Radar-based steel coil hoisting monitoring method and system

Abstract

The invention discloses a radar-based steel coil hoisting monitoring method and system, relates to the technical field of safety monitoring, and solves the problems that an image analysis method depends on the environment, and when illumination is insufficient and dust or oil stains cover a camera, the image quality is reduced, so that recognition errors or missing detection is caused; high-resolution image processing needs powerful computing resources, an alarm signal may be delayed, and emergency response is affected. Echo signals reflected by the surface of the steel coil are obtained through the millimeter wave radar sensor; analyzing the echo signal by using a signal processing technology to obtain point cloud data of the surface contour of the steel coil; analyzing and calculating the point cloud data to obtain state parameters of the steel coil; the safety state of the steel coil is analyzed according to the state parameters of the steel coil, and an alarm signal is sent out; the safety condition in the steel coil hoisting process can be effectively identified, and unsafe conditions can be handled in time.

Description

Technical Field

[0001] This application belongs to the field of safety monitoring and relates to steel coil hoisting monitoring technology, specifically a radar-based steel coil hoisting monitoring method and system. Background Technology

[0002] Steel coil hoisting is an operation that uses specialized lifting equipment to move steel coils vertically or horizontally. Monitoring steel coil hoisting allows for real-time monitoring of whether the lifting equipment correctly and completely cradles the center of the steel coil, fundamentally preventing slippage caused by misalignment or improper equipment positioning. Furthermore, during hoisting, parameters such as the steel coil weight, lifting equipment tilt angle, and hoisting height are monitored in real time. When these parameters exceed set thresholds, an alarm is issued promptly, preventing slippage, tipping, or collisions. In the event of a slippage or other accident, analysis of the monitoring data allows for rapid identification of the cause of the malfunction, preventing similar accidents and improving the safety of steel coil hoisting.

[0003] Current technologies for monitoring steel coil hoisting involve capturing the hoisting process with cameras and using image recognition algorithms to monitor the coil's position, size, and surface defects in real time. This allows for some feedback on safety conditions during the hoisting process. However, image analysis methods are dependent on the environment. Insufficient lighting, dust, or oil covering the camera can degrade image quality, leading to identification errors or missed detections. Furthermore, high-resolution image processing requires significant computing resources, which may delay alarm signals and affect emergency response. Summary of the Invention

[0004] This application aims to solve at least one of the technical problems existing in the prior art; to this end, this application proposes a radar-based steel coil hoisting monitoring method and system to solve the technical problems that the image analysis method depends on the environment, and when there is insufficient light, dust or oil covering the camera, the image quality degrades, leading to identification errors or missed detections; and high-resolution image processing requires powerful computing resources, which may delay alarm signals and affect emergency response.

[0005] To achieve the above objectives, the first aspect of this application provides a radar-based method for monitoring the hoisting of steel coils, comprising: The echo signal reflected from the surface of the steel coil is obtained by a millimeter-wave radar sensor; Signal processing techniques are used to analyze the echo signal to obtain point cloud data of the steel coil surface profile; The point cloud data is analyzed and calculated to obtain the state parameters of the steel coil; The safety status of the steel coil is analyzed based on its condition parameters, and an alarm signal is issued.

[0006] Based on the above steps, millimeter-wave radar sensors are used to acquire the echo signals reflected from the surface of the steel coil, which is a non-contact detection method. Compared with traditional contact measurement methods, it does not directly contact the surface of the steel coil, thus avoiding physical damage such as scratches and wear that may be caused by contact, ensuring the quality and integrity of the steel coil. In some industrial production environments, steel coils may be under dangerous conditions such as high temperature, high pressure, or corrosiveness. Non-contact detection allows operators to measure without approaching the steel coil, greatly reducing the risk of personnel exposure to dangerous environments and effectively protecting the personal safety of operators. Point cloud data accurately describes the three-dimensional shape of the steel coil surface in the form of dense points, which can comprehensively and meticulously reflect the contour features of the steel coil surface. Compared to some traditional two-dimensional measurement methods, point cloud data provides richer information and can more accurately detect minute defects and complex shape changes on the surface of steel coils, helping to improve the accuracy and reliability of steel coil quality inspection. It can analyze the safety status of steel coils based on their state parameters and issue alarm signals in a timely manner. When the state parameters of steel coils exceed the normal range and there may be safety hazards, the system can quickly issue an alarm to remind operators to take appropriate measures, thereby eliminating safety hazards in their infancy and effectively preventing safety accidents.

[0007] Preferably, the acquisition of the echo signal reflected from the surface of the steel coil via a millimeter-wave radar sensor includes: An oscillator inside the millimeter-wave radar sensor generates several high-frequency baseband radar signals; the frequency of the high-frequency radar baseband signals increases linearly with time, forming several frequency-modulated continuous waves. The frequency-modulated continuous wave is transmitted towards the steel coil after passing through the amplifier inside the millimeter-wave radar sensor; the directional antenna array of the millimeter-wave radar sensor receives the reflected electromagnetic wave; the electromagnetic wave is amplified using a low-noise amplifier, and the amplified electromagnetic wave is passed through a mixer to obtain the echo signal.

