Fault detection method and device for conveyor roller, medium and equipment
Through deep learning model and frequency domain analysis, the accuracy and efficiency of roller fault detection of belt conveyors are solved, efficient detection of roller faults is achieved, and operating costs and risks are reduced.
Patent Information
- Application Number
- CN202411036412.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-11-22
AI Technical Summary
The prior art is difficult to effectively monitor and prevent the failure of belt conveyor drums, especially drum blockage, which leads to increased friction, accelerated wear and potential fire risks, and relies on traditional manual detection methods, which increases costs and risks.
The deep learning model based on vibration acceleration signal data is adopted to judge faults through the LSTM network, and the frequency domain data is analyzed in combination with the short-time Fourier transform to determine the fault frequency threshold range to achieve accurate detection of conveyor roller faults.
It significantly improves the accuracy and efficiency of fault detection, reduces the dependence of manual detection, reduces maintenance costs, and achieves early abnormal detection and reliable operation of equipment.
Smart Images

Figure CN118992452A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of fault diagnosis, and in particular to a fault detection method, device, medium and equipment for conveyor rollers. Background Art
[0002] With the rapid improvement of science and technology and the continuous growth of production demand, belt conveyors are gradually developing in the direction of structured, fast and intelligent, so as to better adapt to complex working environments and ensure higher stability in long-term and long-distance transportation. As an important equipment in the production process, belt conveyors are used to continuously transport various materials. Its simple structure includes forward and reverse rollers, belt tensioning systems, and belt guide rollers as main components. The wide application highlights the importance of reducing operating costs and ensuring structural reliability. In conveyors, rollers are a vital component, and their technical condition directly affects the working efficiency of conveyors.
[0003] Ensuring the quality of the rollers used and performing effective condition monitoring are essential for the reliable operation of belt conveyors. Poor technical condition of the rollers can lead to mechanical failure of the belt, and roller clogging is a common problem. This clogging increases friction between the pipe and the belt, accelerates wear and can cause a fire risk on the conveyor route. Bearing damage is one of the main causes of roller clogging. In addition, aging bearings significantly increase the resistance to movement, which in turn increases energy consumption. In the actual operating environment, roller bearings are subject to a variety of factors that accelerate wear, including dynamic loads, humidity, operating temperature, vibration and dust.
[0004] Because roller damage usually has no obvious signs, it is crucial to preventively monitor the technical condition of the roller. When the conveyor is running, the main focus is on indicators such as the noise level, vibration and temperature fluctuations of the roller. Vibroacoustic signal analysis is one of the most commonly used diagnostic methods, including measuring the vibration level of the conveyor belt node, peak factor, signal envelope analysis, etc.
[0005] On the other hand, with the continuous advancement of technology, artificial intelligence has ushered in great development and has been applied across various fields, thereby improving production efficiency and benefits. However, despite this, belt conveyors still rely on traditional manual detection methods, and there are still limitations on the application of intelligent technology, which leads to increased costs and risks. Therefore, it has become an urgent task to integrate artificial intelligence technology into the production and transportation process of conveyors, and a new fault detection method based on conveyor rollers is needed. Summary of the invention
[0006] In view of the above problems, the present invention is proposed to provide a method, device, medium and equipment for fault detection of conveyor rollers that overcome the above problems or at least partially solve the above problems.
[0007] Other features and advantages of the present invention will become apparent from the following detailed description, or may be learned in part by practice of the present invention.
[0008] According to a first aspect of an embodiment of the present invention, a fault detection method for a conveyor roller is provided, the fault detection method for a conveyor roller comprising:
[0009] Collecting and acquiring status data of a conveyor roller, wherein the status data includes vibration acceleration signal data of the conveyor roller;
[0010] Inputting the vibration acceleration signal data into a trained fault judgment model for fault judgment processing to obtain a first fault judgment result;
[0011] Performing signal transformation based on the vibration acceleration signal data to obtain corresponding frequency domain data;
[0012] Determine a threshold range of the fault frequency, and perform analysis and processing based on the frequency domain data and the threshold range of the fault frequency to obtain a second fault judgment result;
[0013] The fault detection result of the conveyor roller is determined according to the first fault judgment result and the second fault judgment result.
