Double-tower drum transportation ovality on-line monitoring method based on double-camera width measurement

Abstract

The application discloses a kind of based on dual-camera width measurement's double-tower cylinder transport ovality on-line monitoring method, belong to wind power generation equipment state monitoring and intelligent operation and maintenance technical field.The method includes: after the side surface image of tower cylinder collected by camera is preprocessed, by edge extraction and least square method fitting, and further calculate the instantaneous projection pixel width;Scale factor and illumination compensation model are introduced, the actual physical width of the two sides of tower cylinder is obtained, and the weighted average method is used to obtain the short axis diameter of tower cylinder;According to the short axis diameter of tower cylinder and the ovality calculation model, the deformed long axis of ellipse is obtained, the real-time ovality of tower cylinder at present time is further calculated, and whether the real-time ovality meets the ovality upper limit requirement, the ovality change requirement and the ovality fluctuation amplitude requirement is judged, whether alarm signal is generated according to the determination result.The application method effectively solves the technical problem that tower cylinder is difficult to measure directly in double-tower cylinder parallel transport caused by tower cylinder shielding.

Description

Technical Field

[0001] This invention belongs to the field of wind power equipment condition monitoring and intelligent operation and maintenance technology, and specifically relates to an online monitoring method for the transport ellipticity of dual towers based on dual-camera width measurement. Technical Background

[0002] As a core pillar of the global clean energy system, wind turbine power generation technology is rapidly developing towards larger single-unit capacity and ultra-tall tower structures. Consequently, the size and weight of wind turbine towers are increasing dramatically, with diameters reaching several meters, single tower sections exceeding thirty meters in length, and weights often reaching hundreds of tons. These behemoths, after manufacturing, typically require a long and complex land and sea transportation process before finally arriving at the wind farm site for installation.

[0003] In current engineering practice, parallel transport of two tower sections has become a common and important transportation method in the industry to improve transportation efficiency and reduce transportation costs. In this mode, two tower sections are placed vertically and parallel on the same transport vehicle, forming a unique spatial constraint. However, this configuration presents a bottleneck for implementing strain gauge monitoring methods on a single tower section: the upper tower section is completely obscured, making sensor deployment difficult, wiring complex, and effective monitoring of the obscured tower section challenging.

[0004] The mechanical behavior during the transportation of twin towers is more complex. The interaction between the two towers, the non-uniformity of the support system, and the dynamic loads during transportation all generate complex stress distributions within the towers. Among these, elliptic deformation of the cross-section is one of the main structural risks. Excessive ellipticity not only affects the subsequent tower docking and installation, but more importantly, it alters the ideal stress state of the tower structure, introducing additional bending moments within the tower wall and leading to stress concentration. This effect is particularly significant in weak areas such as the circumferential welds of the towers.

[0005] Traditional offline measurement methods cannot meet the real-time monitoring needs of the twin-tower transportation process. The industry urgently needs to develop a systematic technology that adapts to the characteristics of twin-tower transportation and realizes real-time monitoring and intelligent diagnosis throughout the entire process. This will break through the limitations of traditional methods, establish an intelligent monitoring system based on visual sensing, and complete the transformation from a passive response to an active early warning risk management model, thus providing reliable protection for the safety of major equipment transportation. Summary of the Invention

[0006] To address the technical challenge of real-time monitoring and safety assessment of ellipticity deformation of wind turbine towers during parallel transport of two towers, this invention provides an online monitoring method for ellipticity during transport of two towers based on dual-camera width measurement. This method is used for online monitoring, availability assessment, and safety early warning of ellipticity changes during parallel transport of two towers.

[0007] In a first aspect, this invention proposes an online monitoring method for the ellipticity of a dual-tower transport system based on dual-camera width measurement, comprising:

[0008] S1, images of the left and right sides of the tower are acquired at time intervals by dual cameras installed on both sides of the tower, and Gaussian filtering and image enhancement are used for preprocessing in sequence to obtain the preprocessed images of both sides;

[0009] S2, after preprocessing the two side images, edge extraction and least squares fitting are used to obtain the straight line equations corresponding to the upper and lower edges of the tower, and the instantaneous projection pixel widths on both sides are further calculated.

