Belt conveyor intelligent monitoring method and system based on distributed optical fiber sensing

Abstract

The invention provides an intelligent monitoring method and system for a belt conveyor based on distributed optical fiber sensing, and relates to the technical field of industrial Internet of Things and optical fiber sensing signal processing.The intelligent monitoring method comprises the steps that Rayleigh scattering light signals of distributed optical fibers laid along the belt conveyor are collected in real time, a space-time vibration matrix is constructed by means of the signals and preprocessed, and the space-time vibration matrix is obtained; determining linear track parameters of movement of the conveyor belt joint based on the space-time energy diagram, determining the belt speed according to the slope in the linear track parameters, predicting the time when the conveyor belt joint passes through each spatial position of the optical fiber based on the intercept and the belt speed in the linear track parameters, and generating a dynamic space-time mask matrix according to the time; signal cleaning is conducted on the vibration matrix through the mask matrix, the cleaned vibration matrix is obtained, the running state of the belt conveyor is diagnosed based on the cleaned vibration matrix, the technical problem that an existing belt conveyor is insufficient in monitoring reliability is solved, and the technical effect of improving the monitoring reliability of the conveyor is achieved.

Description

Technical Field

[0001] This invention relates to the fields of industrial Internet of Things and fiber optic sensing signal processing technology, and in particular to an intelligent monitoring method and system for belt conveyors based on distributed fiber optic sensing. Background Technology

[0002] As a key piece of equipment for continuous transportation of bulk materials in industries such as coal mining, ports, power, and cement, the real-time and accurate monitoring of the operating status of belt conveyors is a core link in ensuring production safety and improving operation and maintenance efficiency. With the rapid development of industrial Internet of Things and intelligent operation and maintenance technology, the monitoring requirements for conveyors have evolved from traditional single-point, passive alarms to full-line, proactive predictive maintenance. The core is to obtain highly reliable data that can comprehensively reflect the health status of the equipment.

[0003] Among numerous operating parameters, conveyor belt speed is one of the most critical parameters characterizing the operating status of a conveyor. Abnormal fluctuations or deviations in speed are directly related to serious faults such as drive drum slippage, conveyor belt overload, belt breakage, and tension imbalance. Meanwhile, the strong impact signals generated when the conveyor belt joint periodically passes over the idler rollers, although normal operating conditions, have extremely high energy and can easily interfere with the diagnosis of critical faults such as idler roller damage and belt tearing based on vibration analysis. This leads to a high false alarm rate in the monitoring system, seriously affecting its reliability and practicality. Therefore, a new monitoring method for belt conveyors is urgently needed.

[0004] Currently, the industry mainly relies on contact or single-point sensing technologies for belt speed monitoring. For example, one type of solution uses magnetic roller speed measurement technology, which embeds a magnet in the driven roller and uses Hall elements or inductor coils to detect the periodic changes in the magnetic field to calculate the roller speed and estimate the belt speed. However, the drawback is that the measurement object is the roller rather than the belt itself, which cannot reflect the true speed of the belt. When slippage occurs between the drive roller and the belt, the measurement results will be severely distorted. In addition, it requires structural modification of the existing roller and installation of active circuits. It is difficult to deploy in explosion-proof and harsh environments such as underground mines, has low reliability, and high maintenance costs. Another type of solution uses rubber wheel-code disk contact speed measurement, which uses a rubber wheel pressed against the belt surface to follow the movement and drive the photoelectric code disk to output pulses to calculate the speed. Although it can directly contact the belt, its mechanical structure is complex, easily worn, and greatly affected by the impact and vibration of the belt joint. The rubber wheel is prone to jumping, which increases the speed measurement error.

[0005] In summary, existing solutions can only reflect local speed information at monitoring points, and cannot perceive the differences in speed distribution along the entire belt caused by uneven tension. Furthermore, they cannot provide spatiotemporal data for identifying and suppressing the global interference signal of joint impact, thus resulting in poor reliability of belt conveyor monitoring. Summary of the Invention

[0006] In view of the above-mentioned shortcomings in the existing technology, the purpose of this invention is to provide an intelligent monitoring method for belt conveyors based on distributed optical fiber sensing, which has the characteristic of improving the reliability of belt conveyor monitoring.

[0007] The above-mentioned objective of this invention is achieved through the following technical solution: A method for intelligent monitoring of belt conveyors based on distributed optical fiber sensing includes: Rayleigh scattering light signals from distributed optical fibers laid along the belt conveyor are acquired in real time, and a spatiotemporal vibration matrix is ​​constructed based on the Rayleigh scattering light signals. The spatiotemporal vibration matrix is ​​preprocessed to obtain a spatiotemporal energy map; Based on the spatiotemporal energy map, the linear trajectory parameters of the conveyor belt joint movement in the belt conveyor are determined, and the real-time operating speed of the conveyor belt is determined based on the slope in the linear trajectory parameters. Based on the intercept in the linear trajectory parameters and the real-time running speed, the time for the conveyor belt joint to pass through each spatial position of the optical fiber is predicted, and a dynamic spatiotemporal mask matrix is ​​generated based on the predicted time. Based on the dynamic spatiotemporal mask matrix, the spatiotemporal vibration matrix is ​​cleaned to obtain the cleaned spatiotemporal vibration matrix; The operating status of the belt conveyor is diagnosed based on the spatiotemporal vibration matrix after cleaning.

[0008] By adopting the above technical solution, distributed optical fibers are laid along the belt conveyor line. Without contacting the conveyor belt or modifying it, the system intelligently identifies and tracks the movement trajectory of the belt joint from the spatiotemporal vibration signal. This not only accurately reflects the real-time operating speed of the belt, effectively overcoming the distortion problem caused by slippage in traditional contact speed measurement, but also accurately predicts the joint position based on the trajectory and generates a dynamic mask. This actively suppresses the interference of high-intensity joint impact on downstream idler damage, belt tearing, and other fault diagnosis algorithms at the signal level, reducing the false alarm rate. Ultimately, without the need for underground power supply or hardware modification, the system simultaneously achieves the dual goals of high-precision belt speed monitoring and high-reliability fault diagnosis, effectively improving the monitoring reliability of the belt conveyor.

[0009] Preferably, the construction of the spatiotemporal vibration matrix based on the Rayleigh scattering light signal includes: The Rayleigh scattered light signal is coherently demodulated to obtain a demodulated signal; The demodulated signal is filtered to obtain a filtered vibration signal sequence; The filtered vibration signal sequence and its corresponding spatial location are recombined with the acquisition time to construct the spatiotemporal vibration matrix.

