Belt conveyor state monitoring method and system based on distributed sound wave sensing

Abstract

The invention discloses a belt conveyor state monitoring method and system based on distributed sound wave sensing, and relates to the technical field of coal mine underground safety monitoring. Comprising the steps that vibration data, sound data and temperature data acting on the belt conveyor in normal operation and different fault states are collected; obtaining processed vibration features, sound features and temperature features in normal operation and different fault states, and obtaining a normal feature image and a fault feature image; comparing the normal feature image with the fault feature image to obtain critical thresholds of different states, and when the critical thresholds are exceeded, identifying and marking the feature image of the fault state; and a fault occurrence area is positioned by using a distributed sonic sensor, a plurality of overlapped fault intersection areas are taken, the type of a feature image is determined, the fault position is positioned and marked, and the state monitoring of the belt conveyor is realized. According to the invention, high-frequency real-time monitoring is realized, the risk of fault accumulation and expansion is reduced, and the position of an abnormal signal is accurately positioned.

Description

Technical Field

[0001] The present invention relates to the technical field of underground coal mine safety monitoring, and particularly to a belt conveyor condition monitoring method and system based on distributed acoustic sensing. Background Art

[0002] Coal has always been an important basic energy source and industrial raw material in China, supporting the sustainable and rapid development of the economy and society. The efficient development and rational utilization of coal are major events in the world's energy industry today. The fully mechanized coal mining technology integrating support, mining, loading, transportation, etc. is an important mining technology. Among them, belt conveyors are widely used in coal transportation and undertake the arduous tasks of transporting a large amount of coal, materials, etc. As a key equipment for coal enterprise transportation, once a problem occurs, it will directly affect the entire coal production, and further affect whether the entire coal enterprise can operate efficiently and stably.

[0003] In recent years, with the continuous expansion of coal mine production scale, the use of overweight belt conveyors has become increasingly widespread. Belt conveyors often suffer from various faults due to the harsh underground environment: conveyor belt faults, such as belt deviation, slipping, fracture and wear; problems with drive devices: motor overload, bearing wear, reducer failure; problems with rollers and idlers: abnormal vibration, bearing failure, insufficient lubrication, etc. In order to improve the overall transportation efficiency of coal enterprises, strict maintenance management and fault prevention should be carried out on belt conveyors.

[0004] At present, there have been certain breakthroughs in the fault monitoring methods and systems for underground conveyors in China. Common traditional monitoring methods mostly use point sensors, which are sparsely arranged and difficult to achieve full-coverage monitoring of the entire conveying system. However, the distributed fiber optic acoustic sensing technology is a new data acquisition technology that uses optical fibers as carriers, perceives through light waves, realizes signal acquisition and transmission, accurately locates abnormal signal sources, and measures them. Compared with traditional geophones, distributed acoustic sensing has obvious advantages such as full-well section reception, all-time, high density, anti-electromagnetic interference, high temperature and high pressure resistance, and permanent use after one deployment.

[0005] Therefore, it is an urgent problem for those skilled in the art to propose a belt conveyor condition monitoring method and system based on distributed acoustic sensing to solve the difficulties existing in the prior art. Summary of the Invention

[0006] In view of this, the present invention provides a belt conveyor condition monitoring method and system based on distributed acoustic sensing, which realizes high-frequency real-time monitoring, quickly captures abnormal signals, reduces the risk of fault accumulation and expansion, accurately locates the position of abnormal signals, realizes the identification of early faults, and gives early warnings of potential problems to avoid small problems evolving into serious faults.