[0010] S3, introduce scale factor and illumination compensation model, and calculate the actual physical width of both sides by combining the instantaneous projected pixel width of both sides; calculate the minor axis diameter of the tower by using the weighted average method of the actual physical width of both sides.

[0011] S4. Based on the tower's minor axis diameter and ellipticity calculation model, the deformed elliptic major axis is obtained, and the real-time ellipticity of the tower at the current moment is further calculated; the ellipticity calculation model is established based on the assumption of equal perimeter.

[0012] S5 determines whether the real-time ellipticity meets the upper limit requirement, the ellipticity change requirement, and the ellipticity fluctuation range requirement. If all requirements are met, transportation and monitoring continue, and S1-S5 are cycled. Otherwise, transportation and monitoring are stopped, and an alarm signal is generated.

[0013] Furthermore, in S1, the camera must also meet the requirements for field of view coverage and camera installation distance. The field of view coverage requirement is that the horizontal field of view of each camera can completely cover the maximum projected size of the tower in the horizontal direction. The camera installation distance requirement is that the distance between the camera and the tower is less than or equal to the maximum installation distance. The maximum installation distance is obtained by dividing the product of the camera focal length and the nominal diameter of the tower by the product of the horizontal dimension of the camera sensor and the safety factor.

[0014] Furthermore, in S1, the image enhancement is performed using the CLAHE algorithm.

[0015] Furthermore, in S2, edge extraction and least squares fitting are specifically as follows:

[0016] First, the gradient magnitude and direction of the image are calculated using the Sobel operator;

[0017] Secondly, an improved Canny algorithm is used to calculate the edge binary image based on the gradient magnitude and direction of the image;

[0018] Finally, the random Hough transform is applied to the binary edge image to detect the boundary lines. Based on the detected boundary lines, the least squares method is used to fit the lines of the upper and lower edges of the tower, respectively, to obtain the line equations corresponding to the upper edge and the lower edge of the tower.

[0019] Furthermore, in step S2, the specific formula for calculating the instantaneous projected pixel width at the current moment is as follows:

[0020] ;

[0021] in, Number of sampling lines , The first The upper and lower boundary y-coordinates of the sampling line at the current moment are obtained from the equations of the corresponding straight lines at the upper and lower edges of the tower, where t represents time. The instantaneous projection pixel width at the current moment.

[0022] Furthermore, in S3, the specific calculation method for the actual physical width is as follows:

[0023] First, using the illumination compensation model, the error compensation value is calculated based on the average gray value of the image on one side of the tower at the current moment;

[0024] Secondly, the actual physical width is obtained by multiplying the scale factor by the instantaneous projected pixel width at the current moment and adding it to the error compensation value.

[0025] Furthermore, the scaling factor is obtained through calibration experiments, specifically as follows:

[0026] Based on the single-side image of the tower acquired by the camera in the calibration experiment, multiple edge feature points on the upper side of the tower and multiple edge feature points on the lower side of the tower were extracted through image processing.

[0027] The least squares method was used to fit the feature points on the upper and lower edges of the tower respectively, and the straight line equations on the upper and lower edges of the tower were obtained.

[0028] The average pixel width of the upper and lower sides of the tower is calculated based on the straight line equations of the upper and lower edges of the tower.

[0029] Divide the nominal diameter of the tower by the average pixel width to obtain the scale factor.

[0030] Furthermore, the illumination compensation model was obtained through calibration experiments, specifically as follows:

[0031] The camera captured images of one side of the tower under different lighting conditions, and the average gray value of all images was calculated.

[0032] Based on the known error compensation values ​​from the calibration experiment, an exponential equation is established between the average gray value and the error compensation value, which is the illumination compensation model.

[0033] Furthermore, in S5, the requirements for the upper limit of ellipticity, the amount of change in ellipticity, and the amplitude of ellipticity fluctuation are specifically as follows:

[0034] The upper limit requirement for ellipticity is that the real-time ellipticity shall not exceed the design allowable upper limit; the requirement for ellipticity variation is that the real-time ellipticity change rate shall not exceed the critical change rate; the requirement for ellipticity fluctuation amplitude is that within a length of Within the time window, the average fluctuation range of the real-time ellipticity deviation from the mean should be less than the fluctuation threshold, that is:

[0035] ;

[0036] in, This represents the average real-time ellipticity within the time window. Let t be the fluctuation threshold, and t be the time interval. This represents the average fluctuation range.