[0010] By adopting the above technical solution, the original Rayleigh scattering light signal is demodulated, filtered and reconstructed to effectively extract the effective phase or intensity modulation information caused by the vibration of the conveyor belt and the equipment along the line, and to filter out optical noise, thus constructing a vibration signal matrix with high signal-to-noise ratio and clear spatiotemporal correspondence.

[0011] Preferably, the preprocessing of the spatiotemporal vibration matrix to obtain the spatiotemporal energy map includes: The spatiotemporal vibration matrix is ​​filtered to obtain the filtered spatiotemporal vibration matrix; The envelope of the filtered spatiotemporal vibration matrix is ​​extracted to obtain the envelope matrix. The envelope matrix is ​​subjected to energy normalization to obtain the spatiotemporal energy map.

[0012] By adopting the above technical solution and using a combination of filtering, envelope extraction and energy normalization, the low-frequency oscillation and random noise of the conveyor structure can be effectively suppressed. At the same time, the high-energy transient characteristics generated by events such as belt joint impact are highlighted, and the original complex vibration waveform is converted into a spatiotemporal image that highlights the energy distribution. This allows the joint to form a clear, continuous, high-brightness linear trajectory in the spatiotemporal image, effectively improving the robustness and accuracy of feature recognition.

[0013] Preferably, determining the linear trajectory parameters of the conveyor belt joint movement in the belt conveyor based on the spatiotemporal energy map, and determining the real-time operating speed of the conveyor belt based on the slope in the linear trajectory parameters, includes: High-energy candidate points are extracted from the spatiotemporal energy map to obtain a candidate point set; The candidate point set is subjected to trajectory model fitting processing with outlier removal to obtain the optimized trajectory model; Linear trajectory parameters are extracted from the optimized trajectory model, including the slope as an initial belt velocity estimate. The initial belt speed estimate is smoothed to obtain the real-time operating speed.

[0014] By adopting the above technical solution, high-energy candidate points are extracted from the energy map to focus on effective signals. After robust fitting to eliminate noise interference, the linear trajectory model of the joint motion is accurately obtained. Then, the initial speed is calculated based on the physical mapping relationship of the trajectory slope. Through smoothing processing, random fluctuations and instantaneous anomalies in a single measurement are effectively suppressed. This not only realizes non-contact, direct, and continuous measurement of the actual operating speed of the conveyor belt, but also ensures the stability and reliability of the speed data.

[0015] Preferably, the step of extracting high-energy candidate points from the spatiotemporal energy map to obtain a candidate point set includes: An adaptive threshold is determined based on the background noise level of the spatiotemporal energy map. The energy values ​​of each spatiotemporal coordinate point in the spatiotemporal energy map are compared with the adaptive threshold, and spatiotemporal coordinate points with energy values ​​greater than the adaptive threshold are extracted to construct the candidate point set. By adopting the above technical solution, the threshold is dynamically determined based on background noise, which effectively distinguishes between effective high-energy events representing joint impact and random environmental noise. This allows the candidate point set to accurately focus on real signal features, greatly reducing false selection and missed selection, thereby achieving adaptive feature point screening for different monitoring segments and different environmental noise levels.

[0016] Preferably, the step of predicting the time for the conveyor belt joint to pass through each spatial position of the optical fiber based on the intercept in the linear trajectory parameters and the real-time running speed, and generating a dynamic spatiotemporal mask matrix based on the predicted time, includes: Based on the intercept and the real-time operating speed, the predicted arrival time of the conveyor belt joint at each spatial position of the optical fiber is calculated. For each of the aforementioned spatial locations, a time tolerance interval is determined with the corresponding predicted arrival time as the center. A dynamic spatiotemporal mask matrix is ​​constructed based on the spatial location, the corresponding predicted arrival time, and the time tolerance range.

[0017] By adopting the above technical solution, and combining the accurately identified joint motion trajectory with the real-time measured belt speed, the precise time window for the joint to reach any point on the optical fiber is calculated. By constructing a dynamic mask matrix that is spatiotemporally aligned with the original data, the spatiotemporal region to be suppressed is pre-marked at the data level before the joint signal actually arrives. This predictive masking based on a physical model can accurately filter out high-intensity joint impacts without affecting other normal vibration signals, thereby fundamentally avoiding false triggering of downstream fault diagnosis algorithms and improving monitoring reliability.

[0018] Preferably, the step of performing signal cleaning on the spatiotemporal vibration matrix based on the dynamic spatiotemporal mask matrix to obtain the cleaned spatiotemporal vibration matrix includes: The dynamic spatiotemporal mask matrix and the spatiotemporal vibration matrix are multiplied at corresponding positions to obtain the cleaned spatiotemporal vibration matrix.

[0019] By adopting the above technical solution and using dot multiplication, a dynamic spatiotemporal mask matrix containing prediction information is applied to the original vibration data. In the spatiotemporal dimension, the impact signal area of ​​a specific joint is zeroed or attenuated, while the original vibration information of other spatiotemporal locations is completely preserved. This allows for the output of clean vibration data without introducing complex algorithms or significant delays, greatly improving the accuracy and reliability of state recognition and fault detection.

[0020] The second objective of this invention is to provide an intelligent monitoring system for belt conveyors based on distributed optical fiber sensing, which improves the reliability of belt conveyor monitoring.

[0021] The second objective of this invention is achieved through the following technical solution: A smart monitoring system for belt conveyors based on distributed optical fiber sensing includes: The data acquisition and construction module is used to acquire Rayleigh scattering light signals from distributed optical fibers laid along the belt conveyor in real time, and to construct a spatiotemporal vibration matrix based on the Rayleigh scattering light signals. The signal preprocessing module is used to preprocess the spatiotemporal vibration matrix to obtain a spatiotemporal energy map; The trajectory parameter determination module is used to determine the linear trajectory parameters of the conveyor belt joint movement in the belt conveyor based on the spatiotemporal energy map, and to determine the real-time operating speed of the conveyor belt based on the slope in the linear trajectory parameters. The dynamic mask generation module is used to predict the time it takes for the conveyor belt joint to pass through each spatial position of the optical fiber based on the intercept in the linear trajectory parameters and the real-time running speed, and to generate a dynamic spatiotemporal mask matrix based on the predicted time. The signal cleaning module is used to clean the spatiotemporal vibration matrix based on the dynamic spatiotemporal mask matrix to obtain the cleaned spatiotemporal vibration matrix. The operation status diagnosis module is used to diagnose the operation status of the belt conveyor based on the spatiotemporal vibration matrix after cleaning.