[0007] To achieve the above object, the present invention adopts the following technical solutions:

[0008] A belt conveyor condition monitoring method based on distributed acoustic sensing, comprising the following steps:

[0009] Collect vibration data, sound data and temperature data acting on the belt conveyor under normal operation and different fault conditions;

[0010] Process the vibration data, sound data and temperature data under normal operation and different fault conditions to obtain the vibration characteristics, sound characteristics and temperature characteristics under normal operation and different fault conditions after processing, so as to obtain normal characteristic images and fault characteristic images;

[0011] Compare the normal characteristic images and the fault characteristic images to obtain the critical thresholds of different states. When the critical thresholds are exceeded, identify and mark the characteristic images of the fault states;

[0012] Use a distributed acoustic sensor to locate the fault occurrence area, take the intersection area where multiple faults overlap, determine the type of the characteristic image, so as to locate and mark the fault position, and realize the condition monitoring of the belt conveyor.

[0013] Optionally, lay a single-mode optical fiber along the entire conveying path of the belt conveyor to cover the entire operating path of the conveyor belt, and randomly set a plurality of distributed acoustic sensor nodes on the entire single-mode optical fiber path, so as to collect vibration data, sound data and temperature data acting on the belt conveyor under normal operation and different fault conditions.

[0014] Optionally, the specific content of collecting vibration data, sound data and temperature data acting on the belt conveyor under normal operation and different fault conditions is:

[0015] Collect normal operation data in real time to generate continuous time-series signal data. Under the normal operation state of the belt conveyor, record the vibration data, sound data and temperature data of each part;

[0016] The fault conditions include friction, impact, belt jamming, motor failure and component looseness. Record the vibration data, sound data and temperature data of each part under different fault conditions.

[0017] Optionally, the specific content of processing the vibration data, sound data and temperature data under normal operation and different fault conditions to obtain the vibration characteristics, sound characteristics and temperature characteristics under normal operation and different fault conditions after processing, so as to obtain normal characteristic images and fault characteristic images is:

[0018] Use deep learning algorithms to extract and optimize features from vibration data, sound data, and temperature data, and fuse them through weighted averaging to obtain normal feature images and fault feature images.

[0019] Optionally, the deep learning algorithm uses a CNN-LSTM model to extract and optimize features from vibration data, sound data, and temperature data. The CNN-LSTM model first extracts spatial features of local vibration patterns and signal waveform morphologies through CNN, reduces the dimensions of the features to form a one-dimensional feature vector; secondly, captures the time series characteristics of vibration signals through LSTM to identify dynamic changes; thirdly, adds a Softmax classifier after the fully connected layer of the CNN-LSTM model to classify the current state; finally, combines DAS positioning to determine the fault occurrence location through time delay and amplitude difference, takes the intersection area of multiple fault occurrences to locate the fault location and mark it, generates a state trend curve according to the change trend of the signal, and uses the sliding window analysis technique to monitor potential faults of the belt conveyor according to the trend curve.

[0020] Optionally, it also includes grading the severity of the fault into low-level faults, medium-level faults, and high-level faults;

[0021] Low-level faults are slight vibrations or short-term abnormal signals, which have no direct impact on the operation of the equipment. A slight fault prompt is displayed through the local display to alert the operator; medium-level faults are continuous vibrations or medium abnormal signals, and the audible and visual alarm is activated to notify the maintenance personnel for inspection and repair; high-level faults are serious faults or catastrophic signals, which immediately trigger the shutdown mechanism to avoid the spread of the fault, protect the safety of the equipment, and send a high-priority alarm at the same time.

[0022] A belt conveyor status monitoring system based on distributed acoustic sensing, applying a belt conveyor status monitoring method based on distributed acoustic sensing according to any one of the above, includes: an acoustic sensing module, an adjustable filtering module, a data acquisition module, a data transmission module, a ground monitoring center, and an analysis and processing module;

[0023] The acoustic sensing module, connected to the input end of the adjustable filtering module, is used to capture the acoustic signals generated by the operation of the belt conveyor and convert these acoustic signals into electrical signals;

[0024] The adjustable filtering module, connected to the input end of the data acquisition module, is used to denoise and extract the acoustic signals using a filtering algorithm;

[0025] The data acquisition module, connected to the input end of the data transmission module, is used to collect and summarize the status monitoring data of the belt conveyor;

