Suspended cable belt conveyor wind vibration monitoring method and system based on machine learning

Abstract

The invention discloses a suspended cable belt conveyor wind vibration monitoring method and system based on machine learning, and relates to the field of intelligent monitoring of belt conveyors. According to the invention, wind vibration observation and prediction are carried out on the within-span marker post of the suspension cable based on the high-definition camera and various sensors, so that an important auxiliary role is played in ensuring safe operation, danger early warning and emergency control of the suspension cable belt conveyor in conveying work; according to the invention, the method can achieve the monitoring of a few persons, reduces the operation and maintenance cost, can improve the intelligent degree of the suspended cable belt conveyor, improves the automation degree and accuracy of the monitoring process, and guarantees the safe operation of the suspended cable belt conveyor under the condition of wind vibration.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring of belt conveyors, and in particular to a method and system for wind vibration monitoring of a suspended belt conveyor based on machine learning. Background Art

[0002] The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.

[0003] In areas of my country with harsh terrain conditions, the transportation method for materials such as ores usually adopts the solution of curved belt conveyors or circular belt conveyors. However, no matter which method is adopted, the layout of the conveyor line will be very difficult, and it is inevitable to use extremely high support towers and large-span trusses, and it may also be necessary to dig tunnels and build bridges.

[0004] For situations with more complex terrain conditions, there is also a cable-suspended belt conveyor solution. However, during the conveying process of this conveyor, due to factors such as materials, belt tension, wind, rain, and snow, the suspension cable will experience a "pendulum"-like displacement change. Because its structure is different from that of a conventional belt conveyor, there is no readily available monitoring system, and excessive displacement may cause abnormal conditions such as material spillage and structural stress changes. These hidden dangers will not only affect the normal operation of the belt conveyor, but in serious cases may cause economic losses and even lead to safety accidents.

[0005] Machine learning technology, through its powerful nonlinear fitting ability, builds a mapping relationship between input and output, and can realize the analysis and prediction of multiple factors and multiple results. The development of machine learning technology has broadened the application scenarios of intelligent monitoring and provided a theoretical basis for wind vibration monitoring of cable-belt conveyors. Summary of the invention

[0006] The purpose of the present invention is to provide a method and system for wind vibration monitoring of a suspension belt conveyor based on machine learning in response to the problems existing in the prior art, thereby solving the problem of intelligent wind vibration monitoring of a suspension belt conveyor.

[0007] The technical solution of the present invention is as follows:

[0008] The wind vibration monitoring method of the cable-belt conveyor based on machine learning includes:

[0009] Step S1: Data collection and transmission; automatically collect videos with target detection benchmarks, environmental data and cable-belt conveyor operation data, and store and transmit them in chronological order; and transmit them to the local server in real time, and the remote server retrieves the data with the time axis as needed;

[0010] Step S2: image correction and data processing; determine the video frame interval and intercept the time series data corresponding to the video frame; calibrate the internal parameters of the camera and measure the external parameters in real time to complete the correction of the video frame; pre-process the obtained data fragments to obtain time domain features and time-frequency domain features;

[0011] Step S3: Model establishment and mapping construction; establish a target detection model, and obtain the distance to the benchmark and the wind vibration offset at the benchmark; establish a wind vibration prediction model based on the collected data and the real-time detection results of the cable belt conveyor; construct a mapping relationship between the wind vibration data at the benchmark and the wind vibration data at the maximum wind vibration point;

[0012] Step S4: automatic monitoring and early warning; real-time acquisition of wind vibration data calculated by the target detection model, and comparison with the wind vibration threshold analysis; timely notification of maintenance personnel or intervention control to slow down or stop the cable-belt conveyor before the abnormal wind vibration time point predicted by the deep learning model is reached.

[0013] Furthermore, the target detection pole is a long strip mark, which needs to be set to a fixed length according to the detection distance, and the color should be different from the ambient color, and is placed horizontally in a span of the suspension belt conveyor.

[0014] Furthermore, the required length or width of the target detection pole is selected according to the following formula:

[0015]

[0016] in:

[0017] l is the length of the target detection pole;

[0018] α is the ratio of the benchmark to the screen;

[0019] f is the focal length;

[0020] w cam is the actual width of the camera image sensor;

[0021] D is the estimated distance between the pole and the camera.