[0008] It should be noted that, since steel coils are made of metal, their surface is an excellent reflector of electromagnetic waves. When millimeter waves come into contact with the curved, smooth cylindrical surface of the steel coil, they will undergo specular reflection and scattering. Furthermore, because the steel coil is cylindrical in shape, different parts of the steel coil are at different distances and angles relative to the radar antenna. Therefore, radar waves will be reflected back from multiple points on the surface of the steel coil, forming a complex reflection field.

[0009] Preferably, the step of analyzing the echo signal using signal processing technology to obtain point cloud data of the steel coil surface profile includes: retrieve echo signal ;in, Indicates the amplitude of the echo signal; Indicates the frequency of the echo signal; Indicates phase; Indicates noise; A bandpass filter is used to remove high-frequency noise and low-frequency drift; a matched filter is used to improve the signal-to-noise ratio; the envelope of the echo signal is extracted by Hilbert transform to obtain the envelope signal; the position of the echo peak is located in the envelope signal; and the corresponding distance is calculated based on the position of the echo peak. The three-dimensional coordinates are calculated based on the scanning angle of the frequency-modulated continuous wave to obtain point cloud data; the expression for calculating the three-dimensional coordinates is: ; in, Indicates the scanning angle. This indicates the distance of the echo peak.

[0010] Preferably, the step of analyzing and calculating the point cloud data to obtain the state parameters of the steel coil includes: The point cloud data is processed by a clustering analysis algorithm to obtain the target point cloud cluster of the steel coil; the characteristic parameters of the target point cloud cluster of the steel coil are analyzed to obtain the target information of the steel coil. The state parameters of the steel coil are analyzed based on the target information of the steel coil. These state parameters include: horizontal offset, tilt angle, oscillation frequency, and relative distance to obstacles.

[0011] Preferably, the step of processing the point cloud data using a clustering analysis algorithm to obtain the target point cloud cluster of the steel coil includes: Retrieve point cloud data; employ a statistical outlier removal method to calculate the average distance from each point to its k nearest neighbors; remove points whose average distance exceeds a distance threshold; the expression for calculating the average distance is: ; Point With the i-th neighboring point The previous Euclidean distance; Define the neighborhood range and the minimum number of points within the neighborhood range; randomly select an unvisited point P in the point cloud data and analyze all points within the neighborhood of the unvisited point P; if the number of points within the neighborhood is greater than the minimum number of points, create a new cluster; otherwise, mark the unvisited point as a noise point. Recursively check the neighborhoods of other unvisited points Q within the neighborhood of point P; when the number of points in the neighborhood of Q is not less than the minimum number of points, add the corresponding point Q to the created cluster; repeat the operation on all point cloud data until all points are visited, and obtain several steel coil target point cloud clusters.

[0012] Preferably, the step of analyzing the characteristic parameters of the target point cloud cluster of the steel coil to obtain the target information of the steel coil includes: Retrieve target point cloud clusters of several steel coils; based on expressions ; ; The centroid of the point cloud cluster is calculated; where, This indicates the number of points in a point cloud cluster; These are the coordinates of each point; Indicates the coordinates of the centroid; Based on expression The covariance matrix of the point cloud cluster is calculated; the covariance matrix is ​​then processed. Eigenvalue decomposition is performed to obtain eigenvalues ​​and their corresponding eigenvectors; the smallest eigenvalue and its corresponding vector are selected to obtain the direction vector of the steel coil axis; where, The coordinate vector of a point; It is the centroid vector; Project each point cloud cluster onto the axis direction vector. Based on formula Calculate the projection value of point j; obtain the height by calculating the difference between the maximum and minimum projection values; based on the formula... Calculate the perpendicular distance from each point to the axis; calculate the average of all perpendicular distances to obtain the radius of the steel coil; where, This is a dot product operation; Denotes the Euclidean norm; The centroid coordinates, axis direction vector, height, and radius are marked as the target information of the steel coil.

[0013] Preferably, the step of analyzing the state parameters of the steel coil based on the target information of the steel coil includes: Obtain the coordinates of the reference point Based on formula The horizontal offset of the steel coil is calculated. retrieve axis direction vector The normalization of the axis direction vector is expressed as follows: ; Based on formula The tilt angle is calculated; where, Represents the vertical direction vector; The position of the centroid is continuously detected, and the centroid offset sequence relative to the average position is calculated; the Fourier transform of the offset sequence is performed to obtain the spectrum; the frequency corresponding to the main peak in the spectrum is identified to obtain the oscillation frequency; Obtain the center coordinates of the obstacle; calculate the distance between the centroid and the obstacle center to obtain the relative distance to the obstacle; integrate the horizontal offset, tilt angle, swing frequency, and relative distance to the obstacle into state parameters.

