Fault diagnosis methods and systems for belt conveyors
By working together with sensor modules, edge devices, and cloud devices, intelligent and timely fault diagnosis of belt conveyors has been achieved, solving the problems of resource consumption and long time required for manual diagnosis and improving mine production safety.
Patent Information
- Application Number
- CN202410288525.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-13
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2044-03-13
AI Technical Summary
In the existing technology, fault diagnosis of belt conveyors relies on manual diagnosis, which leads to serious consumption of human resources, long diagnosis time, difficulty in timely detection of faults, and affects mine production safety.
By using sensor modules to collect data, edge devices to build fault diagnosis models, and cloud devices to make decisions based on a pre-set database, fault diagnosis information is generated, thus achieving intelligent and timely fault diagnosis.
It improves the timeliness and accuracy of fault diagnosis for belt conveyors, covers the entire fault diagnosis process, reduces the computing pressure on cloud devices, and ensures mine production safety.
Smart Images

Figure CN118183216B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of fault diagnosis technology, and in particular to a fault diagnosis method and system for belt conveyors. Background Technology
[0002] As one of the key pieces of equipment ensuring normal production in mines, the stability of belt conveyors has a significant impact on mine safety. Therefore, it is necessary to monitor and diagnose the status of the numerous operating components of belt conveyors in real time, and to respond promptly to potential faults to ensure the normal operation of the belt conveyor.
[0003] In related technologies, fault diagnosis of belt conveyors relies solely on manual diagnosis, which results in significant consumption of human resources and excessively long fault diagnosis times. Furthermore, limitations in personnel's analytical abilities and work experience make it difficult to detect belt conveyor faults in a timely manner, leading to the operation of belt conveyors with defects, which is detrimental to mine production safety. Summary of the Invention
[0004] The purpose of this application is to propose a fault diagnosis method and system for belt conveyors, which can improve the timeliness of fault diagnosis and realize the intelligentization of the fault diagnosis process, thereby improving the production safety of mines.
[0005] To achieve the above objectives, the technical solution of this application is implemented as follows:
[0006] In a first aspect, embodiments of this application provide a fault diagnosis method for a belt conveyor, the method comprising:
[0007] Based on the data collected by the sensor module, the first dataset is constructed;
[0008] The edge device determines the corresponding fault diagnosis model based on the first dataset, and determines the health assessment result based on the fault diagnosis model;
[0009] The cloud-based equipment makes decisions based on health assessment results and a pre-set database to determine fault diagnosis information for the belt conveyor. The fault diagnosis information includes the cause of the fault and the fault repair plan.
[0010] Secondly, embodiments of this application provide a fault diagnosis system for a belt conveyor. The fault diagnosis system includes a sensor module, an edge device, and a cloud device, wherein:
[0011] The sensor module is used to collect data on the operating status of the belt conveyor and construct the first dataset;
[0012] Edge devices are used to determine the corresponding fault diagnosis model based on the first dataset, and to determine the health assessment result based on the fault diagnosis model;
[0013] The cloud-based device is used to make decisions based on health assessment results and a pre-set database to determine fault diagnosis information for the belt conveyor. The fault diagnosis information includes the cause of the fault and the fault repair plan.
[0014] Thirdly, embodiments of this application provide an edge device, which includes a first transceiver module and a first determination module, wherein:
[0015] The first transceiver module is used to receive the first dataset sent by the sensor module.
[0016] The first determining module is used to determine the corresponding fault diagnosis model based on the first dataset, determine the health assessment result based on the fault diagnosis model, and send the health assessment result to the cloud device.
[0017] Fourthly, embodiments of this application provide a cloud device, which includes a second transceiver module and a second determination module, wherein:
[0018] The second transceiver module is used to receive health assessment results sent by the edge device;
[0019] The second determining module is used to determine the fault diagnosis information of the belt conveyor based on the health assessment results and the preset database. The fault diagnosis information includes the cause of the fault and the fault repair plan.
[0020] Fifthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by at least one processor, implements the method described in the first aspect.
[0021] In a sixth aspect, embodiments of this application provide a computer program product, which includes a computer program or instructions that, when executed by at least one processor, implement the method described in the first aspect.
[0022] This application proposes a fault diagnosis method and system for belt conveyors. Based on data collected by a sensor module, a first dataset is constructed. An edge device determines a corresponding fault diagnosis model based on the first dataset and a health assessment result based on the fault diagnosis model. A cloud device makes decisions based on the health assessment result and a preset database to determine the fault diagnosis information of the belt conveyor, including the fault cause and fault repair plan. This collaborative work between the sensor module, edge device, and cloud device improves the timeliness of belt conveyor fault diagnosis. Furthermore, the collection of multi-sensor data by the sensor group and analysis based on the corresponding fault diagnosis model enhances data mining and multi-data analysis, leading to more accurate fault diagnosis. Moreover, the cloud device's decision-making based on the health assessment result and a preset database (e.g., a rule base formed by expert experience and a case base formed by historical maintenance records) covers the entire fault diagnosis process, including fault identification, fault cause analysis, and fault repair plan generation. This enables intelligent fault diagnosis throughout the entire process, thereby improving mine production safety. Attached Figure Description
[0023] Figure 1 This is a schematic diagram of the composition structure of a fault diagnosis system provided in an embodiment of this application;
[0024] Figure 2 A flowchart illustrating a fault diagnosis method provided in this application embodiment. Figure 1 ;
[0025] Figure 3 A flowchart illustrating a fault diagnosis method provided in this application embodiment. Figure 2 ;
[0026] Figure 4 A schematic diagram of a fault diagnosis process based on the YOLOv5 algorithm provided for an embodiment of this application;
[0027] Figure 5 A flowchart illustrating a fault diagnosis method provided in this application embodiment. Figure 3 ;
[0028] Figure 6 A schematic diagram illustrating the composition structure of a rule base provided in an embodiment of this application;
[0029] Figure 7 This is a schematic diagram illustrating the composition structure of a case library provided in an embodiment of this application;
[0030] Figure 8 A flowchart illustrating a fault diagnosis method provided in this application embodiment. Figure 4 ;
[0031] Figure 9 A flowchart illustrating a fault diagnosis method provided in this application embodiment. Figure 5 ;
[0032] Figure 10 A detailed flowchart illustrating a fault diagnosis method provided in an embodiment of this application;
[0033] Figure 11 A schematic diagram of the application framework for a fault diagnosis method provided in the embodiments of this application. Figure 1 ;
[0034] Figure 12 A schematic diagram of the application framework for a fault diagnosis method provided in the embodiments of this application. Figure 2 ;
[0035] Figure 13 This is a schematic diagram of the composition structure of an edge device provided in an embodiment of this application;
[0036] Figure 14 This is a schematic diagram of the specific hardware structure of an edge device provided in an embodiment of this application;
[0037] Figure 15 This is a schematic diagram of the composition structure of a cloud device provided in an embodiment of this application;
[0038] Figure 16 This is a schematic diagram of the specific hardware structure of a cloud device provided in an embodiment of this application. Detailed Implementation
[0039] In order to gain a more detailed understanding of the features and technical content of the embodiments of this application, the implementation of the embodiments of this application will be described in detail below with reference to the accompanying drawings. The accompanying drawings are for reference and illustration only and are not intended to limit the embodiments of this application.
[0040] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.