[0026] A data transmission module, connected to the input end of the ground monitoring center, is used to transmit the collected and aggregated status monitoring data to the ground monitoring center using an industrial-grade optical fiber network or 5G wireless communication;

[0027] The ground monitoring center, connected to the input end of the analysis and processing module, is used to process, display, query, store, and print the status monitoring data of the belt conveyor underground, distinguish and analyze the danger of various monitored status parameters, determine the severity of the accident, delineate the intersection area of multiple faults, locate the fault occurrence position, conduct real-time monitoring and abnormal early warning, and immediately display the fault level to the staff;

[0028] The analysis and processing module is used for data analysis and can process and explain the reasons for various status data occurrences.

[0029] It can be seen from the above technical solutions that compared with the prior art, the present invention provides a belt conveyor status monitoring method and system based on distributed acoustic sensing, having the following beneficial effects:

[0030] (1) Based on distributed acoustic sensors, the present invention breaks through the limitation of sparse point sensor arrangement in traditional monitoring methods, which makes it difficult to achieve full coverage monitoring of the entire conveying system. The distributed sensing nodes in the optical fiber can collect thousands of data points per second, realizing high-frequency real-time monitoring, quickly capturing abnormal signals, reducing the risk of fault accumulation and expansion. The distributed acoustic sensing technology can accurately locate the position of abnormal signals according to the spatio-temporal distribution of optical fiber signals, capture minute vibrations or acoustic signal changes, realize the identification of early faults, give early warnings of potential problems, and prevent small problems from evolving into serious faults;

[0031] (2) By analyzing the frequency changes, amplitude characteristics, etc. of vibration signals, the present invention predicts the trend of equipment operation status, plans maintenance tasks in advance, reduces unplanned outages. The distributed acoustic sensing monitoring combined with a deep learning model intelligently classifies and grades the severity of fault signals, triggers different alarm responses according to the fault level, solves the problem of frequent but inaccurate alarms in traditional monitoring systems, avoids production interference caused by false alarms, realizes accurate alarms, effectively guides maintenance personnel to respond quickly, formulates targeted maintenance plans through fault location and classification, avoids unnecessary full-system maintenance, improves equipment utilization rate, reduces maintenance time and labor costs, analyzes the operation trend of equipment by combining long-term monitoring data, optimizes the maintenance cycle, reduces production losses caused by sudden faults, protects the safety of underground operation personnel, reduces accident risks, optimizes the conveyor operation parameters through data analysis, improves the overall efficiency, and provides strong support for the development of intelligent mines. Description of the Drawings

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on the provided drawings.

[0033] Figure 1 Flow chart of a belt conveyor status monitoring method based on distributed acoustic sensing provided by the present invention;

[0034] Figure 2 Flow chart of belt conveyor status monitoring provided by the present invention;

[0035] Figure 3 Schematic diagram of the distributed acoustic sensing feature algorithm provided by the present invention;

[0036] Figure 4 Schematic diagram of the partial layout structure of the distributed optical fiber sensor provided by the present invention;

[0037] Figure 5 Schematic diagram of another partial layout structure of the distributed optical fiber sensor provided by the present invention;

[0038] Among them, 1 - Arranged in a winding manner along the drive motor and the driving drum, 2 - Arranged along the support frame and the belt tension device, 3 - Arranged along the commutator and the guiding device, 4 - Arranged along the belt joint and the connecting part. Specific embodiments

[0039] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0040] Referring to Figure 1 and Figure 2 As shown, the present invention discloses a belt conveyor status monitoring method based on distributed acoustic sensing, including the following steps:

[0041] Collect vibration data, sound data, and temperature data acting on the belt conveyor under normal operation and different fault states;

[0042] Process the vibration data, sound data, and temperature data under normal operation and different fault states to obtain the vibration characteristics, sound characteristics, and temperature characteristics under normal operation and different fault states after processing, so as to obtain a normal feature image and a fault feature image;

[0043] Compare the normal feature image and the fault feature image to obtain the critical thresholds for different states. When the critical threshold is exceeded, identify the feature image of the fault state and mark it;

[0044] Use a distributed acoustic wave sensor to locate the fault occurrence area, take the intersection area of multiple faults overlapping, determine the type of feature image, so as to locate and mark the fault location, and realize the condition monitoring of the belt conveyor.