[0037] Secondly, this invention proposes an online monitoring system for the ellipticity of dual-tower transport based on dual-camera width measurement, which is used to implement the above-mentioned online monitoring method for the ellipticity of dual-tower transport based on dual-camera width measurement.

[0038] The beneficial effects of this invention are:

[0039] This invention proposes a non-contact monitoring scheme based on dual-camera width measurement for parallel transport of twin towers, effectively solving the technical problem of difficulty in direct measurement caused by tower obstruction during parallel transport.

[0040] By establishing a precise optical calibration model and an adaptive lighting compensation mechanism, the accuracy and robustness of visual measurement under complex transportation conditions are significantly improved.

[0041] By employing multi-view data fusion and dynamic weighted optimization algorithms, the system's resistance to interference factors such as vibration and changes in lighting is enhanced.

[0042] The constructed composite criterion two-level response mechanism simplifies the decision-making process while ensuring the accuracy of status determination, realizing a complete technical closed loop from real-time monitoring to intelligent early warning.

[0043] The monitoring system adopts a modular and lightweight design, featuring rapid deployment and flexible configuration. It enables remote real-time management and analysis of monitoring data through a wireless transmission network, significantly improving the safety control efficiency of the entire dual-tower transportation process. Attached Figure Description

[0044] Figure 1 This is an overall framework diagram of the method of the present invention;

[0045] Figure 2 This is a schematic diagram of the structure of a double-tower measuring device;

[0046] Figure 3 A method for real-time ellipticity monitoring of the tower during transportation; Detailed Implementation

[0047] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below. Technical features in the various embodiments of the present invention can be combined accordingly without mutual conflict.

[0048] This invention addresses the typical loading mode of parallel transportation of twin towers by designing a method for measuring and evaluating the ellipticity of the towers during transportation based on dual-camera visual sensing. Through non-contact visual measurement technology, it enables real-time and continuous monitoring of ellipticity deformation during the transportation of twin towers.

[0049] like Figure 2 As shown, the method of this invention relies on industrial cameras 3, camera mounting brackets 4, and a control terminal located on the left and right sides of the tower. The industrial cameras mounted on the camera mounting brackets capture images of the tower in real time and transmit them to the control terminal. The control terminal further calculates the ellipticity of the corresponding tower based on the received images and determines whether to trigger an early warning signal.

[0050] like Figure 1 As shown, the specific steps include:

[0051] (1) Deployment and spatial calibration of visual sensing units.

[0052] The deployment of visual sensing units requires precise spatial geometric constraints to ensure full coverage monitoring of the twin-tower structure. This solution deploys two visual sensing units symmetrically on the left and right sides of each monitored tower. Each sensing unit consists of an industrial camera, with the two cameras symmetrically arranged on both sides of the tower and their optical axes strictly aligned in the horizontal plane, forming a collaborative width measurement pair. For the parallel transportation scenario of the twin towers in this solution, a total of four visual sensing units are required.

[0053] All visual sensing units are mounted on the transverse support structure of the transport base via rigid brackets. During installation, it is necessary to ensure that the optical center line connecting the two cameras in each sensing unit is perpendicular to the preset central axis of the tower. A level is used to ensure that the height of the camera's optical center is consistent with the height of the tower's axis, with the height deviation controlled within millimeters. At the same time, the installation distance is calculated based on the camera parameters to ensure that the tower can be fully imaged.

[0054] Installation distance between the sensing unit and the tower The field of view coverage requirement must be met, and the calculation formula is as follows:

[0055]

[0056] in, For camera lens focal length, The nominal diameter of the tower. The horizontal dimension of the camera sensor. This is for the safety factor.

[0057] Horizontal field of view The calculation formula is During installation, ensure that It can completely cover the maximum projected size of the tower in the horizontal direction.

[0058] like Figure 2 The figure shows a specific embodiment of the present invention, wherein 1 is the outer contour shape of the tower without elliptical deformation; 2 is the outer contour shape of the tower after irreversible elliptical deformation during transportation; 3 is an industrial camera, i.e., a vision sensing unit, and two industrial cameras located on the same horizontal line form a width measurement pair; 4 is a camera fixing bracket; and 5 is the bearing surface of the transportation facility.