[0014] In some embodiments of the present invention, the collecting and obtaining the status data of the conveyor roller includes:
[0015] An accelerometer is arranged on the conveyor roller, and a plurality of groups of first vibration acceleration signals when the conveyor roller stops operating are collected by the accelerometer;
[0016] Taking an average value of the plurality of groups of the first vibration acceleration signals collected as a calibration signal;
[0017] Collecting a second vibration acceleration signal of the conveyor roller during normal operation by means of the accelerometer;
[0018] Calibrate the second vibration acceleration signal based on the calibration signal to obtain a third vibration acceleration signal;
[0019] After eliminating point interference from the third vibration acceleration signal through a one-dimensional median filter, vibration acceleration signal data of the conveyor roller is obtained.
[0020] In some embodiments of the invention, the method comprises:
[0021] Collecting multiple groups of historical data of vibration acceleration signals of the conveyor rollers as training data;
[0022] An LSTM network is established and trained using the training data to obtain the fault judgment model.
[0023] In some embodiments of the present invention, the LSTM network is processed based on the training data through a gating mechanism to control the flow and storage of information of the training data in the LSTM network, wherein the LSTM network includes a forget gate for controlling the forgetting of the information, an input gate for controlling how the input information updates the cell state, and an output gate for controlling the flow of information from the cell state to the hidden state.
[0024] In some embodiments of the present invention, the method of performing signal transformation based on the vibration acceleration signal data is short-time Fourier transform.
[0025] In some embodiments of the present invention, the state data of the conveyor roller includes working condition data, and the threshold range for determining the fault frequency includes:
[0026] According to the vibration acceleration signal data corresponding to the different working condition data, the corresponding frequency domain data are respectively obtained;
[0027] The threshold range of the fault frequency is determined according to the frequency data corresponding to different operating condition data.
[0028] In some embodiments of the present invention, determining the fault detection result of the conveyor roller according to the first fault judgment result and the second fault judgment result includes:
[0029] When the first fault judgment result indicates that the conveyor roller has a fault, confirming whether the second fault judgment result indicates that the conveyor roller has a fault;
[0030] When the first fault judgment result and the second fault judgment result both indicate that the conveyor roller has a fault, the fault detection result indicates that the conveyor roller has a fault.
[0031] According to a second aspect of an embodiment of the present invention, there is provided a fault detection device for a conveyor roller, the fault detection device for a conveyor roller comprising:
[0032] A data acquisition module, used for acquiring status data of a conveyor roller, wherein the status data includes vibration acceleration signal data of the conveyor roller;
[0033] A first fault diagnosis module, used for inputting the vibration acceleration signal data into a trained fault judgment model for fault judgment processing to obtain a first fault judgment result;
[0034] A signal conversion module, used to perform signal conversion based on the vibration acceleration signal data to obtain corresponding frequency domain data;
[0035] A second fault diagnosis module, used to determine a threshold range of a fault frequency, and to analyze and process the frequency domain data and the threshold range of the fault frequency to obtain a second fault judgment result;
[0036] A fault determination module is used to determine a fault detection result of the conveyor roller according to the first fault judgment result and the second fault judgment result.
[0037] According to a third aspect of an embodiment of the present invention, a computer-readable storage medium is provided, in which computer program instructions are stored. The computer program instructions are loaded and executed by a processor to implement the operations performed by any of the methods described above.
[0038] According to a fourth aspect of an embodiment of the present invention, there is provided an electronic device, comprising a processor and a memory, wherein the memory stores computer program instructions that can be executed by the processor, and when the processor executes the computer program instructions, it implements instructions of any of the methods described above.