[0045] Furthermore, lay a single-mode optical fiber along the entire conveying path of the belt conveyor to cover the entire operating path of the conveyor belt. Randomly set multiple distributed acoustic wave sensor nodes along the entire single-mode optical fiber path, so as to collect vibration data, sound data and temperature data acting on the belt conveyor under normal operation and different fault states.

[0046] Specifically, as Figure 4 and Figure 5 shown, the sensing medium selects a single-mode optical fiber with high sensitivity, long-distance transmission and durability. Lay the optical fiber along the entire conveying path of the belt conveyor to ensure that it covers the entire operating path of the conveyor belt. Arrange it in a winding manner 1 along the driving motor and the driving drum, arrange it 2 along the support frame and the belt tension device, arrange it 3 along the commutator and the guiding device, and arrange it 4 along the belt joint and the connecting part. Focus on monitoring the vibration data, sound data and temperature data in these areas; Set multiple distributed acoustic wave sensor nodes along the entire optical fiber path. These sensor nodes are installed at random positions to ensure that the nodes cover the monitoring area of the entire belt conveyor and are not restricted by fixed positions.

[0047] Furthermore, the specific content of collecting vibration data, sound data and temperature data acting on the belt conveyor under normal operation and different fault states is as follows:

[0048] Collect normal operation data in real time to generate continuous time-series signal data. Under the normal operation state of the belt conveyor, record the vibration data, sound data and temperature data of each part;

[0049] The fault states include friction, impact, belt jamming, motor failure and component loosening. Record the vibration data, sound data and temperature data of each part under different fault states.

[0050] Specifically, signal acquisition is performed every 100 ms to generate continuous time-series signal data. During normal operation data acquisition, while the belt conveyor is in normal operation, vibration data, sound data, and temperature data of each part are recorded to form a characteristic image under normal conditions. Fault state data is collected, and the fault data includes abnormal vibrations, temperatures, and sound waves caused by friction, impact, belt jamming, motor faults, and component looseness. The images include vibration amplitude waveforms, sound wave spectrograms, and temperature curves. Based on the analysis of the signals in the normal operation state, the critical thresholds of vibration data, sound data, and temperature data are extracted. When the collected signals exceed the critical thresholds, the fault state is identified and the image waveforms of the fault signals are marked. The signal data is formed into a spatio-temporal two-dimensional signal matrix, including vibration, temperature, sound wave amplitudes, and monitoring positions. Adaptive filtering technology is used to eliminate noise interference in the underground environment. The vibration signal is transformed from the time domain to the time-frequency domain characteristics using wavelet transform. The amplitudes and frequencies of the collected signals are normalized to eliminate errors and enhance signal accuracy. The continuous time-series vibration signals are cut into multiple signal segments with a time window of 2 seconds for each window, and each signal segment is state-labeled to form a signal data set.

[0051] Furthermore, the vibration data, sound data, and temperature data under normal operation and different fault states are processed to obtain the vibration characteristics, sound characteristics, and temperature characteristics under normal operation and different fault states after processing. The specific content of the normal characteristic image and the fault characteristic image is as follows:

[0052] Deep learning algorithms are used to extract and optimize the characteristics of vibration data, sound data, and temperature data, and fusion is performed through weighted averaging to obtain the normal characteristic image and the fault characteristic image.