[0041] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.
[0042] It should also be noted that the terms "first, second, and third" used in the embodiments of this application are only used to distinguish similar objects and do not represent a specific order of objects. It is understood that "first, second, and third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.
[0043] It is understandable that belt conveyors offer advantages such as long transport distances, large capacity, low power consumption, and minimal maintenance workload in the continuous loading and transportation of materials, making them one of the crucial pieces of equipment for ensuring normal mine production. However, belt conveyors operate in harsh, complex, and dangerous environments, and the maintenance workload in case of malfunctions is substantial. Therefore, it is necessary to monitor and diagnose the status of numerous operating components throughout the belt conveyor in real time, responding promptly to potential faults to ensure the normal operation of the belt conveyor.
[0044] As one of the most important pieces of equipment in mine production, the stability of belt conveyors has a significant impact on mine safety. Currently, fault diagnosis of belt conveyors relies solely on manual methods, which leads to problems such as high consumption of human resources and excessively long fault diagnosis times. In addition, limitations in personnel's analytical abilities and work experience make it difficult to detect belt conveyor faults in a timely manner, resulting in belt conveyors operating "with defects," which is detrimental to mine safety.
[0045] While related technologies have provided some fault diagnosis methods, their effectiveness is not ideal and still has some shortcomings. For example, the fault diagnosis methods are relatively simple, and the data mining and utilization are not thorough enough, resulting in an incomplete fault diagnosis. Some methods use a bus-based communication method, with monitoring substations connected in series via a bus. This results in a large amount of useless data occupying the lines, wasting storage resources, making it difficult to extract effective data, and causing slow data transmission with poor accuracy and reliability. The number of monitoring substations in the monitoring system is also limited, and the system's functional scalability is poor. Even with fiber optic communication, the objective reality of long transmission distances remains, and transmitting large amounts of data remains difficult.
[0046] Based on this, this application provides a fault diagnosis method for belt conveyors, specifically a cloud-edge-device collaborative intelligent fault diagnosis method for the entire belt conveyor process. In this application embodiment, sensor modules collect data on the operating status of numerous components in the belt conveyor, improving data mining and making fault diagnosis information more accurate, achieving comprehensive and three-dimensional fault diagnosis of the belt conveyor. Furthermore, by analyzing health assessment results through cloud devices and a preset database, fault causes and fault repair plans are generated, covering the entire fault diagnosis process and achieving intelligent fault diagnosis throughout. In addition, through the collaboration of cloud, edge, and device, the collaborative process of fault diagnosis is differentiated, achieving lightweight data transmission, reducing the computational pressure on cloud devices, ensuring the timeliness of data processing, improving the timeliness of belt conveyor fault diagnosis, and thus improving mine production safety.
[0047] The various embodiments of this application will now be described in detail with reference to the accompanying drawings.
[0048] This application provides a fault diagnosis method, which is applied to a fault diagnosis system. Figure 1 This is a schematic diagram illustrating the structural composition of a fault diagnosis system provided in an embodiment of this application. Figure 1 As shown, the fault diagnosis system 10 may include a sensor module 101, an edge device 102, and a cloud device 103. The transmitter module 101, the edge device 102, and the cloud device 103 are connected in communication.
[0049] In this embodiment, edge device 102 can be referred to as an edge node; cloud device 103 can be a cloud server or a cloud platform. Additionally, the preset database on the cloud device can be referred to as a cloud database or cloud-based database.
[0050] In this embodiment, the sensor module 101 first acquires data collected by various sensors, inspection robots, etc., and constructs a first dataset based on the collected data. Then, the edge device 102 processes the first dataset to obtain health assessment results. Finally, the cloud device 103 analyzes the health assessment results based on the cloud database and a preset database to obtain the cause of the fault and the fault repair plan, thereby realizing the entire process of fault diagnosis for the belt conveyor.
[0051] based on Figure 1 The fault diagnosis system 10 shown is... Figure 2 A flowchart illustrating a fault diagnosis method provided in this application embodiment. Figure 1 .like Figure 2 As shown, the method may include:
[0052] S201: Construct the first dataset based on the data collected by the sensor module.
[0053] In this embodiment, sensors are installed on numerous components of the belt conveyor, and then the sensor module collects the operating data of these components to construct a first dataset. This first dataset can be referred to as the first multi-sensor dataset.
[0054] For example, high-speed cameras and tension sensors are installed on the belt conveyor; industrial cameras and infrared thermal imagers are installed on the rollers; noise and vibration sensors are installed on the idlers and reducers; and current and voltage sensors are installed on the motors. By installing sensors on numerous components of the belt conveyor, faults such as conveyor belt tearing, slippage, roller stalling, idler jamming, abnormal reducer vibration, and motor current imbalance can be monitored. Additionally, inspection robots can be used to monitor the operating status of the belt conveyor. For instance, an inspection robot can inspect the belt conveyor along a predetermined path, using cameras, robotic arms, microphone arrays, and other supporting equipment, along with sensors, to monitor the belt conveyor's operating status.
[0055] In this embodiment of the application, the data collected by the sensor module includes at least one of the following: data collected by sensors installed on the conveyor belt; data collected by sensors installed on the roller; data collected by sensors installed on the idler roller; data collected by sensors installed on the reducer; data collected by sensors installed on the motor; and data collected by sensors installed on the inspection robot.
[0056] In other words, the sensor module collects data reflecting the operating conditions of the belt conveyor transmitted between various sensors, including real-time operating data, normal operating data, abnormal operating data, and various interference information, constructs a first dataset, and sends the first dataset to the edge device so that the edge device can determine the health assessment result based on the first dataset.
[0057] S202: The edge device determines the corresponding fault diagnosis model based on the first dataset, and determines the health assessment result based on the fault diagnosis model.
[0058] In this embodiment, the edge device determines a fault diagnosis model by receiving a first dataset sent by the sensor module, thereby determining a health assessment result based on the fault diagnosis model, and then sends the health assessment result to the cloud device so that the cloud device can make a decision based on the health assessment result and determine the fault diagnosis information of the belt conveyor.
[0059] In some embodiments, the edge device determines the corresponding fault diagnosis model based on the first dataset, which may include: preprocessing the first dataset to determine the second dataset; and determining the corresponding fault diagnosis model based on the second dataset.
[0060] In this embodiment of the application, preprocessing may include redundancy reduction, noise reduction, and other processes to filter out various types of interference information.
[0061] In other words, after receiving the first dataset, the edge device needs to perform redundancy reduction and noise reduction on it. The first dataset can include real-time running data, normal running data, abnormal running data, and various interference information, and the interference information is the target of redundancy reduction and noise reduction.
[0062] In one specific embodiment, the edge device may include a data processing center, which performs redundancy reduction, noise reduction, and filtering of various interference information on the first dataset to obtain a second dataset.
[0063] In this embodiment, the second dataset can be referred to as the second multi-sensor dataset. Then, the normal operation data and abnormal operation data of the second dataset are used as the sample feature set of the fault diagnosis model. The real-time operation data is compared with the sample feature set to determine the health assessment result.
[0064] In addition, in this embodiment of the application, the health assessment result can be referred to as the health status level assessment result. Here, the health assessment result can be the fault type or the lifespan prediction, and there is no specific limitation on the health assessment result.