[0022] By adopting the above technical solution, using a single distributed optical fiber as the passive sensing nerve of the entire line, and through a modular processing flow, the original Rayleigh scattering signal is transformed into operating status information with clear physical meaning. Through the collaboration of the trajectory parameter determination module and the dynamic mask generation module, the interference source in the monitoring of the joint motion trajectory is transformed into an information source that provides both speed measurement and interference suppression functions. The signal cleaning module outputs high-quality data. The entire system does not require modification of field equipment or underground power supply, effectively improving the reliability of belt conveyor monitoring.

[0023] The third objective of this invention is to provide an electronic device that improves the reliability of monitoring belt conveyors.

[0024] The above-mentioned objective three of this invention is achieved through the following technical solution: An electronic device includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described above for the intelligent monitoring method of belt conveyor based on distributed optical fiber sensing.

[0025] The fourth objective of this invention is to provide a computer-readable storage medium capable of storing corresponding programs, which facilitates the improvement of the reliability of belt conveyor monitoring.

[0026] The fourth objective of this invention is achieved through the following technical solution: A computer-readable storage medium storing a computer program that can be loaded by a processor and executed by any of the above-described intelligent monitoring methods for belt conveyors based on distributed optical fiber sensing.

[0027] In summary, the present invention has at least one of the following beneficial technical effects: This invention utilizes a single distributed optical fiber as the sensing medium. By analyzing the spatiotemporal trajectory of conveyor belt joint impact, it simultaneously achieves high-precision, non-contact measurement of the actual belt speed across the entire line and accurate prediction of periodic strong interference signals. It transforms the joint impact, a source of interference in traditional monitoring, into a useful information source.

[0028] Based on a precisely measured belt speed and trajectory model, this invention dynamically generates a spatiotemporal mask, which can proactively suppress joint impact signals at the data level, thereby reducing their interference with key fault diagnosis algorithms such as idler damage and belt tearing, and greatly improving the accuracy and reliability of monitoring.

[0029] This invention requires no modification to the existing structure of the conveyor, no underground power supply, and only the laying of a passive optical cable. It has the advantages of simple deployment, resistance to electromagnetic interference, and long-term stability, and is particularly suitable for intelligent monitoring of belt conveyors in long-distance and harsh industrial environments. Attached Figure Description

[0030] Figure 1 This is a flowchart illustrating the steps of an intelligent monitoring method for belt conveyors based on distributed optical fiber sensing, provided in Embodiment 1 of the present invention.

[0031] Figure 2 This is a structural block diagram of an intelligent monitoring system for belt conveyors based on distributed optical fiber sensing, provided in Embodiment 2 of the present invention. Detailed Implementation

[0032] This invention provides an intelligent monitoring method and system for belt conveyors based on distributed optical fiber sensing. It solves the technical problem that existing solutions can only reflect local speed information at monitoring points, failing to perceive the overall speed distribution differences along the entire belt caused by uneven tension, and further failing to provide spatiotemporal data for identifying and suppressing global interference signals such as joint impact, thus leading to poor reliability in belt conveyor monitoring. This invention effectively improves the reliability of belt conveyor monitoring.

[0033] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0034] It should be noted that, in the embodiments of this invention, when the relevant object information and other related data are used in specific products or technologies, permission or consent from the object is required, and the collection, use, and processing of the relevant data must comply with the relevant laws, regulations, and standards of the relevant countries and regions. In other words, if the embodiments of this invention involve data related to an object, it must be obtained with the object's authorization and consent, the authorization and consent of relevant departments, and in accordance with the relevant laws, regulations, and standards of the country and region. If personal information is involved in the embodiments, the acquisition of all personal information requires the individual's consent; if sensitive information is involved, the separate consent of the information subject is required. The embodiments also need to be implemented with the object's authorization and consent.

[0035] It should be noted that the terms "first," "second," etc., used in this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The implementations described in the following exemplary embodiments do not represent all implementations consistent with this disclosure.

[0036] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship. Example 1:

[0037] Please see Figure 1The present invention provides an intelligent monitoring method for belt conveyors based on distributed optical fiber sensing, comprising: Step 101: Real-time acquisition of Rayleigh scattering light signals from distributed optical fibers laid along the belt conveyor, and construction of a spatiotemporal vibration matrix based on the Rayleigh scattering light signals.

[0038] Belt conveyors are mechanical equipment used in industrial settings such as coal mines, ports, power plants, and cement plants to transport bulk materials such as coal, ore, and bulk materials over long distances and continuously. Core components include, but are not limited to, a ring conveyor belt, drive rollers, idler rollers, load-bearing idlers, and a frame.

[0039] Distributed optical fiber refers to a continuously laid single-mode or multimode communication optical cable that is passive and has no electronic components. It can be laid along the entire length of a belt conveyor frame and used as a sensing medium in a distributed acoustic sensing (DAS) system. When external vibrations act on the optical fiber, they cause weak phase or intensity modulation of the Rayleigh scattering light that transmits optical signals in the fiber. By demodulating and analyzing these modulated signals, vibration information at different locations along the length of the optical fiber can be reconstructed.

[0040] Distributed acoustic sensing (DAS) system refers to a fiber optic sensing system based on the principle of coherent optical time domain reflectometry (Φ-OTDR). This system periodically emits narrow-linewidth laser pulses into a distributed optical fiber and detects the phase or intensity changes of the backscattered Rayleigh light generated when the pulse light propagates in the fiber.

[0041] It is worth mentioning that the distributed acoustic sensing system is implemented by the Φ-OTDR / DAS host located in the ground control room. The host emits narrow pulse light with a repetition frequency of 2–10 kHz into the optical fiber to collect Rayleigh scattering signals along the line.

[0042] It should be noted that the deployment method of distributed optical fibers can be optimized according to the site conditions to ensure effective coupling and long-term reliability of vibration signals. Specifically, the following two typical deployment methods can be adopted: The first method is to arrange the fiber close to the idler frame: the fiber is laid along the side beam of the idler and fixed with clamps or stainless steel cable ties at intervals of 1-2m. To obtain the best vibration transmission effect, the fiber can be kept at a distance of about 3-10cm from the idler frame, so that the mechanical vibration generated by the conveyor belt and idler during operation can be efficiently coupled to the fiber.