[0039] The technical solution provided in the embodiments of the present invention has at least the following technical effects or advantages:
[0040] An embodiment of the present invention provides a fault detection method, device, medium and equipment for a conveyor roller. Compared with the prior art, the fault detection method for a conveyor roller described in the embodiment of the present invention first performs preliminary fault diagnosis on the collected vibration acceleration signal data through a trained fault judgment model to obtain a first fault judgment result. After determining that a fault exists, further analysis and processing is performed based on the frequency domain data to obtain a second fault judgment result, thereby greatly improving the accuracy of fault detection through secondary diagnosis.
[0041] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0043] Figure 1 A schematic flow chart of a fault detection method for a conveyor roller provided in an embodiment of the present invention;
[0044] Figure 2 It is a reference schematic diagram of frequency domain images under different working conditions;
[0045] Figure 3 It is a reference diagram of average peak acceleration under different working conditions;
[0046] Figure 4 It is a reference schematic diagram of vibration signals under different working conditions;
[0047] Figure 5 A schematic flow chart of a fault detection device for a conveyor roller provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0048] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings.
[0049] Various structural schematic diagrams according to embodiments of the present disclosure are shown in the accompanying drawings. These figures are not drawn to scale, and some details are magnified and some details may be omitted for the purpose of clear expression. The shapes of various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are only exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may further design regions / layers with different shapes, sizes, and relative positions according to actual needs.
[0050] In the context of the present disclosure, when a layer / element is referred to as being "on" another layer / element, the layer / element may be directly on the other layer / element, or there may be an intervening layer / element therebetween. Additionally, if a layer / element is "on" another layer / element in one orientation, the layer / element may be "below" the other layer / element when the orientation is reversed. In the context of the present disclosure, similar or identical components may be denoted by the same or similar reference numerals.
[0051] In order to better understand the above-mentioned technical scheme, the above-mentioned technical scheme will be described in detail below in combination with specific implementation methods. It should be understood that the embodiments of the present disclosure and the specific features in the embodiments are detailed descriptions of the technical scheme of the present invention, rather than limitations on the technical scheme of the present invention. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.
[0052] Figure 1 is a flow chart of a fault detection method for a conveyor roller provided by an embodiment of the present invention, such as Figure 1 As shown, the fault detection method for conveyor rollers includes the following steps:
[0053] S1. Collecting and acquiring status data of a conveyor roller, wherein the status data includes vibration acceleration signal data of the conveyor roller;
[0054] In an embodiment of the present invention, the state data of the conveyor roller includes the vibration acceleration signal data and the working condition data of the conveyor roller, and the working condition data includes the tension force and the belt speed. Since the conveyor roller is actually running, it will be disturbed by the external environment, resulting in errors in the vibration acceleration signal data, which needs to be calibrated. Based on this, the state data of the conveyor roller collected and obtained in the embodiment of the present invention includes:
[0055] An accelerometer is arranged on the conveyor roller, and multiple groups of first vibration acceleration signals when the conveyor roller is stopped are collected by the accelerometer; an average value of the multiple groups of first vibration acceleration signals collected is taken as a calibration signal; a second vibration acceleration signal when the conveyor roller is normally operated is collected by the accelerometer; the second vibration acceleration signal is calibrated based on the calibration signal to obtain a third vibration acceleration signal; and the vibration acceleration signal data of the conveyor roller is obtained after eliminating point interference from the third vibration acceleration signal through a one-dimensional median filter.
[0056] After installing the accelerometer on the conveyor roller, its position is adjusted to more accurately obtain the vibration acceleration signal data. At the same time, the first vibration acceleration signal when the conveyor is stopped is measured multiple times, and its average value is calculated as a calibration signal to calibrate the second vibration acceleration signal collected when the conveyor roller is operating normally. Specifically, the calibration signal as the interference part is subtracted from the second vibration acceleration signal to obtain the third vibration acceleration signal as the actual acceleration signal. At the same time, in order to further improve the accuracy of the data, the third vibration acceleration signal is eliminated by a one-dimensional median filter to eliminate point interference, so as to obtain a pure vibration signal through noise reduction processing, while avoiding the problem of slope distortion and effectively reducing the error.