[0065] S203: The cloud-based device makes decisions based on the health assessment results and the preset database to determine the fault diagnosis information of the belt conveyor. The fault diagnosis information includes the cause of the fault and the fault repair plan.
[0066] In this embodiment, the cloud device makes decisions based on health assessment results and a preset database, generates fault diagnosis information for the belt conveyor, and performs maintenance on the belt conveyor in its current state based on the fault diagnosis information. This fault diagnosis information includes cause analysis of the current belt conveyor fault and intelligent generation of fault causes and repair plans.
[0067] In this embodiment of the application, the cloud device can be deployed on a central server. After the fault diagnosis model determines the health assessment results, the fault-related data can also be transmitted to the central server, and then the central server makes a decision in conjunction with a preset database.
[0068] In this embodiment of the application, after obtaining fault diagnosis information, the cloud device can send the fault diagnosis information to the user device so that the user can view the current fault cause and fault repair plan, enabling the user to promptly eliminate the fault.
[0069] This application provides a fault diagnosis method. Based on data collected by a sensor module, a first dataset is constructed. An edge device determines a corresponding fault diagnosis model based on the first dataset and a health assessment result based on the fault diagnosis model. A cloud device makes decisions based on the health assessment result and a preset database to determine the fault diagnosis information of the belt conveyor. The fault diagnosis information includes the cause of the fault and a fault repair plan. Thus, through the collaborative work of the sensor module, edge device, and cloud device, the timeliness of belt conveyor fault diagnosis can be improved. Furthermore, by collecting multiple sensing data from the sensor group and analyzing it according to the corresponding fault diagnosis model, data mining and analysis of multiple data sources can be improved, making fault diagnosis more accurate. Moreover, the cloud device's decision-making based on the health assessment result and a preset database (such as a rule base formed by expert experience and a case database formed by historical maintenance records) can cover the entire fault diagnosis process, thereby achieving intelligent fault diagnosis and improving mine production safety.
[0070] In another embodiment of this application, based on the fault diagnosis method described in the foregoing embodiments, Figure 3 A flowchart illustrating a fault diagnosis method provided in this application embodiment. Figure 2 .like Figure 3 As shown, the method may include:
[0071] S301: Determine the data type corresponding to the second dataset.
[0072] In this embodiment, the fault diagnosis model is determined by identifying the data type corresponding to the real-time running data in the second dataset. Here, the data type may include a first type, a second type, current data, etc.; wherein, the first type may be image data, and the second type may be sound data.
[0073] S302: If the data type is the first type, then the fault diagnosis model is determined to be the first algorithm model.
[0074] S303: Classify faults based on the second dataset and the first algorithm model to determine the health assessment results.
[0075] In this embodiment, if the data type is a first type, the fault diagnosis model is a first algorithm model. Fault classification is performed using a second dataset and the first algorithm model to determine the health assessment result. Here, the first algorithm model can be the YOLOv5 algorithm model for processing image data.
[0076] For example, roller operation data is acquired, which corresponds to image data, and the YOLOv5 algorithm model is selected. This YOLOv5 algorithm model is mainly used for image data to identify faults such as roller weld defects, rubber wear, and rubber detachment.
[0077] In some embodiments, for the YOLOv5 algorithm model, such as Figure 4 As shown, the process of establishing the YOLOv5 algorithm includes the following steps:
[0078] S401: Obtain the image of the roller.
[0079] S402: Divide the dataset into training set, validation set, and test set.
[0080] S403: Load the YOLOv5 algorithm for pre-training.
[0081] S404: Modify parameters such as iteration count, batch size, freeze training, and thaw training.
[0082] S405: Train the YOLOv5 algorithm using the training set, and use the validation set to initially evaluate the training effect.
[0083] S406: Use the test set to test the trained YOLOv5 algorithm to detect faults in the roller.
[0084] S407: Requirements met?
[0085] S408: Adjust algorithm parameters based on test results.
[0086] S409: Detect roller faults using the YOLOv5 algorithm that meets the requirements.
[0087] In this embodiment, images of the drum are first acquired using an industrial camera. The obtained dataset is then used to train the YOLOv5 algorithm. Before training begins, the training, validation, and test sets need to be divided into appropriate proportions. Then, to achieve better training results, the YOLOv5 algorithm is pre-trained. After modifying relevant parameters, training begins. The loss functions of the training and validation sets change with the number of iterations. When the loss values of the training and validation sets overlap and tend to stabilize, it indicates that the loss function has converged, and the YOLOv5 algorithm shows good training performance. Finally, the model's performance is evaluated using the test set. If it meets the actual requirements, the YOLOv5 algorithm is deployed on an embedded computing module close to the drum; otherwise, the algorithm parameters should be modified and retrained.
[0088] S304: If the data type is the second type, then the fault diagnosis model is determined to be the second algorithm model.
[0089] S305: Classify faults based on the second dataset and the second algorithm model to determine the health assessment results.
[0090] In this embodiment, if the data type is a second type, the fault diagnosis model is a second algorithm model. Fault classification is performed using the second dataset and the second algorithm model to determine the health assessment result. Here, the second algorithm model can be a one-dimensional convolutional neural network (1DCNN)-SVM algorithm model for processing sound data.
[0091] In this embodiment, it is assumed that the fault diagnosis algorithm for diagnosing idler roller bearing faults is a 1DCNN-SVM algorithm. First, the acquired vibration signal is subjected to Fourier transform to highlight its fault characteristics. Then, utilizing the powerful feature extraction capability of 1DCNN, relevant feature vectors are extracted from the transformed signal. SVM classifies the samples based on the extracted features to determine the health assessment result.
[0092] In this embodiment of the application, if the data in the second dataset includes a first type of data and a second type of data, then the fault diagnosis model is determined to be a first algorithm model and a second algorithm model, and the different data in the second dataset are processed respectively to obtain the health assessment result.
[0093] It is also important to note that if the data type in the second dataset is current data, then the fault diagnosis model can be an algorithm model specifically for current fault detection. In other words, the edge device can include an embedded computing module containing various fault diagnosis models, such as the YOLOv5 algorithm for processing image data, the 1DCNN-SVM algorithm for processing sound data, and algorithms for processing current data, etc.
[0094] This application provides a fault diagnosis method that collects multiple sensing data through a sensor array and analyzes it according to the corresponding fault diagnosis model. This improves data mining and analysis of multiple data sources, making fault diagnosis more accurate and achieving comprehensive three-dimensional fault diagnosis of belt conveyors. In addition, by coordinating cloud, edge, and terminal, the collaborative process of fault diagnosis is differentiated, achieving lightweight data transmission, reducing the computing pressure on cloud devices, ensuring the timeliness of data processing, improving the timeliness of belt conveyor fault diagnosis, and thus improving the production safety of mines.
[0095] In another embodiment of this application, based on the fault diagnosis method described in the foregoing embodiments, Figure 5 A flowchart illustrating a fault diagnosis method provided in this application embodiment. Figure 3 .like Figure 5 As shown, the method may include:
[0096] S501: Based on the health assessment results and rule base, perform rule-based reasoning analysis to determine the cause of belt conveyor failure.
[0097] In this embodiment of the application, the preset database may include a rule base and a case base. The rule base may be based on the experience of domain experts, and the case base may be based on historical maintenance records.
[0098] In some embodiments, the construction of the rule base may include: acquiring domain expert experience regarding the belt conveyor; and constructing the rule base based on the domain expert experience.