[0022] Furthermore, the step S1 comprises:

[0023] Step S1-1: acquiring running video data of a cable-belt conveyor with a target detection pole through a camera and transmitting the data;

[0024] Step S1-2: Obtain wind direction and wind speed data through an anemometer and store them in chronological order; obtain wind and rain load information through a rain and snow sensor and store and transmit them in chronological order;

[0025] Step S1-3: Obtain the vibration data of the cable belt conveyor through the vibration sensor and store it in chronological order; obtain the conveyor belt speed through the tachometer, and store and transmit it in chronological order.

[0026] Furthermore, the step S2 comprises:

[0027] Step S2-1: Select a suitable video frame interval according to the number of frames of the monitoring video; intercept the data collected by the anemometer, rain and snow sensor, vibration sensor and tachometer into time sequence segments corresponding to the time interval of the video frames;

[0028] Step S2-2: Determine the internal parameters of the camera, and obtain the focal length, pixel size, optical axis principal point coordinates and camera distortion parameters of the camera through the Zhang Zhengyou calibration method; and measure the external parameters of the camera in real time, that is, the rotation matrix and the translation matrix;

[0029] Step S2-3: pre-process the intercepted time series data, the anemometer, rain and snow sensor and tachometer obtain their time-frequency characteristics, and the vibration sensor obtains its time-frequency domain characteristics.

[0030] Furthermore, the step S3 comprises:

[0031] Step S3-1: Establish a target detection model and train it based on the benchmark dataset; pass the video frame into the model to obtain the benchmark distance and the wind vibration offset at the benchmark.

[0032] Step S3-2: Based on the time series data of wind direction and speed, rain and snow load, and cable-belt conveyor operation, a wind-induced vibration prediction model is established and trained until the evaluation index meets the requirements;

[0033] Step S3-3: According to the coordinates of the two cable towers and the benchmark, a mapping relationship between the coordinates of the theoretical maximum wind vibration position of the cable-belt conveyor and its wind vibration data and the wind vibration data at the benchmark is constructed.

[0034] Furthermore, the step S4 comprises:

[0035] Step S4-1: Establish an autonomous monitoring algorithm to receive video data in real time; use the target detection model to calculate the theoretical maximum wind vibration position coordinates and wind vibration data, and compare and analyze them with the danger judgment threshold to judge the wind vibration abnormality in real time;

[0036] Step S4-2: receiving environmental conditions and operation data of the suspension belt conveyor in real time, predicting and calculating the wind vibration danger time point and its wind vibration data;

[0037] Step S4-3: Build an automated early warning and control system; when abnormal wind vibration is observed or predicted, the early warning control function is automatically enabled, and the early warning information is pushed to the on-duty personnel. In an emergency, the conveyor control system is intervened to achieve speed reduction or shutdown.

[0038] The present invention also proposes a suspension belt conveyor wind vibration monitoring system based on machine learning, which is used to implement the above-mentioned suspension belt conveyor wind vibration monitoring method, including:

[0039] Data acquisition module, information processing module, wind vibration monitoring module, early warning control module;

[0040] The data acquisition module includes: an anemometer, a rain and snow sensor, a high-definition camera, a vibration sensor and a tachometer; it is used to collect environmental data of wind, rain and snow on site, videos with benchmarks and operation data of the cable-belt conveyor, and transmit the data to the information processing module;

[0041] The information processing module includes: a lens distortion correction algorithm, a fast Fourier transform algorithm, a time domain feature extraction method, and a time-frequency domain feature extraction method; it is used to process various types of data transmitted by the receiving data acquisition module and transmit the data to the wind vibration monitoring module;

[0042] The wind vibration monitoring module includes: machine vision algorithm, ranging offset algorithm, wind vibration prediction algorithm, and maximum wind vibration calculation method; it is used to receive various data transmitted by the data acquisition module, establish and train the model; observe and predict the wind vibration data and abnormal time points of the suspension belt conveyor, and transmit the data to the early warning control module;

[0043] The early warning control module includes: an autonomous monitoring algorithm, an alarm push device, an automatic control algorithm, and a control linkage system; it is used to receive wind vibration data transmitted by the wind vibration monitoring module, monitor the operation of the cable-belt conveyor in real time, and realize autonomous monitoring and early warning.

[0044] Furthermore, the machine vision algorithm is used to identify the benchmark and obtain its position information in the image; the ranging offset algorithm is used to calculate the distance from the benchmark to the camera and obtain the wind vibration offset of the benchmark relative to the static position; the wind vibration prediction algorithm is used to predict the future wind vibration data of the benchmark under the current data and obtain the dangerous time point; the maximum wind vibration calculation method is used to calculate the position information and wind vibration data of the benchmark across the maximum wind vibration point.