[0053] Furthermore, as Figure 3 shown, the deep learning algorithm uses the CNN-LSTM model to extract and optimize the characteristics of vibration data, sound data, and temperature data. The CNN-LSTM model first extracts the spatial characteristics of local vibration patterns and signal waveform morphologies through CNN, reduces the dimensions of the characteristics to form a one-dimensional feature vector. Secondly, it captures the time-series characteristics of the vibration signal through LSTM to identify dynamic changes. Thirdly, a Softmax classifier is added after the fully connected layer of the CNN-LSTM model to classify the current state. Finally, combined with DAS positioning, the fault occurrence location is determined through time delay and amplitude difference. The intersection area of multiple fault occurrences is taken to locate the fault location and mark it. A state trend curve is generated according to the change trend of the signal. According to the trend curve, the sliding window analysis technology is used to monitor potential faults of the belt conveyor.

[0054] Furthermore, it also includes grading the severity of the faults, which are divided into low-level faults, medium-level faults, and high-level faults;

[0055] Low-level faults are minor vibrations or short-term abnormal signals that have no direct impact on the operation of the equipment. A minor fault prompt is displayed on the local monitor to alert the operator; medium-level faults are continuous vibrations or medium abnormal signals, and the audible and visual alarm is activated to notify the maintenance personnel for inspection and repair; high-level faults are serious faults or catastrophic signals that immediately trigger the shutdown mechanism to prevent the spread of the fault, protect the safety of the equipment, and send a high-priority alarm at the same time.

[0056] and Figure 1 Corresponding to the method described above, an embodiment of the present invention further provides a belt conveyor status monitoring system based on distributed acoustic sensing, including: an acoustic sensing module, an adjustable filtering module, a data acquisition module, a data transmission module, a ground monitoring center, and an analysis and processing module;

[0057] The acoustic sensing module is connected to the input end of the adjustable filtering module and is used to capture the acoustic signals generated by the operation of the belt conveyor and convert these acoustic signals into electrical signals;

[0058] The adjustable filtering module is connected to the input end of the data acquisition module and is used to denoise and extract the acoustic signals using a filtering algorithm;

[0059] The data acquisition module is connected to the input end of the data transmission module and is used to collect and summarize the status monitoring data of the belt conveyor;

[0060] The data transmission module is connected to the input end of the ground monitoring center and is used to transmit the collected and summarized status monitoring data to the ground monitoring center using an industrial-grade fiber optic network or 5G wireless communication;

[0061] The ground monitoring center is connected to the input end of the analysis and processing module and is used to process, display, query, store, and print the status monitoring data of the belt conveyor underground in the mine, distinguish and analyze the hazards of various monitored status parameters, determine the severity of the accident, delineate the intersection area of multiple faults, locate the fault occurrence location, conduct real-time monitoring and abnormal early warning, and immediately display the fault level to the staff;

[0062] The analysis and processing module is used for data analysis and can process and interpret the reasons for the occurrence of various status data.

[0063] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description in the method section.

[0064] The foregoing description of the disclosed embodiments enables those skilled in the art to practice or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A belt conveyor state monitoring method based on distributed acoustic wave sensing, characterized in that: The following steps are involved: Collect vibration data, sound data and temperature data acting on the belt conveyor under normal operation and different fault conditions; Processing the vibration data, sound data and temperature data under normal operation and different fault states to obtain processed vibration characteristics, sound characteristics and temperature characteristics under normal operation and different fault states, thereby obtaining normal characteristic images and fault characteristic images; Compare the normal feature image with the fault feature image to obtain the critical thresholds of different states. When the critical threshold is exceeded, identify the feature image of the fault state and mark it; Distributed acoustic wave sensors are used to locate the fault area, and the intersection area where multiple faults overlap is taken to determine the characteristic image type, so as to locate and mark the fault position and realize the condition monitoring of the belt conveyor.

2. A belt conveyor state monitoring method based on distributed acoustic wave sensing according to claim 1, characterized in that: The single-mode optical fiber is laid along the conveying path of the entire belt conveyor to cover the entire running path of the conveyor belt. Multiple distributed acoustic wave sensor nodes are randomly set on the entire single-mode optical fiber path to collect vibration data, sound data and temperature data acting on the belt conveyor under normal operation and different fault conditions.