[0099] In some embodiments, the construction of the case library may include: obtaining historical maintenance records of the belt conveyor; and constructing the case library based on the historical maintenance records.
[0100] In other words, in this embodiment of the application, by acquiring the experience of experts in the field of belt conveyors, a rule base with a hierarchical structure is constructed, such as... Figure 6 As shown, the rule base includes a conclusion table, a threshold table, and a fault tree table; then, based on expert experience, the belt conveyor faults can be analyzed layer by layer from fault type to fault cause, generating inference rules from fault type to fault cause.
[0101] In this embodiment of the application, historical maintenance records of the belt conveyor are obtained to construct a case library with a large number of maintenance records; wherein, the case library is as follows: Figure 7 As shown. The case library's knowledge comes from maintenance records. For example, by extracting keywords, maintenance records are represented by nine attributes: case name, case time, case location, fault area, accident phenomenon, direct loss, case analysis, solution, and fault parameters. In other words, the case library can be composed of case tables, each corresponding to a fault case of a different component of the belt conveyor.
[0102] In other words, in this embodiment of the application, the rule base and case base include faults of various components of the belt conveyor, such as belt misalignment, tearing, slippage, roller bearing jamming, roller coating detachment, etc.
[0103] In this application embodiment, the determination of the cause of the fault is first described. In some embodiments, see Figure 8 The method may include:
[0104] S801: Generate a fault tree based on the rule base.
[0105] S802: Analyze the health assessment results layer by layer according to the fault tree to determine the probability of occurrence of at least one candidate fault cause.
[0106] In this application, knowledge from the rule base is acquired to generate a fault tree. Then, based on rule-based reasoning (RBR), the fault tree is quantitatively analyzed to determine the minimum cut set and its importance, that is, to determine at least one candidate fault cause and its importance.
[0107] S803: Determine the cause of failure based on the probability of occurrence of at least one candidate cause of failure.
[0108] In this embodiment, the relative probability importance of at least one candidate fault cause is determined based on the occurrence probability of at least one candidate fault cause. Here, the health assessment result is taken as the base event, and the candidate fault causes are taken as the top events. First, the occurrence probability of the top event in the fault tree is calculated based on the occurrence probability of the base events:
[0109]
[0110] Secondly, the importance of a minimum cut set to the top event is obtained from the probability of the top event. The method for calculating the importance of the relative probability of a minimum cut set is as follows:
[0111]
[0112]
[0113] Where T is the top event of the fault tree, Q j For the minimum cut set, q i The probability of the base event occurring.
[0114] In this embodiment, the fault cause is determined by assessing the relative probability importance of at least one candidate fault cause. Here, the candidate fault cause with the highest relative probability importance value can be used as the fault cause, or the candidate fault cause with the highest probability of occurrence can be used as the fault cause.
[0115] S502: Based on the health assessment results and case library, conduct case-based reasoning analysis to determine the fault maintenance plan for the belt conveyor.
[0116] In this application, the real-time operating data of the belt conveyor is determined based on the health assessment results. Then, based on case-based reasoning (CBR), the current belt conveyor fault is matched with past cases in the case library according to the real-time operating data, and a fault repair plan is given.
[0117] In this application embodiment, the determination of the fault repair plan is specifically described. In some embodiments, see Figure 9 The method may include:
[0118] S901: Based on the health assessment results, determine the characteristic value of at least one characteristic attribute of the target failure case.
[0119] In this embodiment, real-time operating data of the belt conveyor is obtained based on health assessment results and used as a target failure case. At least one characteristic attribute is determined using the real-time operating data from the target failure case, and the feature value of this attribute is obtained through program recognition or manual input. Here, the characteristic attribute may include temperature, speed, current, voltage, etc. Furthermore, the feature value can be a text-based feature value or a numerical feature value.
[0120] S902: Based on the feature value of at least one feature attribute, perform similarity matching between the target failure case and the case library to determine the similarity between at least one candidate historical case and the target failure case.
[0121] In this embodiment of the application, case matching is performed on the target failure case, and similarity matching is performed according to CBR to determine the similarity between at least one candidate historical case and the target failure case.
[0122] For example, taking a historical candidate case where the feature values of the historical candidate case include both textual and numerical types, the method for calculating the similarity of the numerical feature attributes is as follows:
[0123]
[0124] The method for calculating the similarity of textual feature attributes is as follows:
[0125]
[0126] The similarity between candidate historical cases and target fault cases is determined based on numerical and textual feature attribute similarity. The calculation method is as follows:
[0127]
[0128] Where X represents a historical case, X i Y is the value of the i-th attribute; Y is the target case. i The value of the i-th attribute; max i The maximum value of attribute i; min i ω represents the minimum value of attribute i; n represents the number of feature attributes in the case; ω i Let be the index weight of the i-th attribute.
[0129] S903: Based on similarity, determine the target historical case from at least one candidate historical case, and identify the target historical case as the fault repair plan.
[0130] In the embodiments of this application, determining a target historical case from at least one candidate historical case based on similarity may include: determining one or more candidate historical cases with the highest similarity as the target historical case; or, determining one or more candidate historical cases with a similarity greater than a preset similarity threshold as the target historical case.
[0131] In other words, by selecting one or N corresponding candidate historical cases with the highest similarity, and outputting their case analysis and solutions as a fault repair plan, where N > 1 and is an integer. Alternatively, by setting a preset similarity threshold, one or more candidate historical cases with similarity greater than the preset similarity threshold are selected to meet the requirements, and their case analysis and solutions are output as a fault repair plan.
[0132] This application provides a fault diagnosis method. Based on health assessment results and a rule base, rule-based reasoning analysis is performed to determine the cause of a belt conveyor failure. Based on health assessment results and a case base, case-based reasoning analysis is performed to determine a fault repair plan for the belt conveyor. In this way, by analyzing health assessment results through cloud devices and a pre-set database to generate fault causes and fault repair plans, the entire fault diagnosis process can be covered, achieving intelligent fault diagnosis. Furthermore, through the collaboration of cloud, edge, and terminal, the collaborative process of fault diagnosis is differentiated, achieving lightweight data transmission, reducing the computational burden on cloud devices, ensuring the timeliness of data processing, improving the timeliness of belt conveyor fault diagnosis, and ultimately improving mine production safety.
[0133] In another embodiment of this application, the fault diagnosis method based on the foregoing embodiments, Figure 10 This is a detailed flowchart illustrating a fault diagnosis method provided in an embodiment of this application. The fault diagnosis is applied to a fault diagnosis system. This fault diagnosis system may include a sensor module, an edge device, and a cloud device, wherein the sensor module, edge device, and cloud device can communicate via 5G. Figure 10 As shown, the detailed process may include:
[0134] S1001: The sensor module constructs the first dataset based on the collected data.
[0135] S1002: The sensor module sends the first dataset to the edge device.
[0136] S1003: The edge device preprocesses the first dataset to determine the second dataset.
[0137] S1004: The edge device determines the corresponding fault diagnosis model based on the second dataset, and determines the health assessment result based on the fault diagnosis model.
[0138] S1005: The edge device sends the health assessment results to the cloud device.
[0139] S1006: The cloud-based device performs rule-based reasoning analysis based on the health assessment results and rule base to determine the cause of the belt conveyor failure.
