processing on the received acoustic emission electrical signal to obtain an acoustic emission digital signal, wherein the acoustic emission electrical signal is obtained by amplifying the original acoustic signal by the amplifier module installed in the second area, and the original acoustic signal is collected by the sensor module located in the first area in the acquisition device during the operation of the target bearing component to be tested, and the distance between the position of the first area and the position of the second area meets the preset distance condition;

[0132] S2, inputting at least one acoustic emission digital signal into a state prediction model to obtain a bearing state to be measured that matches each of the at least one acoustic emission digital signal;

[0133] S3, when at least one bearing state to be measured meets the warning condition, generate warning information matching the at least one bearing state to be measured.

[0134] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0135] An embodiment of the present application further provides a computer program product, comprising a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program product, and when the computer program is executed by a processor, the steps of the method described in each embodiment of the present application are implemented.

[0136] Optionally, in this embodiment, the above computer program may be configured to implement the following steps when executed by a processor:

[0137] S1, performing signal conversion processing on the received acoustic emission electrical signal to obtain an acoustic emission digital signal, wherein the acoustic emission electrical signal is obtained by amplifying the original acoustic signal by the amplifier module installed in the second area, and the original acoustic signal is collected by the sensor module located in the first area in the acquisition device during the operation of the target bearing component to be tested, and the distance between the position of the first area and the position of the second area meets the preset distance condition;

[0138] S2, inputting at least one acoustic emission digital signal into a state prediction model to obtain a bearing state to be measured that matches each of the at least one acoustic emission digital signal;

[0139] S3, when at least one bearing state to be measured satisfies the warning condition, generating warning information matching the at least one bearing state to be measured.

[0140] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0141] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0142] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for monitoring the stuck state of a bearing, characterized in that: include: The received acoustic emission electrical signal is subjected to signal conversion processing to obtain an acoustic emission digital signal, wherein the acoustic emission electrical signal is obtained by amplifying the original acoustic signal by the amplifier module installed in the second area, and the original acoustic signal is collected by the sensor module located in the first area of ​​the acquisition device during the operation of the target bearing component to be tested, and the distance between the position of the first area and the position of the second area meets the preset distance condition; Inputting at least one of the acoustic emission digital signals into a state prediction model to obtain a bearing state to be measured that matches each of the at least one acoustic emission digital signals; When at least one of the bearing states to be measured satisfies the warning condition, warning information matching at least one of the bearing states to be measured is generated.

2. The method according to claim 1, characterized in that The step of inputting at least one of the acoustic emission digital signals into a state prediction model to obtain a bearing state to be measured that matches each of the at least one acoustic emission digital signals comprises: Determine a signal feature matching the acoustic emission digital signal through a feature extraction layer of the state prediction model; Performing nonlinear processing on the signal feature according to the activation function to obtain an output result; The output result is subjected to state mapping processing to obtain the bearing state to be measured that matches the acoustic emission digital signal.

3. The method according to claim 2, characterized in that After determining the signal feature matching the acoustic emission digital signal through the feature extraction layer of the state prediction model, the method includes: Performing an important feature extraction operation on the signal feature to obtain a local sound feature that matches the signal feature, wherein the local sound feature includes vibration information of the acoustic emission digital signal; The probability of matching each bearing state to be measured is determined according to the local sound feature.

4. The method according to claim 2, characterized in that: The step of performing state mapping processing on the output result to obtain the bearing state to be tested that matches the acoustic emission digital signal further includes: Determine parameter state information matching the bearing to be tested according to the output result, wherein the parameter state information includes at least one of the following: signal time domain information, signal frequency domain information; The bearing state of the bearing to be tested is determined according to the parameter state information.

5. The method according to claim 4, characterized in that The step of determining the bearing state of the bearing to be tested according to the parameter state information further includes: In a case where the parameter state information is the signal time domain information, performing statistical value processing on the signal time domain information to determine amplitude distribution information matching the parameter state information; In the case where the parameter state information is the signal frequency domain information, spectrum processing is performed on the signal frequency domain information to determine energy distribution information matching the parameter state information.

6. The method according to claim 1, characterized in that Before the received acoustic emission electrical signal is converted into a signal to obtain an acoustic emission digital signal, the method further includes: Setting the distance between the position of the first area and the position of the bearing to be tested to be less than a first distance; It is set that the distance between the position where the second area is located and the position where the bearing to be tested is located is greater than the second distance.

7. The method according to claim 1, characterized in that The received acoustic emission electrical signal is subjected to signal conversion processing to obtain an acoustic emission digital signal, including: The brain-like edge computing device receives the acoustic emission electrical signal, wherein the amplifier module is integrated on the brain-like edge computing device; The acoustic emission electrical signal is amplified by the amplifier module in the brain-like edge computing device; The acoustic emission electrical signal after amplification is filtered and converted into analog-to-digital signals to obtain the acoustic emission digital signal.

8. A device for monitoring the stuck state of a bearing, characterized in that: include: A signal processing unit, used for performing signal conversion processing on the received acoustic emission electrical signal to obtain an acoustic emission digital signal, wherein the acoustic emission electrical signal is obtained by amplifying the original acoustic signal by the amplifier module installed in the second area, and the original acoustic signal is collected by the sensor module located in the first area of ​​the acquisition device during the operation of the target bearing component to be tested, and the distance between the position of the first area and the position of the second area meets the preset distance condition; The state prediction unit is used to input at least one of the acoustic emission digital signals into a state prediction model to obtain a bearing state to be tested that matches each of the at least one acoustic emission digital signals; when at least one of the bearing states to be tested meets the warning condition, generate warning information matching at least one of the bearing states to be tested.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method according to any one of claims 1 to 7 when executed.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images