[0059] (2) System benchmark calibration and measurement model construction.

[0060] Before the transportation task begins, the measurement system benchmark calibration must be completed when the tower is in a known roundness state.

[0061] Taking a single tower as an example, firstly, the tower is stably placed on a standard platform. The two visual sensing units in the coordinated width measurement pair acquire multiple sets of clear side projection images of the tower. The average value is then taken to reduce noise interference, ultimately obtaining the side image of the tower. Secondly, based on the side image, multiple edge feature points on the upper and lower sides of the tower are extracted using an image processing algorithm (in this specific embodiment, Canny edge detection is used). The linear equations for the upper and lower edges of the tower are then fitted using the least squares method. Finally, the average pixel width on both the upper and lower sides of the tower is calculated. Combined with the nominal diameter of the tower A pixel scale conversion model is established, with the following specific expression:

[0062]

[0063] The final scale factor This will serve as the baseline parameter for subsequent real-time measurements.

[0064] Simultaneously, sample images under different lighting conditions were acquired to establish an illumination compensation model:

[0065]

[0066] in, The error compensation value is calculated by the illumination compensation model, where M is the average gray value of all sample images. The compensation coefficients were calibrated through experiments, thus forming a complete width-to-diameter conversion system.

[0067] (3) Real-time monitoring and deformation feature extraction during transportation.

[0068] During the transportation process, the logical structure diagram of the ellipticity detection method using binocular vision sensing is as follows: Figure 3 As shown. At this time, the visual sensing units on both sides of the tower operate at a preset sampling frequency. Continuous acquisition of images of the tower side Where N is the total number of samples in the current record, The image shows the side view of the tower acquired at time t.

[0069] Preprocessing is performed on all images acquired by the visual sensing unit, specifically including:

[0070] First, the standard deviation is used as A Gaussian filter is used to suppress image noise. The expression for a Gaussian filter is as follows:

[0071]

[0072] in, This represents the coordinate offset of the pixel relative to the center of the filter. Usually taken Pixel;

[0073] Secondly, Limit Contrast Adaptive Histogram Equalization (CLAHE) is used to enhance edge contrast and improve the robustness of edge detection.

[0074] After obtaining the preprocessed side image of the tower, the edge detection and width extraction stage begins. First, the Sobel operator is used to calculate the gradient magnitude and direction of the image. The Sobel operator template is... and The formula for calculating the gradient magnitude is: The formula for calculating the gradient direction is: ,in represents the convolution operation, and I represents the preprocessed side image of the tower.

[0075] Next, an improved Canny algorithm is used to perform non-maximum suppression and double threshold hysteresis processing based on gradient magnitude and direction to obtain an accurate binary edge image.

[0076] Subsequently, in the camera shooting area The random Hough transform is used to detect boundary lines, and the parameter space is represented as follows: ,in The perpendicular distance from the origin to the line is... The angle between the normal to the line and the x-axis.

[0077] Based on the detected boundary lines, the equations of the upper and lower edges of the tower are obtained by fitting the equations of the upper and lower edges using the least squares method. Then, the instantaneous projected pixel width at time t is calculated. ,in The number of sampling lines (determined by image resolution). , The first The y-coordinates of the upper and lower boundaries of the sampling line at time t.

[0078] In the physical width conversion and data fusion stage, the scale factor obtained in the calibration stage is combined with the data fusion stage. And the illumination compensation model, calculate the actual physical width ,in This is the error compensation value calculated by the illumination compensation model based on the average gray level of the current image.

[0079] For each tower section, the width of the left side can be calculated based on the visual sensing units on its left and right sides. and the width on the right , respectively denoted as and The optimized tower minor axis diameter measurement value was obtained by fusion using a weighted average method. Among them, the weighting coefficient and The weights can be dynamically allocated based on the signal-to-noise ratio or edge sharpness of the left and right camera images, or a fixed weight can be used. .

[0080] (4) Ellipticity calculation and deformation assessment.

[0081] In the ellipticity calculation and deformation evaluation stage, an ellipticity calculation model is established based on the assumption of equal perimeter.