[0140] S1007: The cloud-based device performs case-based reasoning analysis based on the health assessment results and case library to determine the fault repair plan for the belt conveyor.
[0141] Figure 11 A schematic diagram of the application architecture of a fault diagnosis method provided in this application embodiment. Figure 1 .like Figure 11 As shown, the application framework includes cloud device a1, edge device a2, and sensor module a3. Here, cloud device a1 can be a cloud database, edge device a2 can be an edge node, and sensor module a3 is a multi-sensing terminal.
[0142] like Figure 11 As shown, sensor module a3 installs sensors on various components of the belt conveyor; the conveyor belt, rollers, idlers, reducers, and motors are installed on the belt conveyor. Sensor module a3 then acquires data from each sensor to establish a first dataset. Specifically, a high-speed camera and tension sensor are installed on the conveyor belt, an industrial camera and infrared thermal imager are installed on the rollers, noise and vibration sensors are installed on the idlers and reducers, and current and voltage sensors are installed on the motor to monitor faults such as conveyor belt tearing, slippage, roller stalling, idler jamming, abnormal reducer vibration, and motor current imbalance. Then, sensor module a3 sends the first dataset to edge device a2 via 5G communication. Edge device a2 determines the health assessment results based on the first dataset. Specifically, it performs redundancy and noise reduction on the first dataset through a mining explosion-proof and intrinsically safe data processing center (i.e., the data processing center) to obtain a second dataset. Then, it uses fault diagnosis algorithms such as YOLOv5 and 1DCNN-SVM to calculate and classify the second dataset to determine the health assessment results. Finally, edge device a2 sends the health assessment results to cloud device a1 via 5G communication. Cloud device a1 uses a rule base and a case base to make decisions based on the health assessment results, obtaining the current fault category, fault cause, and fault maintenance plan for the belt conveyor. Figure 11 The fault diagnosis method of this application is described in detail below:
[0143] Step 1: Perform fusion sensing on multiple sensors to quickly extract the operating data of the belt conveyor and construct the first multi-sensor dataset (i.e., the first dataset);
[0144] In this embodiment, the multiple sensors may include various sensors and an inspection robot, used to acquire the operating data of the belt conveyor for analysis by intelligent diagnostic algorithms. Specifically, high-speed cameras and tension sensors are installed on the conveyor belt, industrial cameras and infrared thermal imagers are installed on the rollers, noise and vibration sensors are installed on the idlers and reducers, and current and voltage sensors are installed on the motors to monitor faults such as conveyor belt tearing, slippage, roller stalling, idler jamming, abnormal reducer vibration, and motor current imbalance. The inspection robot inspects the belt conveyor according to a prescribed path, using cameras, robotic arms, microphone arrays, and other supporting equipment, in conjunction with sensors, to monitor the operating status of the belt conveyor.
[0145] In this embodiment of the application, the first multi-sensor dataset extracts data reflecting the operating conditions of the belt conveyor transmitted between each subsystem, including real-time operating data, normal operating data, abnormal operating data, and various interference information.
[0146] Step 2: The edge nodes match the corresponding fault diagnosis model based on the first multi-sensor dataset and calculate the health assessment results;
[0147] In this application, the health assessment results may include fault type, lifespan prediction, etc. The edge node includes a data processing center that performs redundancy and noise reduction on the data extracted by the sensor module, filtering out various types of interference information. Additionally, the edge node includes an embedded computing module containing various fault diagnosis models, which may include the YOLOv5 algorithm for processing image data, the 1DCNN-SVM algorithm for processing sound data, etc. Here, the normal operation data and abnormal operation data in the second dataset serve as the sample feature set for constructing the fault diagnosis model. The real-time operation data is compared with the sample feature set to determine the health assessment results.
[0148] In some embodiments, the fault diagnosis model for the diagnostic roller is the YOLOv5 algorithm. This YOLOv5 algorithm identifies faults such as weld defects, rubber wear, and rubber detachment in image-type data. The process of establishing the YOLOv5 algorithm is as follows: Figure 4As shown, firstly, images of the drum are acquired using an industrial camera. The resulting dataset is then used to train the network model. Before training begins, the dataset needs to be divided into training, validation, and test sets in appropriate proportions. Next, to achieve better training results, the YOLOv5 algorithm is loaded for pre-training. After modifying the relevant parameters, training begins. The loss functions of the training and validation sets change with the number of iterations. When the loss values of the training and validation sets overlap and tend to stabilize, it indicates that the loss function has converged, and the model's training effect is good. Finally, the model's performance is evaluated using the test set. If it meets the actual requirements, the algorithm model is deployed on an embedded computing card close to the drum; otherwise, the network parameters should be modified and retrained.
[0149] In some embodiments, the fault diagnosis model for idler roller bearings is a 1DCNN-SVM algorithm. First, the acquired vibration signal is subjected to Fourier transform to highlight its fault characteristics. Then, utilizing the powerful feature extraction capability of 1DCNN, relevant feature vectors are extracted from the transformed signal, and SVM classifies the samples based on the extracted features.
[0150] Step 3: Based on the health assessment results and the decision generated by the cloud database, perform maintenance on the belt conveyor in its current state.
[0151] In this embodiment, the cloud database-generated decision may include cause analysis of the current belt conveyor failure and intelligent generation of fault repair plans, etc. Furthermore, the cloud database can be deployed on a central server. After the fault diagnosis model identifies the fault type, it transmits the fault-related data to the central server. The central server makes decisions based on a preset database, which includes a rule base and a case base. The construction of the rule base and case base specifically includes the following steps:
[0152] Step 3.1.1: Acquire the experience of experts in the field of belt conveyors and build a rule base with a hierarchical structure.
[0153] In this embodiment of the application, the failure of the belt conveyor is analyzed layer by layer from failure type to failure cause based on expert experience, and rules from failure type to failure cause are generated.
[0154] Step 3.1.2: Obtain historical maintenance records of the belt conveyor and build a case library with a large number of maintenance records.
[0155] In this embodiment of the application, the case library knowledge comes from maintenance records. By extracting keywords, the maintenance records are represented by nine attributes, namely case name, case time, case location, fault area, accident phenomenon, direct loss, case analysis, solution, and fault parameters. The case library consists of case tables, which correspond to the fault cases of each subsystem of the belt conveyor.
[0156] In steps 3.1.1 and 3.1.2, the domain expert experience and historical maintenance records include faults in various subsystems of the belt conveyor, which may include conveyor belt misalignment, tearing, slippage, idler bearing jamming, roller coating detachment, etc.
[0157] Step 3.2: Based on RBR and CBR, analyze the causes of the fault layer by layer, and generate fault causes and fault repair plans.
[0158] Figure 12 A schematic diagram of the application framework for a fault diagnosis method provided in the embodiments of this application. Figure 2 .like Figure 12 As shown, a rule base and a case base are obtained through a preset database b1. The establishment process of the rule base and case base is as shown in steps 3.1.1 and 3.1.2. The rule-based intelligent diagnostic algorithm b2 for belt conveyors can obtain the cause of the fault. Specifically, it generates a fault tree based on the health assessment results and the rule base, then determines the fault type, calculates the probability of occurrence of at least one candidate fault cause, and determines the fault cause of the belt conveyor in the current state according to the inference rules. The case-based intelligent diagnostic algorithm b3 for belt conveyors can obtain a fault repair plan. Specifically, it identifies the target fault case, calculates the similarity between at least one candidate historical case and the target fault case from the case base, then determines the case analysis and fault repair plan, and then sends the fault diagnosis information b4 (i.e., the fault cause and fault repair plan) to the user equipment so that maintenance personnel can repair the belt conveyor according to the fault diagnosis information. Here, the fault repair plan includes case analysis and solutions.