[0045] Furthermore, the autonomous monitoring algorithm is used to monitor the operating status of the suspension belt conveyor in real time, and to determine whether the wind vibration data of the suspension belt conveyor exceeds a threshold; the automatic control algorithm is used to start the early warning device to remind the on-duty personnel to reduce the conveying volume, slow down or make an emergency stop, and intervene to control the suspension belt conveyor in an emergency.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1. The present invention is based on high-definition cameras and various sensors to observe and predict the wind vibration of the benchmark within the cable span, which plays an important auxiliary role in ensuring the safe operation, danger warning and emergency control of the cable belt conveyor during transportation work.

[0048] 2. The present invention uses an anemometer, rain and snow sensor, high-definition camera, vibration sensor and tachometer to ensure real-time acquisition of operating conditions and operating data of the suspension belt conveyor, thereby achieving less-managed monitoring and reducing operation and maintenance costs.

[0049] 3. The present invention realizes real-time analysis of wind vibration data of suspension belt conveyor and predicts abnormal time point of wind vibration through machine vision algorithm, distance measurement offset algorithm, wind vibration prediction algorithm and maximum wind vibration calculation method, thereby improving the intelligence level of suspension belt conveyor.

[0050] 4. The present invention realizes real-time monitoring of the operation of the cable-belt conveyor through autonomous monitoring algorithms, automatic control algorithms, and alarm push equipment, and can provide timely early warning and emergency control when wind vibration is abnormal, thereby improving the automation and intelligence of the monitoring process. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a flow chart of a method for monitoring wind vibration of a cable-belt conveyor based on machine learning proposed by the present invention;

[0052] Figure 2 It is a suspension belt conveyor wind vibration monitoring system based on machine learning proposed by the present invention, which is a structural diagram for realizing the wind vibration monitoring method;

[0053] Figure 3 It is a schematic diagram of the implementation layout of a suspension belt conveyor wind vibration monitoring system based on machine learning proposed by the present invention;

[0054] Figure 4 It is a schematic diagram of the camera arrangement of the wind vibration monitoring system of the cable-belt conveyor of the present invention;

[0055] Figure 5 It is a schematic diagram of a benchmark in a video frame of the suspension belt conveyor wind vibration monitoring system of the present invention;

[0056] Figure 6 It is a schematic diagram of the wind vibration offset between the current position and the marked position of the benchmark of the suspension belt conveyor wind vibration monitoring system of the present invention.

[0057] in, Figure 3 to Figure 4 The corresponding relationship between the reference numerals and the component names is as follows:

[0058] 1-wire rope; 2-cable frame; 3-roller group; 4-tower; 5-target detection pole; 6-camera. DETAILED DESCRIPTION

[0059] It should be noted that relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0060] The features and performance of the present invention are further described in detail below in conjunction with the embodiments.

[0061] Embodiment 1

[0062] The present invention provides a method for monitoring wind vibration of a cable-belt conveyor based on machine learning, such as Figure 1 ,include:

[0063] Step S1: Data collection and transmission. Automatically collect videos with target detection benchmarks, environmental data such as wind, rain, snow, and cable-belt conveyor operation data, and store and transmit them in chronological order.

[0064] Step S1 specifically includes:

[0065] Step S1-1: Obtain and transmit the running video data of the cable-belt conveyor with the target detection benchmark through a camera.

[0066] Step S1-2: Obtain wind direction and wind speed data through an anemometer and store them in chronological order; obtain wind and rain load information through a rain and snow sensor and store and transmit them in chronological order.

[0067] Step S1-3: Obtain the vibration data of the cable belt conveyor through the vibration sensor and store it in chronological order; obtain the conveyor belt speed through the tachometer and store and transmit it in chronological order.

[0068] Step S2: Image correction and data processing. Determine the video frame interval and intercept the time series data corresponding to the video frame; calibrate the internal parameters of the camera and measure the external parameters in real time to complete the correction of the video frame; pre-process the obtained data fragments to obtain time domain features and time-frequency domain features.

[0069] Step S2 specifically includes:

[0070] Step S2-1: Select a suitable video frame interval according to the number of frames of the monitoring video; intercept the data collected by the anemometer, rain and snow sensor, vibration sensor and tachometer into time-series segments corresponding to the time interval of the video frames.

[0071] Step S2-2: Measure the internal parameters of the camera, and obtain the focal length, pixel size, optical axis principal point coordinates and camera distortion parameters of the camera through the Zhang Zhengyou calibration method; and measure the external parameters of the camera in real time, namely the rotation matrix and translation matrix.