[0014] Preferably, the step of analyzing the safety status of the steel coil based on its state parameters and issuing an alarm signal includes: Retrieve the state parameters of the steel coil; compare the state parameters of the steel coil with the corresponding parameter thresholds; wherein, the parameter thresholds include: offset threshold, tilt angular frequency, frequency threshold, and distance threshold. If the relative distance to the obstacle in the status parameters is within the corresponding distance threshold, then other data in the status parameters are analyzed; otherwise, an alarm signal is generated; if all status parameters are less than the corresponding parameter thresholds, then the steel coil is determined to be in a safe state; otherwise, an alarm signal is generated and transmitted to the corresponding technical personnel.

[0015] The second aspect of this application provides a radar-based steel coil hoisting monitoring system, including: a data acquisition module, a data analysis module, and a safety alarm module; The data acquisition module is used to acquire the echo signal reflected from the surface of the steel coil through a millimeter-wave radar sensor; The data analysis module is used to analyze the echo signal using signal processing technology to obtain point cloud data of the steel coil surface profile; and to analyze and calculate the point cloud data to obtain the state parameters of the steel coil. The safety alarm module is used to analyze the safety status of the steel coil based on its status parameters and issue an alarm signal.

[0016] A third aspect of this application provides a computer-readable storage medium storing instructions that, when executed on a radar-based steel coil hoisting monitoring system, cause the radar-based steel coil hoisting monitoring system to perform the methods described in the first aspect and any possible implementation thereof.

[0017] Compared with the prior art, the beneficial effects of this application are: 1. The millimeter-wave radar sensor of this application generates a high-frequency baseband radar signal based on a time-linearly increasing frequency, forming a frequency-modulated continuous wave. This provides rich range and velocity information during detection, enabling more accurate differentiation of targets at different distances, effectively improving the detection capability of minute features on the surface of the steel coil, and reducing the possibility of false positives and false negatives. The frequency-modulated continuous wave is amplified and transmitted towards the steel coil, and the reflected electromagnetic wave is received by a directional antenna array. The wave is then amplified by a low-noise amplifier and processed by a mixer to obtain the echo signal. This improves signal transmission efficiency and reception quality, reduces noise interference in the signal, and ensures high-quality echo signals. A bandpass filter can selectively filter out these interference signals, retaining only signals within the effective frequency band, thereby improving the signal-to-noise ratio of the echo signal. The matched filter is designed based on known signal characteristics, maximizing the enhancement of useful signals and suppressing noise in the received signal. Through matched filtering, target features in the echo signal become more prominent. The echo signal envelope is extracted using Hilbert transform, and the peak position of the echo signal is located within it. Hilbert transform can convert real signals into analytic signals, thus facilitating the extraction of the signal envelope. The envelope signal contains information about the amplitude variation of the echo signal, and the echo peak positions correspond to the points of strongest reflection on the steel coil surface. Accurately locating these peak positions is crucial. Based on the scanning angle of the frequency-modulated continuous wave and the calculated distance, the three-dimensional coordinates of the reflection points on the steel coil surface can be obtained using a given three-dimensional coordinate calculation expression, thus generating point cloud data. The point cloud data accurately describes the three-dimensional shape of the steel coil surface in the form of dense points, comprehensively and meticulously reflecting the contour features of the steel coil surface.

[0018] 2. This application employs a statistical outlier removal method, calculating the average distance from each point to its k nearest neighbors and comparing it with a distance threshold to remove outliers. This method effectively removes outliers from point cloud data, improving the purity and accuracy of the data. It sets a neighborhood range and a minimum number of points, randomly selecting unvisited points and analyzing their neighborhoods to create clusters or label noise points. It then recursively checks other unvisited points within the neighborhood and continuously expands the clusters until all points are visited. Based on the spatial distribution characteristics of the point cloud data, it accurately groups points belonging to the same steel coil, achieving precise identification and segmentation of the steel coil target. By calculating the centroid and covariance matrix of the point cloud clusters and performing eigenvalue decomposition on the covariance matrix, it selects the vector corresponding to the smallest eigenvalue to obtain the direction vector of the steel coil axis. This method accurately extracts the direction vector of the steel coil axis. The system uses a multi-dimensional coordinate system to analyze the data. It projects points from a point cloud cluster onto an axial direction vector, calculates the difference between the projected values ​​to obtain the height of the steel coil, and then calculates the vertical distance from each point to the axis, averaging the distances to obtain the coil's radius. This comprehensive analysis of the coil's geometry and dimensions allows for a complete description of its shape and dimensions. Reference point coordinates are used to calculate the coil's horizontal offset. The angle between the axial direction vector and the vertical direction vector, after normalization, is calculated to obtain the tilt angle, enabling timely detection of any abnormalities such as coil offset or tilt. Continuous monitoring of the centroid position is performed, and the offset sequence is calculated and Fourier transformed to obtain the spectrum. The frequency corresponding to the main peak is identified as the oscillation frequency, reflecting the coil's dynamic stability during operation and helping to prevent safety accidents caused by excessive coil oscillation. When all state parameters are below a threshold, the coil is considered safe; otherwise, an alarm signal is generated and transmitted to the relevant technical personnel. By setting reasonable parameter thresholds, abnormal coil conditions can be detected and alarms issued promptly, allowing technical personnel to take swift action and effectively prevent safety accidents. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram illustrating the overall steps of the method described in this application; Figure 2 A schematic diagram illustrating the steps involved in generating point cloud data for this application; Figure 3 This is a schematic diagram of the steel coil safety status analysis steps in this application; Figure 4 This is a schematic diagram of the system structure connection in this application. Detailed Implementation