[0159] Step 3.2.1: Based on the RBR algorithm, analyze the current belt conveyor fault layer by layer according to expert experience until the cause of the fault occurs, including the following steps:
[0160] Step 3.2.1.1: Obtain knowledge from the rule base and generate a fault tree;
[0161] Step 3.2.1.2: Obtain the occurrence probability of each candidate fault cause according to the RBR algorithm, and sort them in descending order;
[0162] In this embodiment, the RBR algorithm is actually a qualitative and quantitative analysis of the fault tree to determine the minimum cut set and its importance, and the priority of the rule is represented by the importance of the bottom event. First, the probability of the top event in the fault tree is calculated based on the probability of the bottom event, and the calculation method is as shown in the above formula (1); second, the importance of the occurrence of a certain minimum cut set to the top event is obtained from the probability of the top event, and the relative probability importance of the minimum cut set is calculated as shown in the above formulas (2) and (3).
[0163] Step 3.2.1.3: Determine the inference rule based on the probability of occurrence of at least one candidate fault cause according to the fault type, and output the fault cause as the result.
[0164] Here, the candidate fault cause corresponding to the largest relative probability importance value can be output as the result, or the candidate fault cause defined by the largest occurrence probability value can be output as the result.
[0165] Step 3.2.2: Based on the CBR algorithm, according to the fault symptoms (i.e., the current real-time operating data of the belt conveyor), match the current belt conveyor fault with historical cases in the case library, and provide a fault repair plan, including the following steps:
[0166] Step 3.2.2.1: Obtain the characteristic attributes of the target fault case, and obtain the characteristic values of each characteristic attribute of the target fault case through program recognition or manual input; the characteristic attributes include, but are not limited to, temperature, speed, current and voltage; the characteristic values are divided into text type and numerical type.
[0167] Step 3.2.2.1: Perform case matching. Based on the CBR algorithm, perform similarity matching to obtain the similarity between at least one candidate historical case and the target case.
[0168] In this embodiment, the main calculation process of the CBR algorithm is as follows: the numerical feature attribute similarity calculation method is as shown in the above formula (4); the text attribute similarity calculation method is as shown in the above formula (5); the overall similarity between candidate historical cases and target fault cases is as shown in the above formula (6).
[0169] Step 3.2.2.1: Select one or N candidate historical cases with the highest similarity, and output their case analysis and solutions. Alternatively, filter candidate historical cases that meet the requirements by setting a preset similarity threshold, and output their case analysis and solutions.
[0170] In the embodiments of this application, see Figure 6 and Figure 7 The pre-set fault database includes a rule base and a case base. The rule base stores expert experience, and the case base stores maintenance records. Faults in the belt conveyor are diagnosed by combining expert experience and maintenance records.
[0171] This application provides a fault diagnosis method, specifically a cloud-edge-device collaborative intelligent fault diagnosis method for the entire process of a belt conveyor. The specific implementation of the aforementioned embodiments is described in detail through the above embodiments. It can be seen that by fusing multiple sensing data (i.e., the first dataset) through sensor modules and determining the corresponding fault diagnosis model, the extraction and analysis of various data types such as images, sounds, vibrations, and currents from the belt conveyor can be achieved. This improves data mining and analysis of multiple data sources, making fault diagnosis more accurate and achieving comprehensive, three-dimensional fault diagnosis of the belt conveyor. Furthermore, by integrating historical maintenance cases and expert experience of the belt conveyor using cloud devices to form a rule base and case base, and then analyzing the causes of faults and generating fault repair plans, the entire fault diagnosis process can be covered, including fault identification, fault location, and fault repair plan generation, achieving intelligent fault diagnosis throughout the entire process. In addition, through the collaboration of cloud, edge, and terminal, tasks such as perception, redundancy reduction, noise reduction, classification, and decision-making are deployed on different terminals according to different computing power requirements. Among them, the sensor module is responsible for extracting various types of data, and the edge device reduces redundancy and noise of the massive amount of data extracted, classifies it, and then transmits it to the cloud device, realizing lightweight data transmission. The cloud device makes decisions, outputs the cause of the fault and the fault repair plan, reduces the computing pressure of the cloud device, ensures the timeliness of data processing, and thus improves the production safety of the mine.
[0172] Based on the same inventive concept as the foregoing embodiments. Figure 13 This is a schematic diagram illustrating the structural composition of an edge device provided in an embodiment of this application. Figure 13 As shown, the edge device 130 may include a first transceiver module 1301 and a first determination unit 1302, wherein:
[0173] The first transceiver module 1301 is used to receive the first dataset sent by the sensor module;
[0174] The first determining module 1302 is used to determine the corresponding fault diagnosis model based on the first dataset, and to determine the health assessment result based on the fault diagnosis model.
[0175] The first transceiver module 1301 is used to send the health assessment results to the cloud device.
[0176] In some embodiments, see Figure 13 The edge device 130 also includes a data processing center 1303 and an embedded computing module 1304, wherein: the data processing center 1304 is used to preprocess the first dataset to determine the second dataset; the embedded computing module 1304 is used to determine the corresponding fault diagnosis model based on the second dataset, and to classify faults based on the second dataset and the fault diagnosis model to determine the health assessment result.
[0177] In some embodiments, the first determining module 1302 is specifically used to determine the data type corresponding to the second dataset; if the data type is a first type, then the fault diagnosis model is determined to be a first algorithm model; and if the data type is a second type, then the fault diagnosis model is determined to be a second algorithm model.
[0178] In some embodiments, the embedded computing module 1304 is specifically used to perform fault classification based on the second dataset and the first algorithm model to determine the health assessment result if the data type is a first type; and to perform fault classification based on the second dataset and the second algorithm model to determine the health assessment result if the data type is a second type.
[0179] Understandably, in this embodiment, a "unit" can be a portion of a circuit, a portion of a processor, a portion of a program or software, etc., and can also be a module or a non-modular component. Furthermore, the components in this embodiment can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional module.
[0180] If the integrated unit is implemented as a software functional module and not sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this embodiment, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the method described in this embodiment. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0181] Therefore, this embodiment provides a computer-readable storage medium applied to an edge device 130. The computer-readable storage medium stores a computer program, which, when executed by a first processor, implements the fault diagnosis method executed by any of the edge devices in the foregoing embodiments.
[0182] Based on the composition of the aforementioned edge device 130 and the computer-readable storage medium, Figure 14 This is a schematic diagram illustrating the specific hardware structure of an edge device provided in an embodiment of this application. For example... Figure 14As shown, the edge device 130 may include: a first communication interface 1401, a first memory 1402, and a first processor 1403; the various components are coupled together via a first bus system 1404. It is understood that the first bus system 1404 is used to implement communication between these components. In addition to a data bus, the first bus system 1404 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 14 The general designated all buses as the first bus system 1404. Among them,
[0183] The first communication interface 1401 is used for receiving and sending signals during the process of sending and receiving information with other external network elements;
[0184] The first memory 1402 is used to store computer programs that can run on the first processor 1403;
[0185] The first processor 1403 is used to execute the following when running computer programs:
[0186] Receive the first dataset sent by the sensor module; determine the corresponding fault diagnosis model based on the first dataset, determine the health assessment result based on the fault diagnosis model, and send the health assessment result to the cloud device.