[0082] Assuming the tower cross-section maintains a constant perimeter during elastic deformation and remains a standard ellipse before and after deformation, it can be deduced that... ,in The circumference of the ellipse. The major axis of the deformed ellipse, The short axis is the width value calculated in (3).

[0083] Using the Ramanujan approximation formula:

[0084]

[0085] Substituting the condition of equal perimeter, we get:

[0086]

[0087] set up Given the ratio of the major and minor axes, the above formula can be simplified to:

[0088]

[0089] The information obtained from the compilation equation

[0090] .

[0091] In practical engineering In cases where the ellipticity is close to 1, let ,when An approximate solution can be obtained at this time:

[0092]

[0093] The formula for calculating the real-time ellipticity index is as follows: Substitute The approximate expression is obtained .

[0094] To improve computational accuracy, iterative methods can also be used to solve for the more accurate results. The value, the iterative formula is:

[0095]

[0096] by Using the initial value, convergence is typically achieved after 3-4 iterations. Then, the ellipticity index is calculated using the real-time ellipticity index calculation formula. .

[0097] The terminal synchronously records the historical change curve of ellipticity. And calculate the mean. ,variance and gradient of change These statistical characteristics provide a quantitative basis for the comprehensive assessment of the tower structure status, while also supporting data traceability and export, meeting the traceability requirements of the transportation process.

[0098] (5) Status determination and two-level response mechanism.

[0099] During the state determination and two-level response mechanism phase, the real-time calculated value of the ellipticity of each tower is used. The terminal performs a binary classification of "normal / abnormal" status based on composite criteria.

[0100] The tower's normal state must meet the following three conditions simultaneously: (1) The real-time ellipticity must not exceed the design allowable upper limit. ,Right now ,in The determination is based on tower design specifications and transportation safety standards (usually taken as...). (2) The change in ellipticity per unit time shall not exceed the set critical rate of change. ,Right now ,in , is the sampling interval; (3) in a length of Within the time window, the average fluctuation of the ellipticity deviation from the mean should be less than the allowable fluctuation threshold. ,Right now:

[0101]

[0102] in The time window length, The mean ellipticity within the window. The allowable fluctuation threshold.

[0103] If all three conditions are met, the terminal determines the tower is in a normal state, only recording, uploading, and visualizing data without triggering active intervention. If any condition is not met, the terminal immediately determines the tower is in an abnormal state, sending an alarm message to the remote monitoring center containing the tower number, abnormality type, severity, and location information, and issuing a tiered parking instruction to the vehicle terminal. This determination mechanism integrates information from three dimensions: ellipticity amplitude characteristics, change trends, and fluctuation characteristics. While simplifying the warning level to a binary classification of "normal / abnormal," it ensures the accuracy of state judgment and response timeliness through composite criteria.

[0104] Based on the same inventive concept, this invention proposes an online monitoring system for the ellipticity of a dual-tower transport system based on dual-camera width measurement, comprising:

[0105] A hybrid device consisting of an industrial camera whose optical center height is consistent with the tower axis height and a camera mounting bracket is used to acquire images from both sides of the tower and transmit them to the control terminal in real time.

[0106] The control terminal is used to calculate the ellipticity of the tower based on the received image and determine whether to generate an alarm signal.

[0107] The control terminal includes:

[0108] The preprocessing module is used to acquire images of the left and right sides of the tower at time intervals using dual cameras installed on both sides of the tower, and then perform Gaussian filtering and image enhancement in sequence to obtain the preprocessed images of both sides.

[0109] The pixel width calculation module is used to obtain the straight line equations corresponding to the upper and lower edges of the tower by edge extraction and least squares fitting of the preprocessed images on both sides, and then calculates the instantaneous projected pixel width on both sides.

[0110] The tower minor axis diameter calculation module is used to introduce scale factor and illumination compensation model, and calculate the actual physical width of both sides by combining the instantaneous projected pixel width of both sides; the tower minor axis diameter is calculated by weighted average method using the actual physical width of both sides.

[0111] The real-time ellipticity calculation module is used to obtain the deformed major axis of the ellipse based on the minor axis diameter of the tower and the ellipticity calculation model, and further calculate the real-time ellipticity of the tower at the current moment; the ellipticity calculation model is established based on the assumption of equal perimeter.