[0043] The second method is to lay the fiber on the frame beam or structural components: This method is more suitable for large-span, long-distance conveyor scenarios. The fiber can be laid on the surface of the main beam or structural components of the conveyor, or it can be buried in a special protective groove or sleeve to avoid mechanical wear and environmental damage, thereby improving the durability of the deployment while ensuring signal quality.

[0044] Regardless of the deployment method, the optical fiber must cover the entire conveyor line, and the length can be adapted from 100 meters to 10 kilometers depending on the actual length of the conveyor.

[0045] In terms of fiber selection, general-purpose G.652 single-mode fiber can be used, or a special distributed sensing fiber optimized for vibration sensing can be selected to further ensure sensing performance and system signal-to-noise ratio.

[0046] Preferably, step 101 may include the following sub-steps: S11. Coherently demodulate the Rayleigh scattered light signal to obtain the demodulated signal.

[0047] Rayleigh scattering light signal refers to the backscattered light signal generated when a narrow pulse laser emitted by a distributed acoustic sensing system propagates through the optical fiber due to the non-uniform microstructure of the optical fiber medium.

[0048] Demodulated signal refers to the electrical signal obtained after coherent demodulation processing.

[0049] It should be noted that coherent demodulation is a process of optical interference between a local reference light of the same origin as the emitted laser and the received Rayleigh scattered light. The local reference light serves as a stable phase and frequency reference, and interferes with the scattered light carrying phase disturbances caused by vibrations, converting the weak phase changes into detectable light intensity changes. This interference light intensity is then converted into a raw electrical signal by a photodetector, and then filtered by a bandpass filter and a phase demodulation circuit. If I / Q quadrature demodulation technology is used, the low-frequency electrical signal component that is only related to vibration is finally extracted, which is the demodulated signal.

[0050] In this embodiment of the invention, coherent demodulation technology is used to process the Rayleigh scattered light signal to obtain a demodulated signal characterizing the vibration state along the line.

[0051] S12. Filter the demodulated signal to obtain the filtered vibration signal sequence.

[0052] A vibration signal sequence refers to a signal sequence obtained after filtering.

[0053] It should be noted that filtering is a crucial step in frequency screening of the demodulated signal to remove noise and retain effective vibration components. The demodulated signal typically contains various interference components, such as its own optical noise, electrical noise, and low-frequency oscillations of the conveyor structure or environmental background vibrations. By setting an appropriate passband frequency range, such as 20Hz to several kilohertz for belt joint impacts and abnormal roller vibrations, and using digital filters, such as Butterworth filters or FIR filters, to filter the demodulated signal, out-of-band noise and irrelevant low-frequency structural vibrations can be effectively removed.

[0054] In this embodiment of the invention, the demodulated signal is subjected to digital bandpass filtering to suppress low-frequency structural oscillation and high-frequency random noise, resulting in a vibration signal sequence. Each data point in the sequence corresponds to the vibration intensity at a fixed spatial location on the optical fiber at a certain sampling time.

[0055] S13. Reassemble the filtered vibration signal sequence and its corresponding spatial location with the acquisition time to construct a spatiotemporal vibration matrix.

[0056] The spatiotemporal vibration matrix refers to a two-dimensional data structure. The row index usually corresponds to a series of discrete spatial locations divided at fixed intervals along the fiber direction, and the column index corresponds to a series of continuous time points collected at fixed time intervals.

[0057] Reassembly refers to filling each vibration signal data point into the corresponding row and column coordinates of a two-dimensional matrix in an orderly manner, based on the spatial location and acquisition timestamp of each data point at the time of acquisition. This process organizes the originally dispersed one-dimensional signal sequence arranged in time flow into a complete two-dimensional data structure.

[0058] It should be noted that all the vibration signal sequences obtained after filtering are systematically arranged and filled into a predefined two-dimensional array structure according to their acquisition time and calculated optical fiber spatial position coordinates. Each row of the array represents a fixed spatial channel, i.e., a spatial sampling point, and each column represents a fixed sampling time. The spatiotemporal vibration matrix S(x, t) is constructed in this way, where x is the spatial coordinate along the optical fiber direction and t is time.

[0059] In this embodiment of the invention, the filtered vibration signal sequence and its corresponding spatial location and acquisition time are recombined to construct a spatiotemporal vibration matrix.

[0060] Step 102: Preprocess the spatiotemporal vibration matrix to obtain the spatiotemporal energy map.

[0061] Preferably, step 102 may include the following sub-steps: S21. Filter the spatiotemporal vibration matrix to obtain the filtered spatiotemporal vibration matrix.

[0062] The filtered spatiotemporal vibration matrix refers to the result obtained by processing the original spatiotemporal vibration matrix separately or simultaneously in both the spatiotemporal dimensions using digital filtering techniques.

[0063] It should be noted that, in the spatial dimension, a sliding window averaging or median filtering method is used to smooth each time slice of the spatiotemporal vibration matrix to suppress spatial burst noise caused by differences in fiber optic coupling or environmental noise; in the temporal dimension, the timing signal corresponding to each spatial channel is subjected to high-pass filtering, for example, 20-50Hz, to remove low-frequency structural vibration components caused by the overall swaying of the conveyor frame or foundation vibration.

[0064] In this embodiment of the invention, the spatiotemporal vibration matrix is ​​filtered in both the spatial and temporal dimensions. Through this joint filtering method, a filtered spatiotemporal vibration matrix with an effectively improved signal-to-noise ratio is obtained.

[0065] S22. Extract the envelope of the filtered spatiotemporal vibration matrix to obtain the envelope matrix.

[0066] The envelope matrix refers to the two-dimensional matrix that represents the instantaneous energy or amplitude change of a signal, extracted from the filtered spatiotemporal vibration matrix using the envelope detection method.

[0067] It is worth mentioning that the envelope signal can be obtained using Hilbert transform or short-time energy method. Specifically, for the filtered spatiotemporal vibration matrix... The Hilbert transform method calculates the analytic signal for each spatial channel along the time dimension and takes the modulus of the analytic signal to obtain the envelope matrix E(x,t):

[0068] The short-time energy method uses a sliding window in the time domain to calculate the sum of squares or absolute values ​​of signal samples within the window as an energy estimate at the center of the window.

