Elevator speed governor testing method and device based on infrared stroboflash and phase difference speed measurement

Abstract

The invention relates to the technical field of elevator safety detection, and provides an elevator speed governor detection method and device based on infrared stroboscopic and phase difference speed measurement, and the method comprises the steps: pasting high-reflection infrared mark points at the edge of a speed governor rope wheel at equal intervals, and calculating the initial emission frequency of an infrared stroboscope according to the rated rotation speed; an infrared stroboscope is used for pulse type infrared light irradiation, and a high-precision camera is used for collecting a rope wheel dynamic image sequence; graying processing is carried out on the image sequence, the positions of the mark points in adjacent frames are identified by adopting a dynamic weighting template matching algorithm, and pixel coordinates are obtained; pixel displacement is converted into actual physical arc length displacement through a dynamic mapping model, and a phase difference is calculated; the real-time rotating speed of the speed limiter is calculated according to the phase difference and converted into the real-time linear speed; and the real-time linear speed is compared with a preset speed threshold value for judgment, and a speed limiter action state judgment result is obtained. According to the invention, the detection precision and reliability of the elevator speed governor are improved, and the strict requirements of elevator safety standards are met.

Description

Technical Field

[0001] This invention relates to the field of elevator safety testing technology, and in particular to a method and device for testing elevator speed governors based on infrared stroboscopic and phase difference speed measurement. Background Technology

[0002] Elevator speed governors are critical elevator safety protection devices. When the elevator's operating speed exceeds a preset safety threshold, they can promptly activate the braking mechanism to prevent the elevator from falling out of control. Over long-term use, speed governors are affected by factors such as mechanical wear, environmental corrosion, and dynamic loads, leading to various performance degradation problems such as delayed action, speed measurement deviations, and inaccurate braking response. Elevator speed governor inspection technology, as an important means of elevator safety maintenance, assesses its safety performance status by testing key parameters such as the speed governor's action speed and response time. However, the unique high-speed rotation characteristics, complex mechanical structure, and stringent precision requirements of elevator speed governors present numerous challenges to inspection technology. The key lies in accurately measuring the real-time rotational speed of the speed governor's sheave, precisely detecting the action response time, and achieving high-precision speed threshold determination.

[0003] In existing technologies, elevator speed governor inspection mainly employs traditional mechanical speed measurement and basic photoelectric detection methods to achieve basic speed measurement functions. However, existing methods do not adequately consider the inherent physical correlation mechanism between the transient characteristics of the speed governor during high-speed rotation and the speed measurement accuracy. They also fail to organically integrate objective kinematic constraints with actual speed fluctuation characteristics, resulting in the inability to achieve high-precision and high-reliability automatic inspection of the elevator speed governor's operating status. Summary of the Invention

[0004] In view of this, the present invention proposes an elevator speed governor inspection method and device based on infrared stroboscopic and phase difference speed measurement. This solves the problem that existing methods do not adequately consider the inherent physical correlation mechanism between the transient characteristics of the speed governor during high-speed rotation and the speed measurement accuracy, making it difficult to organically integrate objective kinematic constraints with actual speed fluctuation characteristics, thus failing to achieve high-precision and high-reliability automatic inspection of the elevator speed governor's operating status.

[0005] The technical solution of this invention is implemented as follows: On one hand, this invention provides a method for testing elevator speed governors based on infrared stroboscopic and phase difference speed measurement, including the following steps: Highly reflective infrared markers are equidistantly distributed along the edge of the speed limiter rope wheel to obtain the completed speed limiter rope wheel. The initial emission frequency of the infrared stroboscope is calculated based on the rated speed of the speed limiter. An infrared stroboscope is used to irradiate the speed limiter rope wheel with pulsed infrared light according to the initial emission frequency of the infrared stroboscope, and a high-precision camera is used to collect dynamic image sequences of the speed limiter rope wheel simultaneously. Image processing is performed on the dynamic image sequence of the speed limiter rope wheel to obtain a processed grayscale image sequence. The dynamic weighted template matching algorithm is used to identify the position of the highly reflective infrared marker in the adjacent frame image of the grayscale image sequence, and the pixel coordinates of the highly reflective infrared marker are obtained. The pixel displacement of highly reflective infrared marker points is converted into actual physical arc length displacement through a dynamic mapping model, and the phase difference is calculated based on the actual physical arc length displacement. Calculate the real-time speed limiter speed based on the phase difference, and convert the real-time speed limiter speed into the real-time linear speed limiter. The speed limiter's real-time linear velocity is compared with a preset speed threshold to determine the speed limiter's operating status.