[0072] Step S2-3: pre-process the intercepted time series data, the anemometer, rain and snow sensor and tachometer obtain their time domain characteristics, and the vibration sensor obtains its time-frequency domain characteristics.

[0073] Step S3: Factor analysis and wind vibration prediction: A deep learning model of wind vibration data and factors such as wind direction and speed, rain and snow load, and cable-belt conveyor operation data is established, and based on this, abnormal wind vibration time points and abnormal wind vibration data of the cable-belt conveyor are predicted.

[0074] Step S3 specifically includes:

[0075] Step S3-1: Establish a target detection model and train it based on the benchmark dataset; pass the video frame into the model to obtain the benchmark distance and the wind vibration offset at the benchmark.

[0076] Step S3-2: Based on the time series data of wind direction and speed, rain and snow load, and cable-belt conveyor operation, a wind-induced vibration prediction model is established and trained until the evaluation index meets the requirements.

[0077] Step S3-3: According to the coordinates of the two cable towers and the benchmark, a mapping relationship between the coordinates of the theoretical maximum wind vibration position of the cable-belt conveyor and its wind vibration data and the wind vibration data at the benchmark is constructed.

[0078] Step S4: Automatic monitoring and early warning. Obtain the wind vibration data calculated by the model in real time and compare it with the wind vibration threshold value; before the abnormal wind vibration time point, promptly notify the maintenance personnel or intervene to control the suspension belt conveyor to slow down or stop.

[0079] Step S4 specifically includes:

[0080] Step S4-1: Establish an autonomous monitoring algorithm to receive video data in real time; use the target detection model to calculate the theoretical maximum wind vibration position coordinates and its wind vibration data, and compare and analyze them with the danger judgment threshold to judge wind vibration abnormalities in real time.

[0081] Step S4-2: receiving environmental conditions and cable-belt conveyor operation data in real time, predicting and calculating wind vibration danger time points and wind vibration data.

[0082] Step S4-3: Build an automated early warning and control system. When abnormal wind vibration is observed or predicted, the early warning control function is automatically enabled, and the early warning information is pushed to the on-duty personnel. In an emergency, the conveyor control system is intervened to reduce speed or stop the conveyor.

[0083] The present invention provides a suspension belt conveyor wind vibration monitoring system based on machine learning, such as Figure 2 , including: data acquisition module, information processing module, wind vibration monitoring module, and early warning control module.

[0084] The data acquisition module is used to collect on-site environmental data of wind, rain and snow, videos with benchmarks and operating data of suspended belt conveyors, and specifically includes: an anemometer, rain and snow sensors, high-definition cameras, vibration sensors, tachometers and benchmarks.

[0085] The information processing module is used to process on-site environmental data of wind, rain and snow, videos with benchmarks and operating data of cable-belt conveyors, and specifically includes: lens distortion correction, fast Fourier transform, time domain feature extraction, and time-frequency domain feature extraction.

[0086] The wind vibration monitoring module is used to receive various types of data transmitted by the data acquisition module, and to observe and predict the wind vibration data and abnormal time points of the suspension belt conveyor in real time, specifically including: machine vision algorithm, ranging offset algorithm, wind vibration prediction algorithm, and maximum wind vibration calculation method.

[0087] The early warning control module is used to receive the wind vibration data transmitted by the wind vibration monitoring module, monitor and judge the safe operation of the suspension belt conveyor in real time, and take corresponding early warning or control actions, specifically including: autonomous monitoring algorithm, alarm push equipment, and automatic control algorithm.

[0088] In a specific implementation of the present invention, the machine vision algorithm gives priority to the target detection algorithm based on the YOLO model of deep learning; the wind vibration prediction algorithm gives priority to the long short-term memory network (LSTM) or the gated recurrent unit (GRU) suitable for time series data.

[0089] The specific implementation of the ranging offset algorithm includes:

[0090] Distance from camera to benchmark plane:

[0091] Distance from the pole to the center of the screen in the x direction:

[0092] The distance from the pole to the center of the screen in the y direction:

[0093]

[0094] Where: f is the focal length, in millimeters; dx is the actual length of the image sensor per pixel in the x direction, in millimeters per pixel; dy is the actual length of the image sensor per pixel in the y direction, in millimeters per pixel; cx and cy are the coordinates of the principal point of the optical axis, in pixels; l is the length of the pole, in meters; X l , X r , Y t , Y b They are the coordinates of the left, right, top and bottom sides of the benchmark target detection box, in pixels.