[0069] It should be noted that both methods can effectively extract the envelope information of the signal. The Hilbert transform has higher time-frequency resolution, while the short-time energy method is simpler and more direct to calculate. The specific method can be designed according to needs.

[0070] It is worth mentioning that, in order to improve the detectability of weak joint impact signals under strong background noise, multi-scale feature extraction methods can also be introduced, specifically: For each spatial channel signal of the filtered spatiotemporal vibration matrix, time-frequency analysis is performed using methods such as wavelet packet transform, short-time Fourier transform, or continuous wavelet transform. By extracting or fusing energy at multiple scales related to the joint impact characteristics, a feature map highlighting a specific time-frequency pattern can be formed. For example, wavelet packet transform can be used to extract energy in a specific frequency band, or the amplitude of the continuous wavelet transform coefficients can be accumulated to obtain a multi-scale energy representation that better reveals the impact traces of weak joints. Finally, this multi-scale energy information is fused with the envelope information obtained through Hilbert transform or short-time energy method to generate an envelope matrix.

[0071] In this embodiment of the invention, the Hilbert transform is applied to the signal of each spatial channel along the time dimension of the filtered spatiotemporal vibration matrix to calculate the modulus of its analytic signal, thereby obtaining the envelope matrix characterizing the instantaneous amplitude of the signal.

[0072] S23. Perform energy normalization on the envelope matrix to obtain the spatiotemporal energy map.

[0073] A spatiotemporal energy map is a two-dimensional grayscale or pseudo-color image obtained by normalizing the envelope matrix, used to visually display the spatiotemporal distribution of vibrational energy. The brightness level in the image directly corresponds to the magnitude of the normalized vibrational energy.

[0074] The envelope matrix E(x,t) is subjected to energy normalization to suppress background noise and highlight the impact characteristics of the joint. Specifically, for each fixed spatial location, the mean energy value of all time points in that row is calculated independently. and standard deviation These two statistics characterize the background vibration energy level and its fluctuation range at this location under normal operating conditions, and each element is calculated using the following row-by-row Z-score normalization:

[0075] After this processing, the background noise of each spatial channel is normalized to a distribution with an approximate mean of 0 and a standard deviation of 1, while the high energy peaks generated by events such as joint impact will show a significant positive shift. In the generated spatiotemporal energy map, the trajectory formed by the periodic impact of the joint will appear as a clear, approximately linear high-brightness stripe, while non-target noise and background undulations in the image are greatly suppressed.

[0076] In this embodiment of the invention, the envelope matrix is ​​Z-score normalized row by row according to the spatial channel to obtain a spatiotemporal energy map with enhanced contrast and suppressed background.

[0077] Step 103: Based on the spatiotemporal energy map, determine the linear trajectory parameters of the conveyor belt joint movement in the belt conveyor, and determine the real-time running speed of the conveyor belt based on the slope in the linear trajectory parameters.

[0078] The linear trajectory parameters of the conveyor belt joint movement refer to the parameters of the mathematical model followed by the sequence of high-energy points formed by the periodic impact of the conveyor belt joint in the space-time coordinate system in the spatiotemporal energy map.

[0079] Preferably, step 103 may include the following sub-steps: S31. Extract high-energy candidate points from the spatiotemporal energy map to obtain a candidate point set.

[0080] Preferably, S31 may include the following sub-steps: S1. Determine the adaptive threshold based on the background noise level of the spatiotemporal energy map.

[0081] Background noise level refers to the level in the spatiotemporal energy map. In this context, the statistical characteristics correspond to a relatively stable and widely distributed energy base formed by factors such as environmental background vibration, inherent system noise, and random vibrations from non-joint impacts.

[0082] Adaptive threshold refers to a numerical threshold dynamically calculated from the real-time statistical characteristics of the background noise level, used to distinguish significant high-energy events in the energy map from background noise.

[0083] It should be noted that since the spatiotemporal energy map has been processed by row-by-row Z-score normalization, the background noise at each spatial location theoretically follows a distribution with a mean of 0 and a standard deviation of 1. Therefore, a uniform adaptive threshold T can be set for all spatial locations. This threshold is a constant selected based on experience, and T can take values ​​between 3 and 5, indicating that only significant points with energy values ​​exceeding k times the standard deviation of the background noise distribution are retained. Alternatively, different thresholds can be set for spatial locations in different segments based on the actual signal-to-noise ratio.

[0084] In this embodiment of the invention, an adaptive threshold is determined based on the background noise level of the spatiotemporal energy map.

[0085] S2. Compare the energy values ​​of each spatiotemporal coordinate point in the spatiotemporal energy map with the adaptive threshold, and extract the spatiotemporal coordinate points with energy values ​​greater than the adaptive threshold to construct a candidate point set.

[0086] The energy value of each spatiotemporal coordinate point refers to the matrix element value corresponding to a specific spatial location x and a specific time point t in the spatiotemporal energy map.

[0087] The candidate point set refers to a set of spatiotemporal coordinate pairs, where each coordinate pair (x,t) corresponds to a position in the spatiotemporal energy map where the energy value exceeds the adaptive threshold T.

[0088] It is understandable that traversing the spacetime energy map... For each element, compare its energy value with the threshold T, and select all elements that meet the condition. The points are extracted to form a candidate point set, that is:

[0089] In this embodiment of the invention, each coordinate point of the spatiotemporal energy map is traversed, and it is determined whether its energy value is greater than the adaptive threshold. All coordinate points that meet the conditions are extracted to form a candidate point set.

[0090] S32. Perform trajectory model fitting processing on the candidate point set with outlier removal to obtain the optimized trajectory model.

[0091] Outlier removal refers to the process of identifying and excluding abnormal data points that do not conform to the distribution pattern of the main data before or during the fitting of a mathematical model using a set of candidate points. These outliers are usually caused by random strong noise from non-joint impacts, other transient vibration events, or detection errors.

[0092] Fitting refers to the process of finding a continuous mathematical model from a set of discrete observation data points that best describes the overall variation of these data points.

[0093] A trajectory model is a mathematical model used to describe the motion of a conveyor belt joint in a space-time diagram. Specifically, the trajectory model is a straight line in the space-time (xt) plane, and the equation can be expressed as x = vt + b, where v is the slope of the line, i.e., the velocity, and b is the intercept, i.e., the initial position.