[0006] Based on the above technical solutions, preferably, the step of using an infrared stroboscope to irradiate the speed limiter rope wheel with pulsed infrared light according to the initial emission frequency of the infrared stroboscope, and simultaneously using a high-precision camera to capture a dynamic image sequence of the speed limiter rope wheel, includes: An infrared stroboscope emits adjustable frequency infrared light pulses to illuminate the speed limiter rope wheel, and a high-precision camera externally triggers the synchronization signal of the infrared stroboscope. Based on the change in the speed limiter sheave rotation speed, an adaptive frequency adjustment algorithm is used to adjust the emission frequency of the infrared stroboscope in real time, so that the position of the highly reflective infrared marker points in adjacent frames remains relatively stable or produces a measurable phase shift, thus obtaining a dynamic image sequence of the speed limiter sheave.

[0007] Based on the above technical solutions, preferably, the adaptive frequency adjustment algorithm adopts a real-time rotation speed monitoring mode, and sets the frame rate of the high-precision camera to no less than 1000 frames per second. The rotational speed fluctuation is calculated by initially acquiring the image sequence. When the stroboscopic frequency matches the rotational speed, the position of the highly reflective infrared marker point remains stable in adjacent frame images. When the rotational speed fluctuation causes the frequency to mismatch, the highly reflective infrared marker point generates a phase shift in adjacent frame images. The timing stability is verified by cross-correlation algorithm to ensure that the synchronization error meets the requirements for no less than 1000 consecutive frames, and the adjusted infrared stroboscopic transmitter emission frequency is obtained.

[0008] Based on the above technical solutions, preferably, the step of converting the pixel displacement of the highly reflective infrared marker point into the actual physical arc length displacement through a dynamic mapping model, and calculating the phase difference based on the actual physical arc length displacement, includes: The radial and tangential distortion parameters of the camera are obtained by Zhang's calibration method, and the image coordinates are distorted. The distance between the camera and the speed limiter pulley is obtained in real time using a laser ranging module. A dynamic mapping model is established to correct the dynamic conversion coefficient in real time. The pixel displacement of the distortion-corrected high reflective infrared marker point is converted into the actual physical arc length displacement. The instantaneous angular displacement is calculated based on the actual physical arc length displacement and the pulley radius to obtain the phase difference.

[0009] Based on the above technical solutions, preferably, the dynamic mapping model adopts a spatiotemporal dual-domain cumulative compensation algorithm to perform moving average processing on the phase difference data of multiple consecutive frames to suppress random noise interference. It uses the equidistant circular distribution characteristics of the marker points on the speed limiter sheave to establish phase difference constraints, performs cross-validation and error compensation through the phase difference of each marker point, corrects the sheave radius by combining the temperature stress compensation coefficient, introduces a Kalman filter model to fuse multi-frame speed data, inputs temperature stress environment parameters, and outputs the phase difference after error suppression.

[0010] Based on the above technical solutions, preferably, the step of equidistantly affixing highly reflective infrared markers to the edge of the speed limiter rope wheel to obtain the completed speed limiter rope wheel, and calculating the initial emission frequency of the infrared stroboscope based on the rated speed of the speed limiter, includes: High-purity infrared reflective film material is used to make square-shaped highly reflective infrared markers, which are then matched with the pixel resolution of a high-precision camera to obtain the designed highly reflective infrared markers. The designed highly reflective infrared markers are pasted on the edge of the speed limiter rope wheel according to the principle of equidistant circumference, resulting in the finished speed limiter rope wheel.

[0011] Based on the above technical solutions, preferably, the step of performing image processing on the dynamic image sequence of the speed limiter rope wheel to obtain a processed grayscale image sequence, and using a dynamic weighted template matching algorithm to identify the position of the highly reflective infrared marker point in adjacent frames of the grayscale image sequence to obtain the pixel coordinates of the highly reflective infrared marker point, includes: The dynamic image sequence of the speed limiter rope wheel is converted to grayscale, median filtering is used to remove noise and edge enhancement, and the processed grayscale image sequence is output. A dynamic weighted template matching algorithm combined with an improved Harris corner detection algorithm is used to identify the position of highly reflective infrared markers in adjacent frames of a grayscale image sequence, and obtain the pixel coordinates of the highly reflective infrared markers.