[0057] S2, inputting the vibration acceleration signal data into a trained fault judgment model for fault judgment processing to obtain a first fault judgment result;
[0058] The fault judgment model can be a trained model obtained directly. The embodiment of the present invention can also train and obtain the fault judgment model in the following manner, including: collecting multiple groups of historical data of vibration acceleration signals of the conveyor rollers as training data; establishing an LSTM network and obtaining the fault judgment model through training with the training data.
[0059] It should be noted that the training data consists of a training set, a validation set, and a test set. In the training data, the training set, the validation set, and the test set can be divided in a ratio of 8:1:1.
[0060] The LSTM (Long Short Term Memory) network processes the training data through a gating mechanism to control the flow and storage of information of the training data in the LSTM network, wherein the LSTM network includes a forget gate for controlling the forgetting of the information, an input gate for controlling how the input information updates the cell state, and an output gate for controlling the flow of information from the cell state to the hidden state.
[0061] Specifically, the forget gate processes information through the following formula:
[0062] f t =σ(W f ·[h t-1 ,x t ]+b f )
[0063] where f t is the output of the forget gate, W f is the weight matrix, h t-1 is the hidden state of the previous time step, x t is the input of the current time step, b f is the bias term, and σ is the sigmoid activation function.
[0064] The input gate processes information through the following formula:
[0065] i t =σ(W i ·[h t-1 ,x t ]+b i )
[0066]
[0067] where it is the output of the input gate, is the new candidate cell state processed by the tanh function, W i and W C is the corresponding weight matrix, b i and b C is the corresponding bias term.
[0068] The output gate processes information through the following formula:
[0069] o t =σ(W o ·[h t-1 ,x t ]+b o )
[0070] h t =o t tanh(C t )
[0071] o t is the output of the output gate, h t is the hidden state of the current time step, W o is the corresponding weight matrix, b o is the corresponding bias term.
[0072] Among them, updating the cell state follows the following formula:
[0073]
[0074] The LSTM network prevents the problems of gradient vanishing and gradient exploding through the synergistic effect of the forget gate, input gate and output gate based on the gating mechanism, captures long-term dependencies, controls the flow and storage of information in the network, effectively solves the shortcomings of traditional RNN in processing long sequence data, and greatly improves the performance of the model in time series data.
[0075] The vibration acceleration signal data is input into the trained fault judgment model, and the vibration acceleration signal data is divided into time series segments of fixed length, each segment is an input sample of the LSTM network, and then the corresponding label is output to represent the state of the conveyor belt in the time period (i.e., the first fault judgment result, indicating whether the conveyor roller is normal or faulty).
[0076] The parameters of the LSTM network can be changed according to the complexity of the data and the required accuracy level to adapt to actual application requirements in different application scenarios, and the embodiments of the present invention have no limitation to this.
[0077] For example, during the training process, the network parameters of the LSTM network, such as the learning rate, are adjusted according to the actual performance. A too high learning rate may cause the model to oscillate and fail to converge during the training process, and a too low learning rate may cause the training process to be slow. The learning rate decay, network search or random search, and adaptive learning rate algorithm are used for adjustment; the batch size is adjusted by testing different batch sizes or combining the optimizer with momentum; the number of iterations is adjusted by using an early stopping mechanism or by observing the training curve, and the model performance is optimized by the above methods. Alternatively, the accuracy and generalization ability of the LSTM network are evaluated on the validation set and the test set, and the fault diagnosis performance of the model is evaluated using indicators such as accuracy and recall rate. The data set used can also be a publicly available conveyor belt vibration data set on the Internet, which counts the vibration acceleration signal data of normal and faulty conveyor belts under different working conditions (such as tension and belt speed). According to the evaluation results of the fault diagnosis performance, the LSTM network is fine-tuned and optimized to improve the accuracy and robustness of fault diagnosis.