[0094] It should be noted that, firstly, the optimal straight line model is searched in the parameter space using Hough transform, Radon transform, or multi-channel time delay correlation method to obtain the initial trajectory parameters. Then, using these initial parameters as a reference, robust straight line fitting is performed using the RANSAC algorithm or M-estimators estimation method to remove outliers that do not conform to the main distribution, thereby obtaining the optimized straight line trajectory model. The fitting confidence γ of the trajectory is output simultaneously, such as the proportion of inliers or the fitting residual index.

[0095] It is worth mentioning that if the conveyor has multiple conveyor belt joints, cluster analysis or multi-target tracking can be performed on the candidate point set to separate the candidate points belonging to different joints into multiple subsets, and the above trajectory fitting process can be executed separately to extract multiple linear trajectory parameters.

[0096] In this embodiment of the invention, the candidate point set is processed by Hough transform to obtain initial trajectory parameters. With these initial values ​​as a reference, the RANSAC algorithm is used for robust straight line fitting. After removing outliers, an optimized straight line trajectory model is obtained.

[0097] S33. Extract linear trajectory parameters from the optimized trajectory model. The linear trajectory parameters include the slope, which serves as the initial belt speed estimate.

[0098] Linear trajectory parameters refer to the set of mathematical variables that uniquely determine the optimized trajectory model.

[0099] The initial belt speed estimate refers to the preliminary estimate of the conveyor belt running speed obtained directly from the single trajectory fitting result.

[0100] In this embodiment of the invention, the slope of the optimized trajectory model is directly read as the initial belt speed estimate at the current moment.

[0101] S34. Smooth the initial belt speed estimate to obtain the real-time operating speed.

[0102] Real-time operating speed refers to the belt speed value, denoted as v(t), which is output after smoothing and filtering and can stably and reliably reflect the macroscopic average operating state of the conveyor belt.

[0103] It is worth mentioning that the initial belt speed estimates at multiple consecutive moments can be smoothed by using sliding time window averaging, Kalman filtering, etc. In the case of using Kalman filtering, the initial belt speed estimate is used as the observation input of Kalman filter. The filter maintains a state estimate of the belt speed and assumes that the speed change follows a stationary process. The filter recursively updates its state estimate according to the preset process noise covariance and observation noise covariance, and calculates the optimal speed estimate under all current historical observation conditions, that is, the real-time running speed v(t).

[0104] If the sliding time window averaging method is used, the arithmetic mean of all initial belt speed estimates within the most recent period, such as the past 10 seconds, can be calculated as the current real-time operating speed v(t) output.

[0105] It is worth mentioning that, based on the trajectory fitting confidence and error covariance information during the smoothing process, an estimated confidence index corresponding to the real-time running speed can also be output. This is used to reflect the reliability of the current speed value.

[0106] It is worth mentioning that the real-time running speed is updated at a frequency of 0.5 to 5 Hz, and each speed update outputs a corresponding confidence index.

[0107] In this embodiment of the invention, the initial belt speed estimate is smoothed by sliding time window averaging or Kalman filtering to obtain the real-time running speed. This effectively suppresses speed value fluctuations caused by random noise, instantaneous measurement errors or local abnormal events in single trajectory fitting, ensuring that the final output real-time running speed has high stability and reliability.

[0108] Step 104: Based on the intercept and real-time running speed in the linear trajectory parameters, predict the time for the conveyor belt joint to pass through each spatial position of the optical fiber, and generate a dynamic spatiotemporal mask matrix based on the predicted time.

[0109] Preferably, step 104 may include the following sub-steps: S41. Based on the intercept and real-time operating speed, calculate the predicted arrival time of the conveyor belt joint at each spatial location of the optical fiber.

[0110] Predicted arrival time refers to the estimated time when the connector will arrive at any specified spatial location on the optical fiber, calculated based on the identified connector trajectory model and the real-time measured belt speed.

[0111] Based on the intercept b and real-time running speed v(t) in the linear trajectory parameters, the predicted arrival time of the conveyor belt joint at each spatial position on the optical fiber is calculated. Within a short prediction window, the conveyor belt speed can be considered to remain stable, and the joint motion follows the identified linear model. Therefore, substituting b and v(t) into the trajectory linear equation x = v(t) × t + b, the predicted arrival time of the conveyor belt joint at any spatial position on the optical fiber can be calculated. :

[0112] In the formula, x represents the coordinates of each spatial sampling point discretized along the fiber direction, i = 1, 2, N, N is the total number of spatial channels, b is the optimized intercept, and v(t) is the real-time running speed.

[0113] In this embodiment of the invention, based on the trajectory parameters and real-time speed obtained at the current moment, the predicted arrival time corresponding to each sampling position along the entire optical fiber is calculated sequentially, thereby realizing the prediction and dynamic tracking of the joint movement.

[0114] S42. For each spatial location, determine the time tolerance range centered on the corresponding predicted arrival time.

[0115] Spatial location refers to the coordinates of discrete sampling points along the length of the optical fiber, divided by a fixed spatial resolution. .

[0116] Time tolerance range refers to the range around the predicted arrival time on the time axis. A symmetrical time window with a certain width is set to accommodate prediction errors and signal duration, ensuring that the joint impact signal can be effectively covered.

[0117] Considering factors such as belt speed estimation error, time broadening of joint impact signals, and system sampling time resolution, a time resolution is set for each spatial location. The time tolerance interval centered on this point can be expressed as:

[0118] In the formula, The preset time tolerance parameter can be set according to actual working conditions, such as belt speed fluctuation range and joint impact duration. For example, take... =0.1-0.5 s.

[0119] In this embodiment of the invention, by setting a dynamic time tolerance interval for each spatial location, the actual arrival time window of the joint impact signal is adaptively covered, thereby improving the robustness and accuracy of signal shielding.

[0120] S43. Based on the spatial location and the corresponding predicted arrival time and time tolerance range, construct a dynamic spatiotemporal mask matrix.

[0121] The dynamic spatiotemporal mask matrix refers to a two-dimensional matrix with the same dimensions as the original spatiotemporal vibration matrix.

[0122] It is understandable that a binary mask matrix M(x,t) with the same dimensions as the spatiotemporal vibration matrix S(x,t) can be constructed. For each element in the matrix, if its time coordinate t falls on the corresponding spatial position... If the tolerance range is within a certain range, then the mask value at that position is set to 0 to indicate that it needs to be suppressed; otherwise, it is set to 1 to indicate that it needs to be retained. The mathematical expression is as follows:

[0123] The mask matrix M(x,t) is the dynamic spatiotemporal mask matrix, which is dynamically updated according to the joint's motion trajectory and real-time belt speed.