[0095] When the maximum wind vibration calculation method is implemented, the vibration of the cable-belt conveyor is considered to be a small-amplitude vibration, and its dynamic equation during vibration is simplified to the vibration equation of the static catenary. The catenary equation of the cable is solved according to the coordinates of the two towers and the coordinates of the benchmark; the straight line equation of the line connecting the two towers is solved according to the coordinates of the two towers; the line connecting the towers is subtracted from the catenary, and the distance between them is considered to be the extreme value of the vibration pendulum length; a plumb line is made through the point where the benchmark is located to intersect the line connecting the towers, and the vibration pendulum length is obtained; based on the proportional relationship between the amplitude at the benchmark and the extreme value of the vibration pendulum length and the vibration pendulum length, the amplitude at the extreme value of the pendulum length is deduced.

[0096] The present invention provides a suspension belt conveyor wind vibration monitoring system based on machine learning, which is specifically arranged as follows: Figure 3 As shown, the target detection pole 5 is arranged at a certain position within the span of two cable towers in the cable belt conveyor, with a fixed length and a color different from the ambient color. Figure 4 As shown, the camera is arranged in the middle above the tower 4, aimed at the target detection pole 5, and placed in the center of the video as much as possible.

[0097] The above-mentioned embodiments only express the specific implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the protection scope of the present application. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the technical solution concept of the present application, and these all belong to the protection scope of the present application.

[0098] This background section is provided to generally present the context of the invention, and the work of the presently named inventors, the work to the extent described in this background section, and aspects of this section that did not constitute prior art at the time of application are neither explicitly nor implicitly admitted to be prior art to the present invention.

Claims

1. A method for monitoring wind vibration of a cable-belt conveyor based on machine learning, characterized in that: include: Step S1: data collection and transmission; Automatically collect videos with target detection benchmarks, environmental data and cable-belt conveyor operation data, and store and transmit them in chronological order; and transmit them to the local server in real time, and the remote server retrieves the data with time axis on demand; Step S2: image correction and data processing; determine the video frame interval and intercept the time series data corresponding to the video frame; calibrate the internal parameters of the camera and measure the external parameters in real time to complete the correction of the video frame; pre-process the obtained data fragments to obtain time domain features and time-frequency domain features; Step S3: Model building and mapping construction; Establish a target detection model and obtain the distance to the benchmark and the wind vibration offset at the benchmark; establish a wind vibration prediction model based on the collected data and the real-time detection results of the cable-belt conveyor; construct a mapping relationship between the wind vibration data at the benchmark and the wind vibration data at the maximum wind vibration point; Step S4: automatic monitoring and early warning; real-time acquisition of wind vibration data calculated by the target detection model, and comparison with the wind vibration threshold analysis; timely notification of maintenance personnel or intervention control to slow down or stop the cable-belt conveyor before the abnormal wind vibration time point predicted by the deep learning model is reached.

2. The method for monitoring wind vibration of a cable-belt conveyor based on machine learning according to claim 1 is characterized in that: The target detection pole is a long strip mark, which needs to be set to a fixed length according to the detection distance, and the color should be different from the ambient color, and is placed horizontally in a span of the suspension belt conveyor.

3. The method for monitoring wind vibration of a cable-belt conveyor based on machine learning according to claim 2 is characterized in that: The required length or width of the target detection pole is selected according to the following formula: in: l is the length of the target detection pole; α is the ratio of the benchmark to the screen; f is the focal length; w cam is the actual width of the camera image sensor; D is the estimated distance between the pole and the camera.

4. The method for monitoring wind vibration of a cable-belt conveyor based on machine learning according to claim 1, characterized in that: The step S1 comprises: Step S1-1: acquiring running video data of a cable-belt conveyor with a target detection pole through a camera and transmitting the data; Step S1-2: Obtain wind direction and wind speed data through an anemometer and store them in chronological order; obtain wind and rain load information through a rain and snow sensor and store and transmit them in chronological order; Step S1-3: Obtain the vibration data of the cable belt conveyor through the vibration sensor and store it in chronological order; obtain the conveyor belt speed through the tachometer, and store and transmit it in chronological order.