[0021] The technical solutions of this application will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0022] Please see Figure 1 The first aspect of this application provides a radar-based method for monitoring the hoisting of steel coils, including: S101. Obtain the echo signal reflected from the surface of the steel coil using a millimeter-wave radar sensor; S102. Analyze the echo signal using signal processing technology to obtain point cloud data of the steel coil surface profile; S103. Analyze and calculate the point cloud data to obtain the state parameters of the steel coil; S104. Analyze the safety status of the steel coil based on its state parameters and issue an alarm signal.

[0023] Based on the above steps, millimeter-wave radar sensors are used to acquire the echo signals reflected from the surface of the steel coil, which is a non-contact detection method. Compared with traditional contact measurement methods, it does not directly contact the surface of the steel coil, thus avoiding physical damage such as scratches and wear that may be caused by contact, ensuring the quality and integrity of the steel coil. In some industrial production environments, steel coils may be under dangerous conditions such as high temperature, high pressure, or corrosiveness. Non-contact detection allows operators to measure without approaching the steel coil, greatly reducing the risk of personnel exposure to dangerous environments and effectively protecting the personal safety of operators. Point cloud data accurately describes the three-dimensional shape of the steel coil surface in the form of dense points, which can comprehensively and meticulously reflect the contour features of the steel coil surface. Compared to some traditional two-dimensional measurement methods, point cloud data provides richer information and can more accurately detect minute defects and complex shape changes on the surface of steel coils, helping to improve the accuracy and reliability of steel coil quality inspection. It can analyze the safety status of steel coils based on their state parameters and issue alarm signals in a timely manner. When the state parameters of steel coils exceed the normal range and there may be safety hazards, the system can quickly issue an alarm to remind operators to take appropriate measures, thereby eliminating safety hazards in their infancy and effectively preventing safety accidents.

[0024] In one possible implementation of the embodiments of this application, combined with Figure 1 ,like Figure 2 As shown, the above S101-102 can be specifically implemented through the following S201-S205, which are explained in detail below: S201, The oscillator inside the millimeter-wave radar sensor generates several high-frequency baseband radar signals; the frequency of the high-frequency radar baseband signals increases linearly with time, forming several frequency-modulated continuous waves.

[0025] S202. The frequency-modulated continuous wave is transmitted to the steel coil after passing through the amplifier inside the millimeter-wave radar sensor; the directional antenna array of the millimeter-wave radar sensor receives the reflected electromagnetic wave; the electromagnetic wave is amplified by a low-noise amplifier, and the amplified electromagnetic wave is passed through a mixer to obtain the echo signal.

[0026] Example: Within 0 to 1 second, the generated frequency linearly increases from 77 GHz to 81 GHz, thus forming multiple frequency-modulated continuous waves (FM waves). These FM waves are amplified by an amplifier inside the millimeter-wave radar sensor and then transmitted through a directional antenna array towards steel coils stacked in the workshop. The electromagnetic waves are reflected upon encountering the steel coils, and the directional antenna array receives the reflected electromagnetic waves. The received electromagnetic wave signal is usually quite weak, so it is first amplified using a low-noise amplifier to reduce noise interference. Then, the amplified electromagnetic wave signal is mixed with the transmitted signal by a mixer to obtain an echo signal containing information about the distance to the steel coils.

[0027] S203. Retrieve the echo signal; use a bandpass filter to remove high-frequency noise and low-frequency drift; use a matched filter to improve the signal-to-noise ratio.

[0028] Among them, the echo signal is ; Indicates the amplitude of the echo signal; Indicates the frequency of the echo signal; Indicates phase; Indicates noise.

[0029] S204. Extract the envelope of the echo signal through Hilbert transform to obtain the envelope signal; locate the echo peak position in the envelope signal.

[0030] Example: A bandpass filter is used to process the echo signal, with the passband frequency range set from 77.1 GHz to 80.9 GHz. This removes high-frequency noise and low-frequency drift. Then, a matched filter is used for further signal processing. The matched filter matches the echo signal according to the characteristics of the transmitted signal, thereby improving the signal-to-noise ratio and making the useful signal stand out more. The envelope of the filtered echo signal is extracted using the Hilbert transform. The Hilbert transform converts a real signal into an analytic signal, and the magnitude of the analytic signal is the envelope of the original signal. After obtaining the envelope signal, the peak position is searched within it. A threshold is set; when the amplitude of the envelope signal exceeds this threshold, a peak is considered to have been found, and the time point corresponding to the peak is recorded.