[0078] S3, performing signal transformation based on the vibration acceleration signal data to obtain corresponding frequency domain data;
[0079] In the embodiment of the present invention, the signal transformation method based on the vibration acceleration signal data is short-time Fourier transform.
[0080] Refer to the following formula:
[0081]
[0082] Where X(m,ω) represents the STFT result at time index m and frequency ω, x(t) represents the value of the original signal at time t, is the value of the window function at time t and window m, is a complex exponential term, and is used to represent the Fourier transform.
[0083] The embodiment of the present invention obtains corresponding frequency domain data after performing signal transformation based on the vibration acceleration signal data, and obtains a corresponding frequency domain image based on the frequency domain data.
[0084] S4, determining a threshold range of the fault frequency, and performing analysis and processing based on the frequency domain data and the threshold range of the fault frequency to obtain a second fault judgment result;
[0085] The embodiment of the present invention determines the threshold range of the fault frequency, including: acquiring corresponding frequency domain data respectively according to each vibration acceleration signal data corresponding to different working condition data; and determining the threshold range of the fault frequency according to the frequency data corresponding to different working condition data.
[0086] The embodiment of the present invention can compare the corresponding frequency domain images under different working condition data, and can intuitively observe the peak values of each frequency domain image to determine the threshold range of the fault frequency.
[0087] See also Figure 2 As shown in the figure, it is a reference diagram of frequency domain images under different working conditions. The fault vibration signals with tension force of 110N, belt speed of 500RPM, 600RPM and 700RPM and the normal vibration signals with tension force of 110N and belt speed of 600RPM are respectively taken for analysis. It can be determined that as the belt speed increases from 500RPM to 700RPM, the abnormal vibration frequency of the fault signal is also increasing, and the amplitude increases at the same time. The frequency increases from 18Hz to 25Hz, and the amplitude increases from 0.015m / s2 to 0.020m / s2. Any fault on the conveyor belt (such as imbalance, looseness, wear, etc.) may cause the vibration frequency of the conveyor belt to change. In this case, as the belt speed increases, the vibration frequency caused by the fault may increase because the excitation frequency of the faulty component becomes higher; at the same time, as the belt speed increases, the dynamic response of the entire system changes, and higher speeds may increase the vibration amplitude of the faulty component on the conveyor belt. Figure 2 It can be seen that when the tension force is 110N and the belt speed is 600PRPM, the fault frequency is 20Hz. Subsequently, a Butterworth bandpass filter is used to retain the vibration signal near the fault frequency and analyze the amplitude difference between the normal signal and the fault signal for subsequent analysis.
[0088] For example, in the embodiment of the present invention, the fault judgment model uses cross entropy as the loss function and uses the Adam optimizer to train it. Based on this model architecture, the prediction performance of the model is evaluated based on the four major model evaluation indicators of accuracy, precision, recall and F1 score according to the different feature dimensions of the input data.
[0089] By adjusting the weights of precision and recall and adjusting other factors, the final results of the model performance with different features added are shown in Table 1. When the data features used only include acceleration, the model's prediction accuracy is only 78.4%, compared with an accuracy of only 77.6%, a recall of only 75.6%, and an F1 score of only 0.766; after adding the feature of conveyor belt speed, the accuracy, precision, recall, and F1 score have all been significantly improved; but after adding only the feature of conveyor belt tension, the parameters of the model have only been slightly improved; the best model prediction effect is after taking all three features into consideration. In this case, the model's prediction accuracy reaches 87.5%, the precision reaches 87.3%, the recall reaches 87.2%, and the F1 score reaches 0.872.