[0012] Based on the above technical solutions, preferably, the step of calculating the real-time speed limiter based on the phase difference and converting the real-time speed limiter speed into the real-time linear velocity of the speed limiter includes: The real-time speed limiter speed is calculated based on the mathematical model of phase difference and speed deviation. A Kalman filter model is introduced to fuse multiple frames of speed data for dynamic correction to obtain the real-time speed limiter speed. Based on the law of circular motion, the real-time rotational speed of the speed limiter is converted into the real-time linear velocity of the speed limiter. A temperature stress compensation coefficient is introduced to correct the radius of the rope wheel, and the real-time linear velocity of the speed limiter is obtained.

[0013] Based on the above technical solutions, preferably, the step of comparing the real-time linear velocity of the speed limiter with a preset speed threshold to obtain the speed limiter's operating state determination result includes: Based on elevator safety standards, electrical and mechanical speed thresholds are determined, and a three-level intelligent speed range determination rule is constructed. The speed range determination result is then combined with the real-time linear velocity output of the speed limiter. Based on the speed range determination result, the time integration algorithm is used to verify the speed duration. Combined with the dual threshold verification, the current speed is compared with the historical speed average to obtain the speed limiter action status determination result.

[0014] On the other hand, the present invention also provides an elevator speed governor testing device based on infrared stroboscopic and phase difference testing, the device comprising: The marker point setting module is used to paste highly reflective infrared marker points at equal intervals on the edge of the speed limiter rope wheel to obtain the finished speed limiter rope wheel, and calculate the initial emission frequency of the infrared stroboscope based on the rated speed of the speed limiter. The illumination acquisition module is used to use an infrared stroboscope to irradiate the speed limiter rope wheel with pulsed infrared light according to the initial emission frequency of the infrared stroboscope, and simultaneously use a high-precision camera to acquire dynamic image sequences of the speed limiter rope wheel. The image processing module is used to process the dynamic image sequence of the speed limiter rope wheel to obtain the processed grayscale image sequence. The dynamic weighted template matching algorithm is used to identify the position of the highly reflective infrared marker in the adjacent frame image of the grayscale image sequence and obtain the pixel coordinates of the highly reflective infrared marker. The phase difference calculation module is used to convert the pixel displacement of highly reflective infrared markers into actual physical arc length displacement through a dynamic mapping model, and calculate the phase difference based on the actual physical arc length displacement. The speed calculation module is used to calculate the real-time speed limiter based on the phase difference and convert the real-time speed limiter speed into the real-time linear speed limiter. The status determination module is used to compare the real-time linear velocity of the speed limiter with the preset speed threshold to determine the speed limiter's action status.

[0015] The elevator speed governor testing method and device based on infrared stroboscopic and phase difference speed measurement of the present invention has the following advantages over the prior art: (1) By integrating infrared strobe technology and phase difference speed measurement algorithm, the motion state of the speed limiter rope wheel is accurately captured by high reflective infrared marker points and the dynamic mapping model is used to convert pixel displacement to physical arc length in real time. The strobe frequency is dynamically adjusted by adaptive frequency adjustment algorithm to achieve accurate measurement of the phase offset of the marker points. The real-time speed is calculated based on the phase difference. The Kalman filter is used to fuse multiple frames of data for dynamic correction, which improves the accuracy and reliability of elevator speed limiter inspection. At the same time, the three-level speed range intelligent judgment rule and time integration algorithm verify that it meets the strict requirements of elevator safety standards. (2) By integrating infrared stroboscopic external trigger synchronization technology with adaptive frequency adjustment algorithm, high frame rate image acquisition and speed fluctuation calculation are performed at a rate of no less than 1000 frames per second using a high-precision camera real-time speed monitoring mode. Combined with cross-correlation algorithm to dynamically verify timing stability, the stroboscopic frequency is adjusted in real time. The synchronization error is precisely controlled for more than 1000 consecutive frames based on the phase offset state of the marker point, which improves the synchronization accuracy and dynamic response capability of speed measurement in elevator speed limiter inspection. (3) By integrating Zhang's calibration distortion correction technology with spatiotemporal dual-domain cumulative compensation algorithm, the laser ranging module is used to perform real-time object distance measurement and dynamic mapping model to accurately convert pixel displacement to physical arc length. The phase difference constraint condition is dynamically established by combining the equidistant circular distribution characteristics of the marker points to achieve cross-validation and error compensation. The radius of the rope wheel is corrected in real time according to the temperature stress environment parameters. The Kalman filter model is used to fuse multiple frames of rotational speed data, which improves the accuracy and stability of phase difference calculation in elevator speed limiter inspection. At the same time, the requirements of high-precision dynamic measurement are met through strict error compensation and noise suppression. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention 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 the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of an elevator speed governor testing method based on infrared stroboscopic and phase difference speed measurement according to the present invention. Figure 2 This is a structural diagram of an elevator speed governor testing device based on infrared stroboscopic and phase difference testing according to the present invention. Figure 3 This is a schematic diagram of a specific embodiment of an elevator speed governor testing device based on infrared stroboscopic and phase difference testing according to the present invention. Detailed Implementation