5. The method for monitoring wind vibration of a cable-belt conveyor based on machine learning according to claim 1 is characterized in that: The step S2 comprises: Step S2-1: Select a suitable video frame interval according to the number of frames of the monitoring video; intercept the data collected by the anemometer, rain and snow sensor, vibration sensor and tachometer into time-series segments corresponding to the time interval of the video frames; Step S2-2: Determine the internal parameters of the camera, and obtain the focal length, pixel size, optical axis principal point coordinates and camera distortion parameters of the camera through the Zhang Zhengyou calibration method; and measure the external parameters of the camera in real time, that is, the rotation matrix and the translation matrix; Step S2-3: pre-process the intercepted time series data, the anemometer, rain and snow sensor and tachometer obtain their time-frequency characteristics, and the vibration sensor obtains its time-frequency domain characteristics.

6. The method for monitoring wind vibration of a cable-belt conveyor based on machine learning according to claim 1, characterized in that: The step S3 comprises: Step S3-1: Establish a target detection model and train it based on the benchmark dataset; pass the video frame into the model to obtain the benchmark distance and the wind vibration offset at the benchmark; Step S3-2: Based on the time series data of wind direction and speed, rain and snow load, and cable-belt conveyor operation, a wind-induced vibration prediction model is established and trained until the evaluation index meets the requirements; Step S3-3: According to the coordinates of the two cable towers and the benchmark, a mapping relationship between the coordinates of the theoretical maximum wind vibration position of the cable-belt conveyor and its wind vibration data and the wind vibration data at the benchmark is constructed.

7. The method for monitoring wind vibration of a cable-belt conveyor based on machine learning according to claim 1, characterized in that: The step S4 comprises: Step S4-1: Establish an autonomous monitoring algorithm to receive video data in real time; use the target detection model to calculate the theoretical maximum wind vibration position coordinates and wind vibration data, and compare and analyze them with the danger judgment threshold to judge the wind vibration abnormality in real time; Step S4-2: receiving environmental conditions and operation data of the suspension belt conveyor in real time, predicting and calculating the wind vibration danger time point and its wind vibration data; Step S4-3: Build an automated early warning and control system; when abnormal wind vibration is observed or predicted, the early warning control function is automatically enabled, and the early warning information is pushed to the on-duty personnel. In an emergency, the conveyor control system is intervened to achieve speed reduction or shutdown.

8. A wind vibration monitoring system for a cable-belt conveyor based on machine learning, characterized in that: The method for monitoring wind vibration of a suspended cable belt conveyor according to any one of claims 1 to 7 comprises: Data acquisition module, information processing module, wind vibration monitoring module, early warning control module; The data acquisition module includes: an anemometer, a rain and snow sensor, a high-definition camera, a vibration sensor and a tachometer; it is used to collect environmental data of wind, rain and snow on site, videos with benchmarks and operation data of the cable-belt conveyor, and transmit the data to the information processing module; The information processing module includes: a lens distortion correction algorithm, a fast Fourier transform algorithm, a time domain feature extraction method, and a time-frequency domain feature extraction method; it is used to process various types of data transmitted by the receiving data acquisition module and transmit the data to the wind vibration monitoring module; The wind vibration monitoring module includes: machine vision algorithm, ranging offset algorithm, wind vibration prediction algorithm, and maximum wind vibration calculation method; it is used to receive various data transmitted by the data acquisition module, establish and train the model; observe and predict the wind vibration data and abnormal time points of the suspension belt conveyor, and transmit the data to the early warning control module; The early warning control module includes: an autonomous monitoring algorithm, an alarm push device, an automatic control algorithm, and a control linkage system; it is used to receive wind vibration data transmitted by the wind vibration monitoring module, monitor the operation of the cable-belt conveyor in real time, and realize autonomous monitoring and early warning.

9. The machine learning-based wind vibration monitoring system for a cable-belt conveyor according to claim 8, characterized in that: The machine vision algorithm is used to identify the benchmark and obtain its position information in the image; the ranging offset algorithm is used to calculate the distance from the benchmark to the camera and obtain the wind vibration offset of the benchmark relative to the static position; the wind vibration prediction algorithm is used to predict the future wind vibration data of the benchmark under the current data and obtain the dangerous time point; the maximum wind vibration calculation method is used to calculate the position information and wind vibration data of the benchmark across the maximum wind vibration point.

10. The machine learning-based wind vibration monitoring system for a cable-belt conveyor according to claim 8, characterized in that: The autonomous monitoring algorithm is used to monitor the operating status of the suspension belt conveyor in real time and determine whether the wind vibration data of the suspension belt conveyor exceeds the threshold; the automatic control algorithm is used to start the early warning device to remind the on-duty personnel to reduce the conveying volume, slow down or make an emergency stop, and intervene to control the suspension belt conveyor in an emergency.

Images