[0187] It is understood that the first memory 1402 in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate Synchronous DRAM (DDRSDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), and Direct Rambus RAM (DRRAM). The first memory 1402 of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.
[0188] The first processor 1403 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the first processor 1403 or by instructions in software form. The first processor 1403 may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules may reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in the first memory 1402. The first processor 1403 reads the information in the first memory 1402 and completes the steps of the above method in conjunction with its hardware.
[0189] It is understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, microcode, or a combination thereof. For hardware implementation, the processing unit can be implemented in one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers, microprocessors, other electronic units for performing the functions described herein, or combinations thereof.
[0190] For software implementation, the techniques described herein can be achieved through modules (e.g., procedures, functions, etc.) that perform the functions described herein. The software code can be stored in memory and executed by a processor. The memory can be implemented within the processor or externally.
[0191] Alternatively, as another embodiment, the first processor 1403 is also configured to execute the fault diagnosis method performed by any of the edge devices in the foregoing embodiments when running the computer program.
[0192] This application provides an edge device that collects multiple sensing data through a sensor array and analyzes it according to a corresponding fault diagnosis model. This improves data mining and analysis of multiple data sets, leading to more accurate fault diagnosis. Furthermore, by coordinating cloud, edge, and terminal operations, the entire fault diagnosis process is differentiated, resulting in lightweight data transmission, reduced computational burden on cloud devices, and ensured timely data processing. This improves the timeliness of belt conveyor fault diagnosis, thereby enhancing mine production safety.
[0193] In yet another embodiment of this application, based on the same inventive concept as the foregoing embodiments, Figure 15 This is a schematic diagram illustrating the structural composition of a cloud device provided in an embodiment of this application. Figure 15 As shown, the cloud device 150 may include a second transceiver module 1501 and a second determining unit 1502, wherein:
[0194] The second transceiver module 1501 is used to receive health assessment results sent by the edge device;
[0195] The second determining unit 1502 is used to determine the fault diagnosis information of the belt conveyor based on the health assessment results and the preset database.
[0196] In some embodiments, see Figure 15 The cloud device 150 also includes a rule reasoning module 1503 and a case reasoning module 1504. The rule reasoning module 1503 is used to perform rule-based reasoning analysis based on the health assessment results and the rule base to determine the cause of the belt conveyor failure. The case reasoning module 1504 is used to perform case-based reasoning analysis based on the health assessment results and the case base to determine the failure maintenance plan for the belt conveyor.
[0197] In some embodiments, the rule reasoning module 1503 is specifically used to generate a fault tree based on a rule base; analyze the health assessment results layer by layer according to the fault tree to determine the probability of occurrence of at least one candidate fault cause; and determine the fault cause based on the probability of occurrence of at least one candidate fault cause.
[0198] In some embodiments, the second determining unit 1302 is specifically used to construct a rule base based on the domain expert experience of the belt conveyor; and to construct a case base based on the historical maintenance records of the belt conveyor.
[0199] In some embodiments, the case reasoning module 1504 is specifically used to determine the feature value of at least one feature attribute of the target fault case based on the health assessment results; to perform similarity matching between the target fault case and the case library based on the feature value of at least one feature attribute, to determine the similarity between at least one candidate historical case and the target fault case; and to determine the target historical case from at least one candidate historical case based on the similarity, and to determine the target historical case as a fault repair plan.
[0200] Understandably, in this embodiment, a "unit" can be a portion of a circuit, a portion of a processor, a portion of a program or software, etc., and can also be a module or a non-modular component. Furthermore, the components in this embodiment can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional module.
[0201] If the integrated unit is implemented as a software functional module and is not sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, embodiments of this application provide a computer-readable storage medium applied to a cloud device 150. This computer-readable storage medium stores a computer program, which, when executed by a second processor, implements the fault diagnosis method executed by any of the cloud devices in the aforementioned embodiments.
[0202] Based on the composition of cloud device 150 and computer-readable storage media, Figure 16 This is a schematic diagram of the specific hardware structure of a cloud device provided in an embodiment of this application. For example... Figure 16 As shown, the cloud device 150 may include: a second communication interface 1601, a second memory 1602, and a second processor 1603; the various components are coupled together via a second bus system 1604. It is understood that the second bus system 1604 is used to implement communication between these components. In addition to a data bus, the second bus system 1604 also includes a power bus, a control bus, and a status signal bus. However, for clarity, in... Figure 16 The various buses are all labeled as the second bus system 1604. Among them,
[0203] The second communication interface 1601 is used for receiving and sending signals during the process of sending and receiving information with other external network elements;
[0204] The second memory 1602 is used to store computer programs that can run on the second processor 1603;
[0205] The second processor 1603 is configured to perform the following when running the computer program:
[0206] Receive health assessment results sent by edge devices; determine fault diagnosis information for the belt conveyor based on the health assessment results and a preset database. The fault diagnosis information includes the cause of the fault and the fault repair plan.
[0207] Alternatively, as another embodiment, the second processor 1603 is also configured to execute the fault diagnosis method performed by any of the cloud devices in the foregoing embodiments when running a computer program.
[0208] It is understood that the second memory 1602 has similar hardware functions to the first memory 1402, and the second processor 1603 has similar hardware functions to the first processor 1403; these will not be described in detail here.
[0209] This application provides a cloud-based device that makes decisions based on health assessment results and a preset database (such as a rule base formed by expert experience and a case base formed by historical maintenance records). It can also cover the entire fault diagnosis process, including fault identification, fault cause analysis, and generation of fault repair plans, thereby enabling intelligent fault diagnosis throughout the entire process and improving mine production safety.
[0210] In another embodiment of this application, based on the same inventive concept as the foregoing embodiments, this application provides a fault diagnosis system 10, such as... Figure 1 As shown, the fault diagnosis system 10 may include a sensor module 101, an edge device 102, and a cloud device 103, wherein:
[0211] Sensor module 101 is used to collect data on the operating status of the belt conveyor and construct the first dataset;
[0212] Edge device 102 is used to determine the corresponding fault diagnosis model based on the first dataset, and to determine the health assessment result based on the fault diagnosis model;
[0213] The cloud device 103 is used to make decisions based on health assessment results and a preset database to determine fault diagnosis information for the belt conveyor. The fault diagnosis information includes the cause of the fault and the fault repair plan.
[0214] In some embodiments, edge device 102 may be edge device 130 as described in any of the foregoing embodiments, and cloud device 103 may be cloud device 150 as described in any of the foregoing embodiments.
[0215] In some embodiments, the sensor module 101 includes at least one of the following: a sensor mounted on a conveyor belt, a sensor mounted on a roller, a sensor mounted on a idler roller, a sensor mounted on a reducer, a sensor mounted on a motor, and a sensor mounted on an inspection robot.
[0216] In some embodiments, the edge device 102 includes a data processing center and an embedded computing module, wherein: the data processing center is used to preprocess a first dataset to determine a second dataset; the embedded computing module is used to determine a corresponding fault diagnosis model based on the second dataset, and to classify faults based on the second dataset and the fault diagnosis model to determine a health assessment result.