[0018] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1 This invention provides a method for testing elevator speed governors based on infrared stroboscopic and phase difference speed measurement, comprising the following steps: Highly reflective infrared markers are equidistantly distributed along the edge of the speed limiter rope wheel to obtain the completed speed limiter rope wheel. The initial emission frequency of the infrared stroboscope is calculated based on the rated speed of the speed limiter. An infrared stroboscope is used to irradiate the speed limiter rope wheel with pulsed infrared light according to the initial emission frequency of the infrared stroboscope, and a high-precision camera is used to collect dynamic image sequences of the speed limiter rope wheel simultaneously. Image processing is performed on the dynamic image sequence of the speed limiter rope wheel to obtain a processed grayscale image sequence. The dynamic weighted template matching algorithm is used to identify the position of the highly reflective infrared marker in the adjacent frame image of the grayscale image sequence, and the pixel coordinates of the highly reflective infrared marker are obtained. The pixel displacement of highly reflective infrared marker points is converted into actual physical arc length displacement through a dynamic mapping model, and the phase difference is calculated based on the actual physical arc length displacement. Calculate the real-time speed limiter speed based on the phase difference, and convert the real-time speed limiter speed into the real-time linear speed limiter. The speed limiter's real-time linear velocity is compared with a preset speed threshold to determine the speed limiter's operating status.

[0020] Specifically, this embodiment integrates infrared strobe technology with a phase difference speed measurement algorithm. It utilizes highly reflective infrared markers to accurately capture the motion state of the governor's sheave and a dynamic mapping model to perform real-time conversion of pixel displacement to physical arc length. An adaptive frequency adjustment algorithm dynamically adjusts the strobe frequency to accurately measure the phase shift of the markers, and the real-time rotational speed is calculated based on the phase difference. This embodiment uses Kalman filtering to fuse multi-frame data for dynamic correction, solving the problems of low speed measurement accuracy and insufficient dynamic response in traditional governor inspection methods. This improves the accuracy and reliability of elevator governor inspection. Furthermore, the three-level speed range intelligent judgment rule and time integration algorithm verify that it meets the stringent requirements of elevator safety standards.

[0021] The process of attaching highly reflective infrared markers at equal intervals along the edge of the speed limiter rope wheel to obtain the completed speed limiter rope wheel, and calculating the initial emission frequency of the infrared stroboscope based on the rated speed of the speed limiter, includes: High-purity infrared reflective film material is used to make square-shaped high-reflectivity infrared markers, which are then matched with the pixel resolution of a high-precision camera to obtain the designed high-reflectivity infrared markers.

[0022] In one specific embodiment, the spectral reflectance curve of the high-purity infrared reflective film material highly coincides with the emission spectrum of the infrared stroboscope, and the reflectivity to the target band is greater than 90%. The reflectivity of the steel rope wheel body to infrared light is less than 30%. The infrared stroboscope emission spectrum parameters are input, and the high-purity infrared reflective film material with spectral matching is output.

[0023] The designed highly reflective infrared markers are pasted on the edge of the speed limiter rope wheel according to the principle of equidistant circumference, resulting in the finished speed limiter rope wheel.