[0124] In this embodiment of the invention, a two-dimensional matrix with the same dimensions as the original spatiotemporal vibration matrix is ​​constructed, which can accurately identify the region where the joint impact signal is located at the spatiotemporal data level.

[0125] Step 105: Perform signal cleaning on the spatiotemporal vibration matrix based on the dynamic spatiotemporal mask matrix to obtain the cleaned spatiotemporal vibration matrix.

[0126] Preferably, step 105 may include the following sub-steps: S51. Perform dot product operations at corresponding positions on the dynamic spatiotemporal mask matrix and the spatiotemporal vibration matrix to obtain the cleaned spatiotemporal vibration matrix.

[0127] The cleaned spatiotemporal vibration matrix refers to the spatiotemporal vibration data matrix after processing by a dynamic spatiotemporal mask matrix. In this matrix, the amplitude of the spatiotemporal region where the joint impact signal is located is suppressed, while the vibration signal in the non-joint region is completely preserved, thus obtaining a clean vibration data that is more suitable for subsequent fault diagnosis.

[0128] The dynamic spatiotemporal mask matrix M(x,t) and the original spatiotemporal vibration matrix S(x,t) are multiplied by the following formula:

[0129] In the formula, This is the cleaned spatiotemporal vibration matrix.

[0130] In this embodiment of the invention, the dynamic spatiotemporal mask matrix and the spatiotemporal vibration matrix are multiplied at corresponding positions, so that the signal within the predicted arrival time window of the joint is effectively suppressed, while the vibration information at other spatiotemporal positions remains unchanged. While retaining the full-line vibration monitoring capability, the interference caused by the joint impact on the downstream state diagnosis algorithm is reduced, thereby improving the monitoring reliability of the belt conveyor.

[0131] Step 106: Diagnose the operating status of the belt conveyor based on the spatiotemporal vibration matrix after cleaning.

[0132] The cleaned spatiotemporal vibration matrix is ​​input into a preset conveyor condition diagnosis algorithm to identify and assess key components and operational anomalies of the belt conveyor, including but not limited to: 1. Idler condition diagnosis: By analyzing the time-frequency characteristics, energy distribution, or abnormal peak values ​​of vibration signals at various spatial locations, faults such as idler damage, jamming, and poor lubrication of the idler can be identified.

[0133] 2. Conveyor belt anomaly detection: Detecting for problems with the conveyor belt itself, such as tearing, misalignment, and abnormal joints; monitoring the drive system.

[0134] 3. Based on the belt speed information and vibration characteristics, determine whether there are slippage, abnormal vibration, or other phenomena in the drive drum and reducer; conduct an overall operation assessment.

[0135] It should be noted that the preset conveyor condition diagnosis algorithm is a known fault diagnosis method based on vibration signal analysis in this field, including but not limited to time domain feature analysis: extracting indicators such as the effective value, peak value, kurtosis, and impulse factor of the vibration signal, and identifying anomalies through threshold comparison or trend analysis; frequency domain analysis: performing fast Fourier transform or power spectral density analysis on the vibration signal to detect whether there are characteristic frequency components corresponding to roller damage, bearing failure, etc.; training a classifier based on historical data to perform pattern recognition of vibration features to achieve automatic fault classification and early warning, etc. No specific limitations are made here, as long as the corresponding functions can be achieved.

[0136] In this embodiment of the invention, the vibration data after cleaning suppresses the interference of high-intensity joint impact signals. Fault diagnosis based on this cleaned signal can reduce the false alarm rate and improve the reliability and practicality of belt conveyor monitoring.

[0137] Example 2: Please see Figure 2 The present invention provides an intelligent monitoring system for belt conveyors based on distributed optical fiber sensing, comprising: The data acquisition and construction module 101 is used to acquire Rayleigh scattering light signals from distributed optical fibers laid along the belt conveyor in real time, and to construct a spatiotemporal vibration matrix based on the Rayleigh scattering light signals.

[0138] The signal preprocessing module 102 is used to preprocess the spatiotemporal vibration matrix to obtain the spatiotemporal energy map.

[0139] The trajectory parameter determination module 103 is used to determine the linear trajectory parameters of the conveyor belt joint movement in the belt conveyor based on the spatiotemporal energy map, and to determine the real-time running speed of the conveyor belt based on the slope in the linear trajectory parameters.

[0140] The dynamic mask generation module 104 is used to predict the time it takes for the conveyor belt joint to pass through each spatial position of the optical fiber based on the intercept and real-time running speed in the linear trajectory parameters, and to generate a dynamic spatiotemporal mask matrix based on the predicted time.

[0141] The signal cleaning module 105 is used to clean the spatiotemporal vibration matrix based on the dynamic spatiotemporal mask matrix to obtain the cleaned spatiotemporal vibration matrix.

[0142] The operation status diagnosis module 106 is used to diagnose the operation status of the belt conveyor based on the spatiotemporal vibration matrix after cleaning.

[0143] Since the above is a system corresponding to a one-to-one intelligent monitoring method for belt conveyors based on distributed optical fiber sensing, and its implementation principle is the same as that of an intelligent monitoring method for belt conveyors based on distributed optical fiber sensing, for the sake of convenience and brevity, those skilled in the art can clearly understand that the specific working process of the system and modules described above can be referred to the corresponding process in the aforementioned method embodiments, and will not be repeated here.

[0144] Example 3: An electronic device according to an embodiment of the present invention includes: a memory and a processor, wherein the memory stores a computer program; when the computer program is executed by the processor, the processor performs the intelligent monitoring method for belt conveyors based on distributed optical fiber sensing as described in any of the above embodiments.

[0145] The memory can be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. The memory has storage space for program code used to perform any of the method steps described above. For example, the storage space for program code may include individual program codes for implementing the various steps in the methods described above. This program code can be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, compact discs (CDs), memory cards, or floppy disks. The program code may be compressed, for example, in a suitable form. When run by a computing processing device, this code causes the computing processing device to perform the various steps in the methods described above.

[0146] Example 4: This invention provides a computer-readable storage medium storing a computer program thereon, which, when executed, implements the intelligent monitoring method for belt conveyors based on distributed optical fiber sensing according to any of the above embodiments.