3. The belt conveyor state monitoring method based on distributed acoustic wave sensing according to claim 1 is characterized in that: The specific contents of collecting vibration data, sound data and temperature data acting on the belt conveyor under normal operation and different fault conditions are as follows: Real-time normal operation data collection, generating continuous time series signal data, recording vibration data, sound data and temperature data of each part when the belt conveyor is in normal operation; Fault conditions include friction, impact, belt jam, motor failure and loose components. The vibration data, sound data and temperature data of each part under different fault conditions are recorded.

4. The belt conveyor state monitoring method based on distributed acoustic wave sensing according to claim 1 is characterized in that: The vibration data, sound data and temperature data under normal operation and different fault states are processed to obtain the processed vibration characteristics, sound characteristics and temperature characteristics under normal operation and different fault states, so that the specific contents of the normal characteristic image and the fault characteristic image are obtained as follows: Deep learning algorithms are used to extract and optimize the features of vibration data, sound data, and temperature data, which are then fused through weighted averaging to obtain normal feature images and fault feature images.

5. A belt conveyor state monitoring method based on distributed acoustic wave sensing according to claim 1 or 4, characterized in that: The deep learning algorithm uses the CNN-LSTM model to extract and optimize the vibration data, sound data, and temperature data. The CNN-LSTM model first extracts the spatial features of the local vibration mode and signal waveform morphology through CNN, reduces the dimension of the features, and forms a one-dimensional feature vector. Secondly, LSTM is used to capture the time series characteristics of the vibration signal and identify dynamic changes. Again, a Softmax classifier is added after the fully connected layer of the CNN-LSTM model to classify the current state. Finally, combined with DAS positioning, the fault location is determined by time delay and amplitude difference, and the fault location is located and marked in the intersection area of ​​multiple faults. A state trend curve is generated according to the change trend of the signal. Based on the trend curve, sliding window analysis technology is used to monitor potential faults of the belt conveyor.

6. A belt conveyor state monitoring method based on distributed acoustic wave sensing according to claim 1, characterized in that: It also includes grading the severity of faults into low-level faults, medium-level faults, and high-level faults; Low-level faults are slight vibrations or short-term abnormal signals that have no direct impact on equipment operation. Minor fault prompts are displayed on the local display to alert operators. Intermediate faults are continuous vibrations or moderate abnormal signals, which activate the sound and light alarms to notify maintenance personnel to inspect and repair; high-level faults are serious faults or catastrophic signals, which immediately trigger the shutdown mechanism to avoid the spread of the fault, protect equipment safety, and send high-priority alarms.

7. A belt conveyor state monitoring system based on distributed acoustic wave sensing, characterized in that: A belt conveyor state monitoring method based on distributed acoustic wave sensing according to any one of claims 1 to 6, comprising: an acoustic wave sensing module, an adjustable filtering module, a data acquisition module, a data transmission module, a ground monitoring center and an analysis and processing module; The acoustic wave sensing module is connected to the input end of the adjustable filter module and is used to capture the acoustic wave signals generated by the operation of the belt conveyor and convert these acoustic wave signals into electrical signals; An adjustable filter module is connected to the input end of the data acquisition module and is used to denoise and extract the sound wave signal using a filtering algorithm; A data acquisition module is connected to the input end of the data transmission module and is used to collect and summarize the status monitoring data of the belt conveyor; A data transmission module, connected to the input terminal of the ground monitoring center, is used to transmit the collected and summarized status monitoring data to the ground monitoring center using an industrial-grade optical fiber network or 5G wireless communication; The ground monitoring center is connected to the input end of the analysis and processing module to realize the processing, display, query, storage and printing of the status monitoring data of the underground belt conveyor, identify and analyze the danger of various monitored status parameters, determine the severity of the accident, circle the intersection area of ​​multiple faults, locate the location of the fault, conduct real-time monitoring and abnormal warning, and immediately display the fault level to the staff; The analysis and processing module is used for data analysis and can process and explain the causes of various status data.

Images