[0090] Table 1 Performance of LSTM model with different features added
[0091]
[0092] At lower speeds, the frequency components associated with the fault tend to be more obvious and prominent. In contrast, at high speeds, the superposition of numerous frequency components may mask specific fault frequencies. Harmonics are reduced at low speeds, and the distinction between fault frequency components and harmonics is clearer, which helps with fault detection and analysis. High-speed operation brings additional background noise and vibration, which interferes with the identification of fault signals and complicates the diagnostic process. High-speed operation can cause mechanical heating, change the characteristics of mechanical components, and may affect the display of fault characteristics. Faults such as bearing defects, misaligned or loose gears will show characteristic frequencies with larger amplitude values at low speeds, so they are easier to identify in the FFT diagram. Therefore, in the subsequent feature analysis, the present invention selected the operating signal of the conveyor belt at a lower speed (from 400RPM to 1100RPM).
[0093] First, the average peak acceleration under different conveyor belt speeds and tensions is statistically calculated. Figure 3 As shown in the figure, the peak acceleration of the faulty conveyor belt exceeds the peak acceleration under normal conditions, and the peak acceleration gradually increases as the conveyor belt speed increases. In practical applications, the peak acceleration under normal operating conditions can be statistically analyzed in advance to set a healthy threshold (the exceeded value is the threshold range of the fault frequency). For example, when the tension is 70N and the belt speed is 400RPM, the threshold can be set to -0.09m / s2-0.09m / s2. During the fault diagnosis process, repeated exceeding of this threshold indicates the existence of a potential fault, which requires analysis in the frequency domain to further determine the fault, reduce misdiagnosis, and ensure the normal operation of the machine.
[0094] By performing Fourier transform analysis on the vibration acceleration signal data of all the collected vibration signals, the results are as follows: Figure 4 As shown in the figure, the fault vibration signals with a tension of 110N and belt speeds of 500RPM, 600RPM and 700RPM, and the normal vibration signal with a tension of 110N and belt speed of 600RPM are compared respectively. Figure 4 In the figure, from top to bottom are 600RPM, 500RPM, 700RPM, and 600RPM normal signals.
[0095] The analysis results show that as the belt speed increases from 500RPM to 700RPM, the abnormal vibration frequency and amplitude of the fault signal increase. The frequency increases from 18Hz to 25Hz, and the amplitude increases from 0.015m / s to 0.015m / s. 2 Increase to 0.020m / s 2This indicates that any fault on the conveyor belt may cause a change in the vibration frequency. Specifically, as the belt speed increases, the excitation frequency of the faulty component may increase, and the dynamic response of the entire system will also change, thereby increasing the vibration amplitude of the faulty component. Through image analysis, it was found that when the tension force was 110N and the belt speed was 600PRPM, the determined fault frequency was 20Hz.
[0096] S5. Determine the fault detection result of the conveyor roller according to the first fault judgment result and the second fault judgment result.
[0097] Regarding step S5, the embodiment of the present invention determines the fault detection result of the conveyor roller based on the first fault judgment result and the second fault judgment result, including: when the first fault judgment result is characterized as a fault in the conveyor roller, confirming whether the second fault judgment result is that the conveyor roller has a fault; when the first fault judgment result and the second fault judgment result are both characterized as a fault in the conveyor roller, the fault detection result is that the conveyor roller has a fault.
[0098] Compared with the prior art, the fault detection method for conveyor rollers described in the embodiment of the present invention first performs preliminary fault diagnosis on the collected vibration acceleration signal data through a trained fault judgment model to obtain a first fault judgment result. After determining that a fault exists, the second fault judgment result is obtained by further analyzing and processing the frequency domain data, thereby greatly improving the accuracy of fault detection through secondary diagnosis; by processing the collected vibration acceleration signal data through a deep learning network, the network is able to learn the characteristics of normal operation and fault status of the conveyor belt, thereby generating a detection model that performs well in practical applications, and then, time-frequency domain analysis and detection is performed on the signal that may have a fault to further determine whether the conveyor has a fault. Compared with traditional diagnostic methods, the embodiment of the present invention does not require shutdown for detection, and at the same time realizes early abnormality detection, reduces manual work during operation, improves diagnostic efficiency, and thus better maintains and manages equipment and reduces maintenance costs.