[0031] S205. Calculate the corresponding distance based on the location of the echo peak; calculate the three-dimensional coordinates based on the scanning angle of the frequency-modulated continuous wave to obtain point cloud data; the expression for calculating the three-dimensional coordinates is: ; in, Indicates the scanning angle. This indicates the distance of the echo peak.

[0032] Example: When ; ; When the coordinates are calculated, the corresponding coordinates are (4.16, 2.41, 1.29). Similarly, the three-dimensional coordinates corresponding to the distances of several echo peaks are calculated, and the several three-dimensional coordinates are integrated to obtain point cloud data.

[0033] Based on the above steps, the millimeter-wave radar sensor's internal oscillator generates a high-frequency baseband radar signal based on a time-linearly increasing frequency, forming a frequency-modulated continuous wave (FM wave). This provides rich range and velocity information during detection, enabling more accurate differentiation of targets at different distances, effectively improving the detection capability of minute features on the steel coil surface, and reducing the possibility of false positives and false negatives. The FM wave is amplified and transmitted towards the steel coil, where a directional antenna array receives the reflected electromagnetic waves. The reflected waves are then amplified by a low-noise amplifier and processed by a mixer to obtain the echo signal. This improves signal transmission efficiency and reception quality, reduces noise interference, and ensures high-quality echo signals. A bandpass filter selectively filters out these interfering signals, retaining only signals within the effective frequency band, thereby improving the signal-to-noise ratio of the echo signal. The matched filter, designed based on known signal characteristics, maximizes the enhancement of useful signals and suppresses noise in the received signal. Through matched filtering, target features in the echo signal become more prominent. The Hilbert transform extracts the echo signal envelope, obtaining the envelope signal, and locates the echo peak position within it. The Hilbert transform converts a real signal into an analytic signal, facilitating the extraction of the signal envelope. The envelope signal contains information about the amplitude variation of the echo signal, and the echo peak positions correspond to the points of strongest reflection on the steel coil surface. Accurately locating these peak positions is crucial. Based on the scanning angle of the frequency-modulated continuous wave and the calculated distance, the three-dimensional coordinates of the reflection points on the steel coil surface can be obtained using a given three-dimensional coordinate calculation expression, thus generating point cloud data. The point cloud data accurately describes the three-dimensional shape of the steel coil surface in the form of dense points, comprehensively and meticulously reflecting the contour features of the steel coil surface.

[0034] In one possible implementation of the embodiments of this application, combined with Figure 1 ,like Figure 3 As shown, the above S103-104 can be specifically implemented through the following S301-S303, which are explained in detail below: S301. Retrieve point cloud data; use the statistical outlier removal method to calculate the average distance from each point to its k nearest neighbors; when the average distance is greater than the distance threshold, remove the corresponding point.

[0035] The formula for calculating the average distance is as follows: ; Point With the i-th neighboring point The previous Euclidean distance.

[0036] Example: Set k=10; set the distance threshold to 0.5; for point p, find the 10 closest points to it and calculate the average distance according to the average distance calculation expression.

[0037] S302. Set the neighborhood range and the minimum number of points within the neighborhood range; randomly select an unvisited point P in the point cloud data and analyze all points within the neighborhood of the unvisited point P; if the number of points within the neighborhood is greater than the minimum number of points, create a new cluster; otherwise, mark the unvisited point as a noise point.

[0038] S303. Recursively check the neighborhoods of other unvisited points Q within the neighborhood of point P; when the number of points in the neighborhood of Q is not less than the minimum number of points, add the corresponding point Q to the created cluster; repeat the operation on all point cloud data until all points are visited, and obtain several steel coil target point cloud clusters.

[0039] Example: Set the neighborhood range to 0.3m and the minimum number of points within the neighborhood to 20. Randomly select an unvisited point P in the point cloud data and analyze all points within its neighborhood. If the number of points in the neighborhood is greater than 20, create a new cluster; otherwise, mark the unvisited point P as a noise point. Then recursively check the neighborhoods of other unvisited points Q within the neighborhood of point P. When the number of points in the neighborhood of Q is not less than 20, add the corresponding point Q to the created cluster. Repeat this operation for all point cloud data until all points are visited, ultimately obtaining 5 target point cloud clusters for steel coils.

[0040] S304, retrieve several target point cloud clusters of steel coils; based on expression ; ; The centroid of the point cloud cluster is calculated; based on the expression The covariance matrix of the point cloud cluster is calculated; the covariance matrix is ​​then processed. Eigenvalue decomposition is performed to obtain eigenvalues ​​and their corresponding eigenvectors; the smallest eigenvalue and its corresponding vector are selected to obtain the direction vector of the steel coil axis.