[0112] The determination module is used to determine whether the real-time ellipticity meets the upper limit requirement of ellipticity, the amount of ellipticity change requirement, and the ellipticity fluctuation range requirement. If all requirements are met, transportation and monitoring continue, and the cycle S1-S5 is repeated. Otherwise, transportation and monitoring are stopped, and an alarm signal is generated.

[0113] For the system embodiments, since they basically correspond to the method embodiments, relevant details can be found in the descriptions of the method embodiments; the implementation methods of the modules will not be repeated here. The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and 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 modules can be selected to achieve the purpose of the present invention according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0114] The system embodiments of the present invention can be applied to any device with data processing capabilities, such as a computer or other similar device. The system embodiments can be implemented in software, hardware, or a combination of both. Taking software implementation as an example, as a logical device, it is formed by the processor of any data processing device loading the corresponding computer program instructions from non-volatile memory into memory for execution.

[0115] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the invention. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention. Therefore, all technical solutions obtained through equivalent substitution or transformation fall within the protection scope of the present invention.

Claims

1. A method for online monitoring of the ellipticity of a dual-tower transport system based on dual-camera width measurement, characterized in that, include: S1, images of the left and right sides of the tower are acquired at time intervals by dual cameras installed on both sides of the tower, and Gaussian filtering and image enhancement are used for preprocessing in sequence to obtain the preprocessed images of both sides; S2, after preprocessing the two side images, edge extraction and least squares fitting are used to obtain the straight line equations corresponding to the upper and lower edges of the tower, and the instantaneous projection pixel widths on both sides are further calculated. S3, introduce scale factor and illumination compensation model, and calculate the actual physical width of both sides by combining the instantaneous projected pixel width of both sides; calculate the minor axis diameter of the tower by using the weighted average method of the actual physical width of both sides. S4. Based on the tower's minor axis diameter and ellipticity calculation model, the deformed elliptic major axis is obtained, and the real-time ellipticity of the tower at the current moment is further calculated; the ellipticity calculation model is established based on the assumption of equal perimeter. S5 determines whether the real-time ellipticity meets the upper limit requirement, the ellipticity change requirement, and the ellipticity fluctuation range requirement. If all requirements are met, transportation and monitoring continue, and S1-S5 are cycled. Otherwise, transportation and monitoring are stopped, and an alarm signal is generated.

2. The method for online monitoring of the ellipticity of a dual-tower transport system based on dual-camera width measurement according to claim 1, characterized in that, In S1, the camera also needs to meet the field of view coverage requirement and the camera installation distance requirement. The field of view coverage requirement is that the horizontal field of view angle of each camera can completely cover the maximum projected size of the tower in the horizontal direction. The camera installation distance requirement is that the distance between the camera and the tower is less than or equal to the maximum installation distance. The maximum installation distance is obtained by dividing the product of the camera focal length and the nominal diameter of the tower by the product of the horizontal dimension of the camera sensor and the safety factor.

3. The method for online monitoring of the ellipticity of a dual-tower transport system based on dual-camera width measurement according to claim 1, characterized in that, In step S1, the image enhancement is performed using the CLAHE algorithm.

4. The method for online monitoring of the ellipticity of a dual-tower transport system based on dual-camera width measurement according to claim 1, characterized in that, In S2, edge extraction and least squares fitting are specifically as follows: First, the gradient magnitude and direction of the image are calculated using the Sobel operator; Secondly, an improved Canny algorithm is used to calculate the edge binary image based on the gradient magnitude and direction of the image; Finally, the random Hough transform is applied to the binary edge image to detect the boundary lines. Based on the detected boundary lines, the least squares method is used to fit the lines of the upper and lower edges of the tower, respectively, to obtain the line equations corresponding to the upper edge and the lower edge of the tower.

5. The method for online monitoring of the ellipticity of a dual-tower transport system based on dual-camera width measurement according to claim 1, characterized in that, In step S2, the specific formula for calculating the instantaneous projected pixel width at the current moment is as follows: ； in, Number of sampling lines , The first The upper and lower boundary y-coordinates of the sampling line at the current moment are obtained from the equations of the corresponding straight lines at the upper and lower edges of the tower, where t represents time. The instantaneous projection pixel width at the current moment.