[0024] In one specific embodiment, the equidistant distribution of the highly reflective infrared markers on the edge of the speed limiter sheave, combined with the rigid rotation characteristics of the sheave, ensures that the displacement of the markers in adjacent frames strictly corresponds to the angular displacement of the sheave. Input the rigid rotation characteristic parameters of the speed limiter sheave and output the correspondence between displacement and angular displacement.

[0025] The infrared marker material selected in this embodiment is a high-purity infrared reflective film, whose spectral reflectance curve highly coincides with the emission spectrum of the infrared stroboscope. According to the law of spectral reflectance, when infrared light is incident, the reflectivity of the marker point to the target wavelength is >90%, while the reflectivity of the steel rope wheel body to infrared light is <30% due to the surface roughness and molecular vibration absorption characteristics. This ensures that the marker point appears as a significant bright spot in the infrared image captured by the high-speed camera, forming a clear separation from the background. In addition, the infrared marker point is only sensitive to the intensity of infrared light and has extremely low response to visible light and ambient thermal radiation. According to the narrowband filtering theory, this embodiment will install an infrared light-transmitting filter in front of the high-precision camera lens, which can further block non-target wavelength light, so that the marker point is excited only by the pulse light of the infrared stroboscope in the image, which can almost completely eliminate ambient light interference and ensure the purity of the feature point.

[0026] The marker is designed as a square, with a diameter of approximately 3mm, adapted to the pixel resolution of the high-speed camera. Based on the feature extraction theory of computer vision, the marker with a regular geometric shape has clear corner and edge features, and can achieve sub-pixel-level positioning through template matching algorithms, with a positioning accuracy of ±0.1 pixels, which translates to a physical displacement error of <0.01mm. The markers are equidistantly distributed along the edge of the sheave, and considering the rigid rotational characteristics of the sheave, the displacement of the markers in adjacent frames strictly corresponds to the angular displacement of the sheave.

[0027] The theoretical angular displacement formula for the marker points in adjacent frame images is: ; in, This indicates the theoretical angular displacement that rotating components such as the speed limiter pulley should produce within the interval between two consecutive infrared stroboscope triggering events under ideal conditions. Indicates the rated speed of the speed limiter; This refers to the rated emission frequency of the infrared stroboscopic light source. The actual angular displacement is calculated by converting the displacement of the marked pixel, using the following formula: ; in, This represents the actual angular displacement, i.e., the actual rotation angle of the speed limiter sheave within the interval between two adjacent image frames. It represents the conversion factor between pixels and physical dimensions, used to convert the pixel displacement of marked points in an image into actual physical displacement. This represents the pixel displacement of the marker point, specifically the difference in pixel coordinates between highly reflective marker points extracted from two adjacent frames using an image recognition algorithm. This comparability between theoretical and actual values ​​provides a stable kinematic benchmark for phase difference calculation, ensuring the accuracy of rotational speed calculation. This represents the radius of the speed limiter pulley.

[0028] Specifically, this embodiment integrates high-purity infrared reflective film material spectral matching technology with square geometric feature design. It utilizes a target band reflectivity greater than 90% for clear segmentation of the marker point from the background and an infrared light-transmitting filter to eliminate ambient light interference. Combined with the principle of equidistant circular distribution, it dynamically establishes a strict correspondence between displacement and angular displacement to achieve sub-pixel-level positioning accuracy. Furthermore, based on the rigid body rotation characteristics, it verifies the comparability between theoretical and actual angular displacements, achieving a positioning accuracy of ±0.1 pixels through a template matching algorithm. This embodiment solves the problems of severe background interference and insufficient positioning accuracy in traditional marker point recognition methods by applying the spectral reflectance law and narrowband filtering theory, improving the accuracy and stability of marker point recognition in elevator speed governor inspection. Simultaneously, it meets the technical requirements of high-precision speed measurement through strict geometric constraints and kinematic benchmarks.

[0029] The process involves using an infrared stroboscope to emit pulsed infrared light onto the speed limiter pulley at its initial emission frequency, while simultaneously using a high-precision camera to capture a sequence of dynamic images of the speed limiter pulley, including: An infrared stroboscope emits adjustable frequency infrared light pulses to illuminate the speed limiter rope wheel, and a high-precision camera externally triggers the synchronization signal of the infrared stroboscope. Based on the change in the speed limiter sheave rotation speed, an adaptive frequency adjustment algorithm is used to adjust the emission frequency of the infrared stroboscope in real time, so that the position of the highly reflective infrared marker points in adjacent frames remains relatively stable or produces a measurable phase shift, thus obtaining a dynamic image sequence of the speed limiter sheave.