[0041] in, This indicates the number of points in a point cloud cluster; These are the coordinates of each point; Indicates the coordinates of the centroid; The coordinate vector of a point; It is the centroid vector.

[0042] Example: Retrieve a point cloud cluster of one steel coil target, assuming the cluster contains n=100 points; calculate the centroid coordinates of the cluster as (2.5). m 1.8 m 0.8 m The covariance matrix of the point cloud cluster is calculated based on the expression; the covariance matrix is ​​decomposed into eigenvalues ​​to obtain three eigenvalues ​​and their corresponding eigenvectors. The direction vector of the steel coil axis is obtained by selecting the smallest eigenvalue and its corresponding vector.

[0043] S305. Project each element in the point cloud cluster onto the axis direction vector. Based on formula Calculate the projection value of point j; obtain the height by calculating the difference between the maximum and minimum projection values; based on the formula... Calculate the perpendicular distance from each point to the axis; calculate the average of all perpendicular distances to obtain the radius of the steel coil; mark the centroid coordinates, axis direction vector, height, and radius as the target information of the steel coil.

[0044] in, This is a dot product operation; This represents the Euclidean norm.

[0045] Example: Project each point in the point cloud cluster onto the axis direction vector, and calculate the projection value of the point based on the formula. By calculating the difference between the maximum and minimum projection values, the height of the steel coil is obtained as 1.2; calculate the vertical distance from each point to the axis based on the formula, and calculate the average of all vertical distances to obtain the radius of the steel coil as 0.6.

[0046] S306. Obtain the coordinates of the reference point. Based on formula The horizontal offset of the steel coil is calculated; the axis direction vector is retrieved. Normalize the axis direction vector; based on the formula The tilt angle is calculated.

[0047] The normalization expression is: ; This represents the vector in the vertical direction.

[0048] S307. Continuously detect the position of the centroid and calculate the centroid offset sequence relative to the average position; perform Fourier transform on the offset sequence to obtain the spectrum; identify the frequency corresponding to the main peak in the spectrum to obtain the oscillation frequency.

[0049] S308. Obtain the center coordinates of the obstacle; calculate the distance between the centroid and the center of the obstacle to obtain the relative distance to the obstacle; integrate the horizontal offset, tilt angle, swing frequency and relative distance to the obstacle into state parameters.

[0050] Example: Obtain the reference point coordinates as (0,0,0); calculate the horizontal offset of the steel coil as 3 based on the formula; normalize the direction vector of the spool and calculate the tilt angle as 5 based on the formula; continuously detect the difficult position, record the centroid coordinates every 0.1 seconds, and calculate the offset sequence of the centroid relative to the average position within 10 seconds; perform Fourier transform on the offset sequence to obtain the spectrum; identify the frequency corresponding to the main peak in the spectrum and obtain the oscillation frequency as 0.2; obtain the center coordinates of the obstacle in the workshop (4m,2m,0m), calculate the distance between the centroid and the center of the obstacle, and obtain the relative distance of 1.5 to the obstacle.

[0051] S309. Retrieve the status parameters of the steel coil; compare the status parameters of the steel coil with the corresponding parameter thresholds; when the relative distance to the obstacle in the status parameters matches the corresponding distance threshold, analyze other data in the status parameters; otherwise, generate an alarm signal; when all other data in the status parameters are less than the corresponding parameter thresholds, determine that the steel coil is in a safe state; otherwise, generate an alarm signal and transmit the alarm signal to the corresponding technical personnel.

[0052] Among them, the parameter thresholds include: offset threshold, tilt angular frequency, frequency threshold, and distance threshold.

[0053] Example: Obtain the following parameter thresholds: offset threshold is 4m, tilt angle frequency threshold is 10, frequency threshold is 0.5Hz, and distance threshold is 1m; compare the state parameters of the steel coil with the corresponding parameter thresholds; if the relative distance to the obstacle is greater than the distance threshold, then analyze other data in the state parameters; if all other data of the state parameters are less than the corresponding parameter thresholds, then determine that the state of the steel coil is safe.