6. The method for online monitoring of the ellipticity of a dual-tower transport system based on dual-camera width measurement according to claim 1, characterized in that, In S3, the specific calculation method for the actual physical width is as follows: First, using the illumination compensation model, the error compensation value is calculated based on the average gray value of the image on one side of the tower at the current moment; Secondly, the actual physical width is obtained by multiplying the scale factor by the instantaneous projected pixel width at the current moment and adding it to the error compensation value.

7. The method for online monitoring of the ellipticity of a dual-tower transport system based on dual-camera width measurement according to claim 6, characterized in that, The scaling factor was obtained through calibration experiments, specifically: Based on the single-side image of the tower acquired by the camera in the calibration experiment, multiple edge feature points on the upper side of the tower and multiple edge feature points on the lower side of the tower were extracted through image processing. The least squares method was used to fit the feature points on the upper and lower edges of the tower respectively, and the straight line equations on the upper and lower edges of the tower were obtained. The average pixel width of the upper and lower sides of the tower is calculated based on the straight line equations of the upper and lower edges of the tower. Divide the nominal diameter of the tower by the average pixel width to obtain the scale factor.

8. The method for online monitoring of the ellipticity of a dual-tower transport system based on dual-camera width measurement according to claim 6, characterized in that, The illumination compensation model was obtained through calibration experiments, specifically as follows: The camera captured images of one side of the tower under different lighting conditions, and the average gray value of all images was calculated. Based on the known error compensation values ​​from the calibration experiment, an exponential equation is established between the average gray value and the error compensation value, which is the illumination compensation model.

9. The method for online monitoring of the ellipticity of a dual-tower transport system based on dual-camera width measurement according to claim 1, characterized in that, In S5, the requirements for the upper limit of ellipticity, the amount of change in ellipticity, and the amplitude of ellipticity fluctuation are as follows: The upper limit requirement for ellipticity is that the real-time ellipticity shall not exceed the design allowable upper limit; the requirement for ellipticity variation is that the real-time ellipticity change rate shall not exceed the critical change rate; the requirement for ellipticity fluctuation amplitude is that within a length of Within the time window, the average fluctuation range of the real-time ellipticity deviation from the mean should be less than the fluctuation threshold, that is: ； in, This represents the average real-time ellipticity within the time window. Let t be the fluctuation threshold, and t be the time interval. This represents the average fluctuation range.

10. An online monitoring system for the ellipticity of a dual-tower transport system based on dual-camera width measurement, used to implement the online monitoring method for the ellipticity of a dual-tower transport system based on dual-camera width measurement as described in claim 1, characterized in that... include: A hybrid device consisting of an industrial camera and a camera mounting bracket is used to acquire images from both sides of the tower and transmit them to the control terminal in real time. The control terminal is used to calculate the ellipticity of the tower based on the received image and determine whether to generate an alarm signal. The control terminal includes: The preprocessing module is used to acquire images of the left and right sides of the tower at time intervals by dual cameras installed on both sides of the tower, and then perform preprocessing by Gaussian filtering and image enhancement in sequence to obtain the preprocessed images of both sides. The pixel width calculation module is used to obtain the straight line equations corresponding to the upper and lower edges of the tower by edge extraction and least squares fitting of the preprocessed images on both sides, and then calculates the instantaneous projected pixel width on both sides. The tower minor axis diameter calculation module is used to introduce scale factor and illumination compensation model, and calculate the actual physical width of both sides by combining the instantaneous projected pixel width of both sides; the tower minor axis diameter is calculated by weighted average method using the actual physical width of both sides. The real-time ellipticity calculation module is used to obtain the deformed major axis of the ellipse based on the minor axis diameter of the tower and the ellipticity calculation model, and further calculate the real-time ellipticity of the tower at the current moment; the ellipticity calculation model is established based on the assumption of equal perimeter. The determination module is used to determine whether the real-time ellipticity meets the upper limit requirement of ellipticity, the amount of ellipticity change requirement, and the ellipticity fluctuation range requirement. If all requirements are met, transportation and monitoring continue, and the cycle S1-S5 is repeated. Otherwise, transportation and monitoring are stopped, and an alarm signal is generated.

Images