[0030] The adaptive frequency adjustment algorithm adopts a real-time rotation speed monitoring mode, setting the frame rate of the high-precision camera to no less than 1000 frames per second. The rotational speed fluctuation is calculated by initially acquiring the image sequence. When the stroboscopic frequency matches the rotational speed, the position of the highly reflective infrared marker point remains stable in adjacent frame images. When the rotational speed fluctuation causes the frequency to mismatch, the highly reflective infrared marker point generates a phase shift in adjacent frame images. The timing stability is verified by cross-correlation algorithm to ensure that the synchronization error meets the requirements for no less than 1000 consecutive frames, and the adjusted infrared stroboscopic transmitter emission frequency is obtained.

[0031] In one specific embodiment, before conducting synchronous image sequence acquisition, it is necessary to determine the rated speed of the speed limiter. The formula is as follows: ; in, This is a positive integer coefficient used to adjust the matching relationship between the flash frequency and the rotation speed; its physical meaning is the number of flash pulses per revolution of the speed limiter. By calculating using the above formula, the rated emission frequency of the infrared stroboscope can be reasonably set. Meanwhile, the frame rate of the high-speed camera is set to no less than 1000fps, and the external trigger synchronization mode is enabled through the camera control software, so that the camera can receive the synchronization trigger signal from the infrared stroboscope.

[0032] After data acquisition begins, the initial image sequence and marker points can be used to preliminarily calculate the speed fluctuations. This allows for real-time monitoring of the speed limiter pulley speed changes, triggering an adaptive frequency adjustment algorithm based on the following formula: ; In the above formula, express The real-time transmission frequency of the infrared stroboscope, that is, the infrared light pulse transmission frequency that changes with time, is a dynamic control parameter output by the algorithm. It is the frequency adjustment coefficient, used to control the amplitude and sensitivity of frequency adjustment. In engineering, it is usually taken between 0.8 and 1.2. express The real-time rotational speed of the speed limiter pulley is a dynamic speed value calculated in real time using image recognition technology, reflecting the actual operating speed of the pulley. The adjusted frequency is then used by a synchronization circuit to re-trigger the camera, continuously acquiring image sequences. During this process, when the stroboscopic frequency matches the rotational speed, based on the principle of stroboscopic imaging, the highly reflective markers on the pulley appear in an approximately "frozen" state in the image. If fluctuations in rotational speed cause a frequency mismatch, the markers in adjacent frames will experience a phase shift due to the pulley's rotation, providing dynamic data for subsequent phase difference calculations.

[0033] After acquiring a sufficient number of image frames, a cross-correlation algorithm is used to verify temporal stability. This involves examining adjacent image frames. (No. (frame) and (No. (frames), the time offset is calculated using the following formula: ; in, It represents the time offset between any two adjacent image frames, that is, the optimal displacement compensation value in the horizontal direction of the (t+1)th frame image relative to the tth frame image. This is a mathematical operator for finding the optimal solution; in this case, it indicates finding the expression within the parentheses that maximizes the result. value. Represents the coordinates of all pixels in the image. Perform a traversal and summation. Indicates the first Frame image in pixel coordinates The pixel value at that location reflects the brightness information at that position; Indicates the first Frame image at horizontal offset Coordinates after pixels The pixel value at that location.

[0034] Based on the calculation in the above formula, it can be determined that... Is it less than 0.1 pixels? If the offset exceeds the limit, the trigger signal is recalibrated through the redundancy check and retransmission mechanism of the synchronization circuit to ensure that the synchronization error meets the requirements for more than 1000 consecutive frames, thus laying a solid foundation for the accurate calculation of the phase difference. Furthermore, multiple sets are pre-set for the entire dynamic speed range of the speed limiter from start to stop. The corresponding flicker frequency is used to form a frequency adjustment library. During data acquisition, the appropriate frequency is automatically selected based on the real-time rotational speed range to ensure that the phase shift is always kept within a accurately detectable range.