[0054] Based on the above steps, a statistical outlier removal method is adopted. The average distance from each point to its k nearest neighbors is calculated and compared with a distance threshold to remove outliers. This method can remove outliers from point cloud outputs, improving the purity and accuracy of the point cloud data. A neighborhood range and minimum number of points are set. Clusters are created or noise points are marked by randomly selecting unvisited points and analyzing the situation of points within their neighborhoods. Then, other unvisited points within the neighborhood are recursively checked, and the clusters are continuously expanded until all points are visited. Based on the spatial distribution characteristics of the point cloud data, points belonging to the same steel coil can be accurately grouped together, achieving precise identification and segmentation of the steel coil target. By calculating the centroid and covariance matrix of the point cloud clusters and performing eigenvalue decomposition on the covariance matrix, the direction vector of the steel coil axis is obtained by selecting the vector corresponding to the smallest eigenvalue. This method can accurately extract the axis of the steel coil. Directional information is generated by projecting points in a point cloud cluster onto the axis direction vector and calculating the difference between the projected values ​​to obtain the height of the steel coil. The vertical distance from each point to the axis is then calculated and averaged to obtain the radius of the steel coil, providing a comprehensive description of its geometry and dimensions. Reference point coordinates are obtained to calculate the horizontal offset of the steel coil. After normalizing the axis direction vector, the angle between it and the vertical direction vector is calculated to obtain the tilt angle, enabling timely detection of any abnormalities such as offset or tilt. Continuous monitoring of the centroid position is performed, and the offset sequence is calculated and Fourier transformed to obtain the spectrum. The frequency corresponding to the main peak is identified to obtain the oscillation frequency, reflecting the dynamic stability of the steel coil during operation and helping to avoid safety accidents caused by excessive coil oscillation. When all state parameters are below a threshold, the steel coil is considered safe; otherwise, an alarm signal is generated and transmitted to the corresponding technical personnel. By setting reasonable parameter thresholds, abnormal steel coil conditions can be detected promptly and alarms issued, allowing technical personnel to take swift action and effectively prevent safety accidents.

[0055] Please see Figure 4 The second aspect of this application provides a radar-based steel coil hoisting monitoring system, including: a data acquisition module, a data analysis module, and a safety alarm module; The data acquisition module is used to acquire the echo signal reflected from the surface of the steel coil through a millimeter-wave radar sensor; The data analysis module is used to analyze the echo signal using signal processing technology to obtain point cloud data of the steel coil surface profile; and to analyze and calculate the point cloud data to obtain the state parameters of the steel coil. The safety alarm module is used to analyze the safety status of the steel coil based on its status parameters and issue an alarm signal.

[0056] A third aspect of this application provides a computer-readable storage medium storing instructions that, when executed on a radar-based steel coil hoisting monitoring system, cause the radar-based steel coil hoisting monitoring system to perform the methods described in the first aspect and any possible implementation thereof.

[0057] Some of the data in the above formula are calculated by removing dimensions and taking their numerical values. The formula is the closest to the real situation obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0058] The working principle of this application is as follows: This application acquires the echo signal reflected from the surface of the steel coil through a millimeter-wave radar sensor; analyzes the echo signal using signal processing technology to obtain point cloud data of the steel coil surface contour; analyzes and calculates the point cloud data to obtain the state parameters of the steel coil; analyzes the safety status of the steel coil based on the state parameters of the steel coil, and issues an alarm signal.

[0059] The above embodiments are only used to illustrate the technical methods of this application and are not intended to limit it. Although this application has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of this application without departing from the spirit and scope of the technical methods of this application.

Claims

1. A radar-based method for monitoring the hoisting of steel coils, characterized in that, include: The echo signal reflected from the surface of the steel coil is obtained by a millimeter-wave radar sensor; Signal processing techniques are used to analyze the echo signal to obtain point cloud data of the steel coil surface profile; The point cloud data is analyzed and calculated to obtain the state parameters of the steel coil; The safety status of the steel coil is analyzed based on its condition parameters, and an alarm signal is issued.

2. The radar-based steel coil hoisting monitoring method according to claim 1, characterized in that, The acquisition of the echo signal reflected from the surface of the steel coil via a millimeter-wave radar sensor includes: An oscillator inside the millimeter-wave radar sensor generates several high-frequency baseband radar signals; the frequency of the high-frequency radar baseband signals increases linearly with time, forming several frequency-modulated continuous waves. The frequency-modulated continuous wave is transmitted towards the steel coil after passing through the amplifier inside the millimeter-wave radar sensor; the directional antenna array of the millimeter-wave radar sensor receives the reflected electromagnetic wave; the electromagnetic wave is amplified using a low-noise amplifier, and the amplified electromagnetic wave is passed through a mixer to obtain the echo signal.

3. The radar-based steel coil hoisting monitoring method according to claim 1, characterized in that, The method of analyzing the echo signal using signal processing technology to obtain point cloud data of the steel coil surface profile includes: Retrieve echo signal ;in, Indicates the amplitude of the echo signal; Indicates the frequency of the echo signal; Indicates phase; Indicates noise; A bandpass filter is used to remove high-frequency noise and low-frequency drift; a matched filter is used to improve the signal-to-noise ratio; the envelope of the echo signal is extracted by Hilbert transform to obtain the envelope signal; the position of the echo peak is located in the envelope signal; and the corresponding distance is calculated based on the position of the echo peak. The three-dimensional coordinates are calculated based on the scanning angle of the frequency-modulated continuous wave to obtain point cloud data; the expression for calculating the three-dimensional coordinates is: ； in, Indicates the scanning angle. This indicates the distance of the echo peak.