[0099] Based on the above embodiment, as the above Figure 1 The present invention provides an embodiment of a fault detection device for a conveyor roller, and the device embodiment is similar to Figure 1 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices, see Figure 5 As shown, the fault detection device for the conveyor roller includes:
[0100] The data acquisition module 100 is used to acquire the status data of the conveyor roller, wherein the status data includes the vibration acceleration signal data of the conveyor roller;
[0101] A first fault diagnosis module 200, used for inputting the vibration acceleration signal data into a trained fault judgment model for fault judgment processing to obtain a first fault judgment result;
[0102] A signal conversion module 300, used to perform signal conversion based on the vibration acceleration signal data to obtain corresponding frequency domain data;
[0103] A second fault diagnosis module 400 is used to determine a threshold range of a fault frequency, and to analyze and process the frequency domain data and the threshold range of the fault frequency to obtain a second fault judgment result;
[0104] The fault determination module 500 is used to determine the fault detection result of the conveyor roller according to the first fault judgment result and the second fault judgment result.
[0105] The fault detection device for conveyor rollers described in the embodiment of the present invention can execute the fault detection method for conveyor rollers provided in the above embodiment. The fault detection device for conveyor rollers has the corresponding functional steps and beneficial effects of the fault detection method for conveyor rollers described in the above embodiment. Please refer to the above embodiment of the fault detection method for conveyor rollers for details. The embodiment of the present invention will not be repeated here.
[0106] The embodiment of the present invention also provides an electronic device, which may include a processor and a memory, wherein the processor and the memory may be connected via a bus or other means. The processor may be a central processing unit (CPU). The processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components and other chips, or a combination of the above-mentioned various types of chips. The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs, non-transient computer executable programs and modules, such as the program instructions / modules corresponding to the fault detection method for conveyor rollers in the embodiment of the present invention. The processor executes various functional applications and data processing of the processor by running the non-transient software programs, instructions and modules stored in the memory, that is, realizing the fault detection method for conveyor rollers in the above method embodiment.
[0107] The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required by at least one function; the data storage area may store data created by the processor, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. The one or more modules are stored in the memory, and when executed by the processor, the fault detection method for the conveyor roller in the above-mentioned method embodiment is executed. The specific details of the above-mentioned electronic device can be understood by corresponding to the corresponding descriptions and effects in the above-mentioned method embodiment, and will not be repeated here. Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and the program, when executed, may include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a read-only memory (ROM), a random access memory (RAM), a flash memory (Flash Memory), a hard disk drive (HDD) or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above types of memory.
[0108] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.
[0109] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the intention that the claimed invention requires more features than those expressly recited in each claim. Rather, as reflected in the claims, inventive aspects lie in less than all of the features of the individual embodiments previously disclosed. Therefore, the claims that follow the detailed description are hereby expressly incorporated into the detailed description, with each claim itself serving as a separate embodiment of the present invention.
[0110] It should be noted that the above-mentioned embodiments illustrate rather than limit the invention and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims.
Claims
1. A method for detecting a fault of a conveyor roller, characterized in that: The fault detection method for a conveyor roller comprises: Collecting and acquiring status data of a conveyor roller, wherein the status data includes vibration acceleration signal data of the conveyor roller; Inputting the vibration acceleration signal data into a trained fault judgment model for fault judgment processing to obtain a first fault judgment result; Performing signal transformation based on the vibration acceleration signal data to obtain corresponding frequency domain data; Determine a threshold range of the fault frequency, and perform analysis and processing based on the frequency domain data and the threshold range of the fault frequency to obtain a second fault judgment result; The fault detection result of the conveyor roller is determined according to the first fault judgment result and the second fault judgment result.