[0147] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0148] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0149] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0150] Furthermore, the functional units in the various embodiments of the present invention 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 unit.

[0151] If the integrated unit is implemented as a software functional unit and 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 the present invention, 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.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. 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.

[0152] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for intelligent monitoring of belt conveyors based on distributed optical fiber sensing, characterized in that, include: Rayleigh scattering light signals from distributed optical fibers laid along the belt conveyor are acquired in real time, and a spatiotemporal vibration matrix is ​​constructed based on the Rayleigh scattering light signals. The spatiotemporal vibration matrix is ​​preprocessed to obtain a spatiotemporal energy map; Based on the spatiotemporal energy map, the linear trajectory parameters of the conveyor belt joint movement in the belt conveyor are determined, and the real-time operating speed of the conveyor belt is determined based on the slope in the linear trajectory parameters. Based on the intercept in the linear trajectory parameters and the real-time running speed, the time for the conveyor belt joint to pass through each spatial position of the optical fiber is predicted, and a dynamic spatiotemporal mask matrix is ​​generated based on the predicted time. Based on the dynamic spatiotemporal mask matrix, the spatiotemporal vibration matrix is ​​cleaned to obtain the cleaned spatiotemporal vibration matrix; The operating status of the belt conveyor is diagnosed based on the spatiotemporal vibration matrix after cleaning.

2. The intelligent monitoring method for belt conveyors based on distributed optical fiber sensing according to claim 1, characterized in that, The construction of the spatiotemporal vibration matrix based on the Rayleigh scattered light signal includes: The Rayleigh scattered light signal is coherently demodulated to obtain a demodulated signal; The demodulated signal is filtered to obtain a filtered vibration signal sequence; The filtered vibration signal sequence and its corresponding spatial location are recombined with the acquisition time to construct the spatiotemporal vibration matrix.

3. The intelligent monitoring method for belt conveyors based on distributed optical fiber sensing according to claim 1, characterized in that, The preprocessing of the spatiotemporal vibration matrix to obtain the spatiotemporal energy map includes: The spatiotemporal vibration matrix is ​​filtered to obtain the filtered spatiotemporal vibration matrix; The envelope of the filtered spatiotemporal vibration matrix is ​​extracted to obtain the envelope matrix. The envelope matrix is ​​subjected to energy normalization to obtain the spatiotemporal energy map.

4. The intelligent monitoring method for belt conveyors based on distributed optical fiber sensing according to claim 1, characterized in that, The process of determining the linear trajectory parameters of the conveyor belt joint movement in the belt conveyor based on the spatiotemporal energy map, and determining the real-time operating speed of the conveyor belt based on the slope of the linear trajectory parameters, includes: High-energy candidate points are extracted from the spatiotemporal energy map to obtain a candidate point set; The candidate point set is subjected to trajectory model fitting processing with outlier removal to obtain the optimized trajectory model; Linear trajectory parameters are extracted from the optimized trajectory model, including the slope as an initial belt velocity estimate. The initial belt speed estimate is smoothed to obtain the real-time operating speed.

5. The intelligent monitoring method for belt conveyors based on distributed optical fiber sensing according to claim 4, characterized in that, The step of extracting high-energy candidate points from the spatiotemporal energy map to obtain a candidate point set includes: An adaptive threshold is determined based on the background noise level of the spatiotemporal energy map. The energy values ​​of each spatiotemporal coordinate point in the spatiotemporal energy map are compared with the adaptive threshold, and spatiotemporal coordinate points with energy values ​​greater than the adaptive threshold are extracted to construct the candidate point set.

6. The intelligent monitoring method for belt conveyors based on distributed optical fiber sensing according to claim 1, characterized in that, The step of predicting the time it takes for the conveyor belt joint to pass through each spatial position of the optical fiber based on the intercept in the linear trajectory parameters and the real-time running speed, and generating a dynamic spatiotemporal mask matrix based on the predicted time, includes: Based on the intercept and the real-time operating speed, the predicted arrival time of the conveyor belt joint at each spatial position of the optical fiber is calculated. For each of the aforementioned spatial locations, a time tolerance interval is determined with the corresponding predicted arrival time as the center. A dynamic spatiotemporal mask matrix is ​​constructed based on the spatial location, the corresponding predicted arrival time, and the time tolerance range.

7. The intelligent monitoring method for belt conveyors based on distributed optical fiber sensing according to claim 1, characterized in that, The step of performing signal cleaning on the spatiotemporal vibration matrix based on the dynamic spatiotemporal mask matrix to obtain the cleaned spatiotemporal vibration matrix includes: The dynamic spatiotemporal mask matrix and the spatiotemporal vibration matrix are multiplied at corresponding positions to obtain the cleaned spatiotemporal vibration matrix.

8. A smart monitoring system for belt conveyors based on distributed optical fiber sensing, characterized in that, include: The data acquisition and construction module is used to acquire Rayleigh scattering light signals from distributed optical fibers laid along the belt conveyor in real time, and to construct a spatiotemporal vibration matrix based on the Rayleigh scattering light signals. The signal preprocessing module is used to preprocess the spatiotemporal vibration matrix to obtain a spatiotemporal energy map; The trajectory parameter determination module is used to determine the linear trajectory parameters of the conveyor belt joint movement in the belt conveyor based on the spatiotemporal energy map, and to determine the real-time operating speed of the conveyor belt based on the slope in the linear trajectory parameters. The dynamic mask generation module is used to predict the time it takes for the conveyor belt joint to pass through each spatial position of the optical fiber based on the intercept in the linear trajectory parameters and the real-time running speed, and to generate a dynamic spatiotemporal mask matrix based on the predicted time. The signal cleaning module is used to clean the spatiotemporal vibration matrix based on the dynamic spatiotemporal mask matrix to obtain the cleaned spatiotemporal vibration matrix. The operation status diagnosis module is used to diagnose the operation status of the belt conveyor based on the spatiotemporal vibration matrix after cleaning.

9. An electronic device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described in any one of claims 1 to 7, which is a method for intelligent monitoring of belt conveyors based on distributed optical fiber sensing.

10. A computer-readable storage medium, characterized in that, The computer program is stored and can be loaded by a processor and executed as described in any one of claims 1 to 7, which is a method for intelligent monitoring of belt conveyors based on distributed optical fiber sensing.

Images