4. The radar-based steel coil hoisting monitoring method according to claim 1, characterized in that, The analysis and calculation of point cloud data to obtain the state parameters of the steel coil includes: The point cloud data is processed by a clustering analysis algorithm to obtain the target point cloud cluster of the steel coil; the characteristic parameters of the target point cloud cluster of the steel coil are analyzed to obtain the target information of the steel coil. The state parameters of the steel coil are analyzed based on the target information of the steel coil. These state parameters include: horizontal offset, tilt angle, oscillation frequency, and relative distance to obstacles.

5. The radar-based steel coil hoisting monitoring method according to claim 4, characterized in that, The algorithm for processing point cloud data through cluster analysis to obtain target point cloud clusters for steel coils includes: Retrieve point cloud data; employ a statistical outlier removal method to calculate the average distance from each point to its k nearest neighbors; remove points whose average distance exceeds a distance threshold; the expression for calculating the average distance is: ; Point With the i-th neighboring point The previous Euclidean distance; Define the neighborhood range and the minimum number of points within the neighborhood range; randomly select an unvisited point P in the point cloud data and analyze all points within the neighborhood of the unvisited point P; if the number of points within the neighborhood is greater than the minimum number of points, create a new cluster; otherwise, mark the unvisited point as a noise point. Recursively check the neighborhoods of other unvisited points Q within the neighborhood of point P; when the number of points in the neighborhood of Q is not less than the minimum number of points, add the corresponding point Q to the created cluster; repeat the operation on all point cloud data until all points are visited, and obtain several steel coil target point cloud clusters.

6. The radar-based steel coil hoisting monitoring method according to claim 4, characterized in that, The analysis of the characteristic parameters of the target point cloud cluster of the steel coil yields the target information of the steel coil, including: Retrieve target point cloud clusters of several steel coils; based on expressions ; ; The centroid of the point cloud cluster is calculated; where, This indicates the number of points in a point cloud cluster; These are the coordinates of each point; Indicates the coordinates of the centroid; Based on expression The covariance matrix of the point cloud cluster is calculated; the covariance matrix is ​​then processed. Eigenvalue decomposition is performed to obtain eigenvalues ​​and their corresponding eigenvectors; the smallest eigenvalue and its corresponding vector are selected to obtain the direction vector of the steel coil axis; where, The coordinate vector of a point; It is the centroid vector; Project each point cloud cluster onto the axis direction vector. Based on formula Calculate the projection value of point j; obtain the height by calculating the difference between the maximum and minimum projection values; based on the formula... Calculate the perpendicular distance from each point to the axis; calculate the average of all perpendicular distances to obtain the radius of the steel coil; where, This is a dot product operation; Denotes the Euclidean norm; The centroid coordinates, axis direction vector, height, and radius are marked as the target information of the steel coil.

7. The radar-based steel coil hoisting monitoring method according to claim 4, characterized in that, The analysis of the state parameters of the steel coil based on the target information of the steel coil includes: Obtain the coordinates of the reference point Based on formula The horizontal offset of the steel coil is calculated. retrieve axis direction vector The normalization of the axis direction vector is expressed as follows: ； Based on formula The tilt angle is calculated; where, Represents the vertical direction vector; The position of the centroid is continuously detected, and the centroid offset sequence relative to the average position is calculated; the Fourier transform of the offset sequence is performed to obtain the spectrum; the frequency corresponding to the main peak in the spectrum is identified to obtain the oscillation frequency; Obtain the center coordinates of the obstacle; calculate the distance between the centroid and the obstacle center to obtain the relative distance to the obstacle; integrate the horizontal offset, tilt angle, swing frequency, and relative distance to the obstacle into state parameters.

8. The radar-based steel coil hoisting monitoring method according to claim 1, characterized in that, The process of analyzing the safety status of the steel coil based on its state parameters and issuing an alarm signal includes: Retrieve the state parameters of the steel coil; compare the state parameters of the steel coil with the corresponding parameter thresholds; wherein, the parameter thresholds include: offset threshold, tilt angular frequency, frequency threshold, and distance threshold. If the relative distance to the obstacle in the status parameters is within the corresponding distance threshold, then other data in the status parameters are analyzed; otherwise, an alarm signal is generated; if all status parameters are less than the corresponding parameter thresholds, then the steel coil is determined to be in a safe state; otherwise, an alarm signal is generated and transmitted to the corresponding technical personnel.

9. A radar-based steel coil hoisting monitoring system, applied to the radar-based steel coil hoisting monitoring method according to any one of claims 1-8, characterized in that, include: Data acquisition module, data analysis module, and security alarm module; The data acquisition module is used to acquire the echo signal reflected from the surface of the steel coil through a millimeter-wave radar sensor; The data analysis module is used to analyze the echo signal using signal processing technology to obtain point cloud data of the steel coil surface profile; and to analyze and calculate the point cloud data to obtain the state parameters of the steel coil. The safety alarm module is used to analyze the safety status of the steel coil based on its status parameters and issue an alarm signal.

10. A computer-readable storage medium storing instructions that, when executed on a radar-based steel coil hoisting monitoring system, cause the radar-based steel coil hoisting monitoring system to perform the method as described in any one of claims 1-8.

Images