2. The fault detection method for conveyor rollers according to claim 1, characterized in that: The acquisition of the conveyor roller status data comprises: An accelerometer is arranged on the conveyor roller, and a plurality of groups of first vibration acceleration signals when the conveyor roller stops operating are collected by the accelerometer; Taking an average value of the plurality of groups of the first vibration acceleration signals collected as a calibration signal; Collecting a second vibration acceleration signal of the conveyor roller during normal operation by means of the accelerometer; Calibrate the second vibration acceleration signal based on the calibration signal to obtain a third vibration acceleration signal; After eliminating point interference from the third vibration acceleration signal through a one-dimensional median filter, vibration acceleration signal data of the conveyor roller is obtained.
3. The fault detection method for conveyor rollers according to claim 1, characterized in that: The method comprises: Collecting multiple groups of historical data of vibration acceleration signals of the conveyor rollers as training data; An LSTM network is established and trained using the training data to obtain the fault judgment model.
4. The fault detection method for conveyor rollers according to claim 3, characterized in that: The LSTM network processes the training data through a gating mechanism to control the flow and storage of information of the training data in the LSTM network, wherein the LSTM network includes a forget gate for controlling the forgetting of the information, an input gate for controlling how the input information updates the cell state, and an output gate for controlling the flow of information from the cell state to the hidden state.
5. The fault detection method for conveyor roller according to claim 1, characterized in that: The method of performing signal transformation based on the vibration acceleration signal data is short-time Fourier transform.
6. The fault detection method for conveyor rollers according to claim 1, characterized in that: The state data of the conveyor roller includes working condition data, and the threshold range for determining the fault frequency includes: According to the vibration acceleration signal data corresponding to the different working condition data, the corresponding frequency domain data are respectively obtained; The threshold range of the fault frequency is determined according to the frequency data corresponding to different operating condition data.
7. The fault detection method for conveyor rollers according to claim 1, characterized in that: Determining the fault detection result of the conveyor roller according to the first fault judgment result and the second fault judgment result includes: When the first fault judgment result indicates that the conveyor roller has a fault, confirming whether the second fault judgment result indicates that the conveyor roller has a fault; When the first fault judgment result and the second fault judgment result both indicate that the conveyor roller has a fault, the fault detection result indicates that the conveyor roller has a fault.
8. A fault detection device for conveyor rollers, applied to the method according to any one of claims 1 to 7, characterized in that: The fault detection device for conveyor rollers comprises: A data acquisition module, used for acquiring status data of a conveyor roller, wherein the status data includes vibration acceleration signal data of the conveyor roller; A first fault diagnosis module, used for inputting the vibration acceleration signal data into a trained fault judgment model for fault judgment processing to obtain a first fault judgment result; A signal conversion module, used to perform signal conversion based on the vibration acceleration signal data to obtain corresponding frequency domain data; A second fault diagnosis module, used to determine a threshold range of a fault frequency, and to analyze and process the frequency domain data and the threshold range of the fault frequency to obtain a second fault judgment result; A fault determination module is used to determine a fault detection result of the conveyor roller according to the first fault judgment result and the second fault judgment result.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, which are loaded and executed by a processor to implement the operations performed by the method according to any one of claims 1 to 7.
10. An electronic device comprising a processor and a memory, characterized in that: The memory stores computer program instructions that can be executed by the processor, and when the processor executes the computer program instructions, the processor implements instructions of any one of the methods described in claims 1-7.
Citation Information
Patent Citations
Fault diagnosis method for roller assembly of belt conveyor
CN105976021A
Fault early warning method and fault early warning device for belt conveyor
CN115593880A
Fault identification method and system for conveyor belt roller of iron remover
CN115783643A
Training method and device for carrier roller fault detection model of belt conveyor
CN115973715A
Intelligent monitoring and adjusting system for carrier roller
CN116692388A