[0035] Specifically, this embodiment integrates infrared stroboscope external trigger synchronization technology with an adaptive frequency adjustment algorithm. It utilizes a high-precision camera in real-time speed monitoring mode to acquire images at a high frame rate of no less than 1000 frames per second and calculate speed fluctuations. Combined with a cross-correlation algorithm, it dynamically verifies timing stability to achieve real-time adjustment of the stroboscope frequency. Furthermore, it precisely controls the synchronization error for more than 1000 consecutive frames based on the phase offset status of the marker points. This embodiment solves the problems of insufficient synchronization accuracy and poor dynamic adaptability in traditional speed measurement methods through a stroboscope frequency and speed matching mechanism, improving the synchronization accuracy and dynamic response capability of speed measurement in elevator speed governor inspection.

[0036] The image processing of the dynamic image sequence of the speed limiter rope wheel yields a processed grayscale image sequence. A dynamic weighted template matching algorithm is then used to identify the position of highly reflective infrared markers in adjacent frames of the grayscale image sequence, obtaining the pixel coordinates of the highly reflective infrared markers. This includes: The dynamic image sequence of the speed limiter rope wheel is converted to grayscale, subjected to median filtering for noise removal and edge enhancement, and the processed grayscale image sequence is output.

[0037] In one specific embodiment, the grayscale conversion adopts a weighted average method, the median filtering replaces the original pixel value with the median of the grayscale values ​​of neighboring pixels, and the edge enhancement processing uses the Sobel operator to calculate the gradient of the image in the horizontal and vertical directions. The input is an RGB color image sequence, and the output is a denoised and enhanced grayscale image sequence.

[0038] The Sobel operator's convolution kernels in the horizontal and vertical directions are shown below: ; ; For each pixel in the image , respectively with and Perform convolution operations to obtain the horizontal gradient. and vertical gradient The calculation formula is as follows: ; ; in, This represents the convolution operation. The pixels in the original image The grayscale value of the pixel is then calculated. The gradient magnitude of that pixel is then calculated. As shown below: ; By calculating the gradient magnitude, the edge characteristics of objects in the image are enhanced, making the outlines of the speed limiter pulley and the action switch clearer, thus providing more accurate image data for subsequent feature extraction. Through the series of operations described above, including camera installation, video capture, image extraction, and processing, high-quality image data of the speed limiter's working process can be obtained, laying the foundation for subsequent analysis of the speed limiter's working status and characteristics.

[0039] A dynamic weighted template matching algorithm combined with an improved Harris corner detection algorithm is used to identify the position of highly reflective infrared markers in adjacent frames of a grayscale image sequence, and obtain the pixel coordinates of the highly reflective infrared markers.

[0040] In one specific embodiment, the dynamic weighted template matching algorithm constructs a matching model that integrates prior shape constraints and adaptive grayscale weights, and the improved Harris corner detection algorithm introduces curvature constraints to suppress edge point misjudgment, achieving sub-pixel level corner positioning accuracy. By inputting the template and the image to be matched, the pixel coordinates of highly reflective infrared marker points with sub-pixel accuracy are obtained.

[0041] The dynamic weighted template matching algorithm constructs a matching model that integrates "prior shape constraints + adaptive grayscale weights": Let the ideal template for the marker point be... The size is The image to be matched is Define dynamic weight function The formula describing the contribution of pixel location to the matching is shown below: ; in, and template The formulas for calculating the mean and variance of gray levels are shown below: ; ; Its purpose is to achieve grayscale consistency constraints; The distance from the pixel to the center of the region of interest (ROI) is the Euclidean distance, which serves to strengthen the region constraint; while This is the grayscale constraint coefficient, which, after experimental calibration, is set to 0.8. This is the area constraint coefficient, calibrated to a value of 1.2. Its function is to strengthen the matching priority of the marker points in the rope wheel edge area. This is achieved through a sliding template. In the image Calculate the weighted cross-correlation similarity. Locate the position of the marker point.