[0217] In some embodiments, the cloud device 103 includes a rule-based reasoning module and a case-based reasoning module, wherein: the rule-based reasoning module is used to perform rule-based reasoning analysis based on health assessment results and a rule base to determine the cause of failure of the belt conveyor; the case-based reasoning module is used to perform case-based reasoning analysis based on health assessment results and a case base to determine a failure maintenance plan for the belt conveyor.
[0218] In some embodiments, the rule reasoning module is specifically configured to: generate a fault tree based on a rule base; analyze the health assessment results layer by layer according to the fault tree to determine the probability of occurrence of at least one candidate fault cause; and determine the fault cause based on the probability of occurrence of at least one candidate fault cause.
[0219] In some embodiments, the case reasoning module is specifically used to determine the feature value of at least one feature attribute of the target fault case based on the health assessment results; to perform similarity matching between the target fault case and the case library based on the feature value of at least one feature attribute, to determine the similarity between at least one candidate historical case and the target fault case; and to determine the target historical case from at least one candidate historical case based on the similarity, and to determine the target historical case as a fault repair plan.
[0220] This application provides a fault diagnosis system that improves the timeliness of fault diagnosis for belt conveyors through the collaborative work of sensor modules, edge devices, and cloud devices. Furthermore, by collecting multiple sensing data from the sensor group and analyzing it according to the corresponding fault diagnosis model, it enhances data mining and analysis of multiple data sources, making fault diagnosis more accurate. Moreover, the cloud device makes decisions based on health assessment results and a preset database (such as a rule base formed from expert experience and a case database formed from historical maintenance records), covering the entire fault diagnosis process, including fault identification, fault cause analysis, and the generation of fault repair plans. This enables intelligent fault diagnosis throughout the entire process, thereby improving mine production safety.
[0221] In yet another embodiment of this application, a computer program product is also provided, including a computer program or instructions that, when executed by a processor, implement the steps of the method as described in any of the foregoing embodiments.
[0222] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, devices, or computer program products. Therefore, this application can take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage) containing computer-usable program code.
[0223] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0224] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0225] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0226] It should be noted that, in this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, product, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, product, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, product, or apparatus that includes that element.
[0227] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.
[0228] The methods disclosed in the several method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.
[0229] The features disclosed in the several product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.
[0230] The features disclosed in the several method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method or device embodiments.
[0231] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
Claims
1. A fault diagnosis method for a belt conveyor, characterized in that, The method includes: Based on the data collected by the sensor module, the first dataset is constructed; The edge device determines the corresponding fault diagnosis model based on the first dataset, and determines the health assessment result based on the fault diagnosis model; Based on the health assessment results and the rule base, rule-based reasoning analysis is performed to determine the cause of the belt conveyor failure. Based on the health assessment results, the feature value of at least one characteristic attribute of the target failure case is determined; based on the feature value of the at least one characteristic attribute, the target failure case is matched with the case library to determine the similarity between at least one candidate historical case and the target failure case; based on the similarity, the target historical case is determined from the at least one candidate historical case, and the target historical case is determined as the failure maintenance plan.
2. The method according to claim 1, characterized in that, The data collected by the sensor module includes at least one of the following: Data collected by sensors installed on the conveyor belt; Data collected by sensors mounted on the roller; Data collected by sensors installed on the idler roller; Data collected by sensors installed on the speed reducer; Data collected by sensors mounted on the motor; Data collected by sensors mounted on the inspection robot.
3. The method according to claim 1, characterized in that, The edge device determines the corresponding fault diagnosis model based on the first dataset, including: The first dataset is preprocessed to determine the second dataset; Based on the second dataset, the corresponding fault diagnosis model is determined.
4. The method according to claim 3, characterized in that, The step of determining the corresponding fault diagnosis model based on the second dataset includes: Determine the data type corresponding to the second dataset; If the data type is the first type, then the fault diagnosis model is determined to be the first algorithm model; If the data type is the second type, then the fault diagnosis model is determined to be the second algorithm model; Accordingly, determining the health assessment result based on the fault diagnosis model includes: If the data type is the first type, then the fault classification is performed based on the second dataset and the first algorithm model to determine the health assessment result; If the data type is the second type, then the fault classification is performed based on the second dataset and the second algorithm model to determine the health assessment result.
5. The method according to claim 1, characterized in that, The method further includes: The rule base is constructed based on the domain expert experience of the belt conveyor. The case library is constructed based on the historical maintenance records of the belt conveyor.
6. The method according to claim 1, characterized in that, The step of determining the cause of failure of the belt conveyor by performing rule-based reasoning analysis based on the health assessment results and rule base includes: Based on the rule base, a fault tree is generated; The health assessment results are analyzed layer by layer according to the fault tree to determine the probability of occurrence of at least one candidate fault cause; The cause of failure is determined based on the probability of occurrence of the at least one candidate cause of failure.
7. The method according to claim 1, characterized in that, The step of determining the target historical case from the at least one candidate historical case based on the similarity includes: One or more candidate historical cases with the highest similarity are identified as the target historical case; or... One or more candidate historical cases with a similarity greater than a preset similarity threshold are identified as the target historical case.
8. A fault diagnosis system for a belt conveyor, characterized in that, The fault diagnosis system includes a sensor module, edge devices, and cloud devices. The cloud devices include a rule-based reasoning module and a case-based reasoning module, wherein: The sensor module is used to collect data on the operating status of the belt conveyor and construct a first dataset; The edge device is used to determine a corresponding fault diagnosis model based on the first dataset, and to determine a health assessment result based on the fault diagnosis model. The rule reasoning module is used to perform rule-based reasoning analysis based on the health assessment results and the rule base to determine the cause of the belt conveyor failure. The case reasoning module is used to determine the feature value of at least one characteristic attribute of the target fault case based on the health assessment results; to perform similarity matching between the target fault case and the case library based on the feature value of the at least one characteristic attribute, and to determine the similarity between at least one candidate historical case and the target fault case; to determine the target historical case from the at least one candidate historical case based on the similarity, and to determine the target historical case as a fault repair plan.
9. The fault diagnosis system according to claim 8, characterized in that, The sensor module includes at least one of the following: a sensor installed on a conveyor belt, a sensor installed on a roller, a sensor installed on a idler roller, a sensor installed on a reducer, a sensor installed on a motor, and a sensor installed on an inspection robot.
10. The fault diagnosis system according to claim 8, characterized in that, The edge device includes a data processing center and an embedded computing module, wherein: The data processing center is used to preprocess the first dataset to determine the second dataset; The embedded computing module is used to determine the corresponding fault diagnosis model based on the second dataset, and to classify faults based on the second dataset and the fault diagnosis model to determine the health assessment result.
11. The fault diagnosis system according to claim 8, characterized in that, The rule reasoning module is specifically used to generate a fault tree based on the rule base; to analyze the health assessment results layer by layer according to the fault tree to determine the probability of occurrence of at least one candidate fault cause; and to determine the fault cause based on the probability of occurrence of the at least one candidate fault cause.
Citation Information
Patent Citations
Intelligent online monitoring system of belt conveyor
CN114275483A
Abnormal vibration fault diagnosis method and system for speed reducer of belt conveyor and medium
CN115979622A