[0042] ; in, template Offset on the image. By searching for the maximum. Determine the largest This refers to the location of the marker point. This algorithm can significantly improve the noise robustness of infrared images, effectively avoiding matching drift caused by noise. Furthermore, to achieve high-precision marker point localization, this embodiment improves the classic Harris corner detection algorithm by introducing curvature constraints to suppress false edge point detection, proposing an improved Harris algorithm for sub-pixel corner detection to further enhance corner localization accuracy. The first step is corner response function reconstruction. The traditional Harris 9-point corner response function is calculated based on the image gradient covariance matrix M as shown below: ; ; In the above formula It is the gradient covariance matrix, used to statistically analyze the distribution characteristics of gradients within local regions of an image. This represents the image frame processed in the previous step, which is the input image for this step; and The images are respectively in The gradient in direction reflects the rate of change of image grayscale in the horizontal and vertical directions, and the formula is shown below: ; ; For matrix The determinant, The trace of the matrix; This is an empirical coefficient, an adjustable parameter in the Harris corner detection algorithm, typically ranging from 0.04 to 0.06. It is used to balance the determinant and adjust the sensitivity of corner detection. This is the Harris corner response value, used to determine whether a pixel is a corner. A large positive value indicates that the grayscale around the pixel changes drastically in both directions, suggesting a high probability that it is a corner pixel; when When the value is small, it indicates that the pixel may be an edge or a flat area, and therefore is not a corner point.

[0043] This embodiment introduces image curvature constraints and calculates the curvature information at each pixel using the Laplacian operator, the calculation formula of which is shown below: ; The formula for the improved corner response function R is defined as follows: ; in, is the Laplacian operator for images, used to calculate the second derivative of image grayscale. and It is an image The second-order partial derivatives in the horizontal and vertical directions reflect the rate of change of gray level in the horizontal direction; It is the improved Harris corner response value, obtained by introducing the Laplace operator to modify the original response. The purpose of this correction is to suppress the misidentification of edge points as corner points, making corner point detection more accurate. These are curvature constraint coefficients, used to control the response value of the Laplacian operator. The correction strength. It is the absolute value of the Laplacian operator result, used to measure the intensity of the second-order gray-level change in a local region of the image. Yes, in the whole image The maximum value is used for... Perform normalization processing, so that It falls within the [0,1] interval to ensure the stability of the exponential calculation.

[0044] According to the principles of optical imaging, a suitable distance and angle should be maintained between the camera and the subject. Let the focal length of the camera be F, the distance from the subject to the camera be U, and the distance from the imaging plane to the optical center of the camera be V. According to the Gaussian imaging formula: ; To obtain a clear and complete image of the speed limiter pulley and actuation switch, proper adjustments are required. and The value should be such that the image is clear and fills the entire frame. At the same time, the installation angle should ensure that the camera's optical axis is perpendicular to the plane where the speed limiter pulley and the action switch are located, to avoid perspective distortion in the image.

[0045] In actual installation, this embodiment employs a three-axis motion platform and inputs precise coordinates to ensure accurate camera positioning, aligning the lens with the center of the speed limiter's key components. Furthermore, environmental factors, such as lighting conditions, must be considered. To ensure the camera is installed in a well-lit and stable area, this embodiment equips the camera with an illumination component to guarantee the brightness and contrast of the captured image. When using the camera for recording, appropriate parameters must be selected; its resolution determines the clarity of the captured image. A high-resolution camera should be used to ensure the subtle movements and characteristics of the speed limiter's pulley and actuation switch are captured. Regarding frame rate, since the speed limiter may experience rapid changes in movement during operation, the frame rate should be no less than 60fps to fully record its working state. During recording, the camera is activated to begin recording, completely documenting the entire working process of the speed limiter from start to stop, generating a single video file.

[0046] Then, video processing software can be used to extract each frame from the generated video file. Each extracted frame will serve as the basis for subsequent processing, used to analyze the speed limiter's operating status and characteristics, and for subsequent speed verification. After extracting each frame, it needs to be processed and calculated. Color images contain three color channels (RGB), making the processing more complex, and color information is not a critical factor for speed limiter feature extraction. Therefore, converting the color image to a grayscale image simplifies the calculations and improves processing efficiency.

[0047] The grayscale conversion uses a weighted average method, and the conversion formula is as follows: ; In the above formula Represents a point in the image The converted grayscale value, and Representing points respectively The three monochromatic components are red, green, and blue. Since the human eye is most sensitive to green, followed by red, and least sensitive to blue, they are given different weights.

[0048] Suppose a pixel in the image The neighborhood of is a The window sorts the grayscale values ​​of all pixels within the window and takes the median value as the pixel value. The new grayscale value, that is: ; in, The pixels in the original image grayscale value, Represents pixels The neighborhood, This indicates the pixel value after median filtering. The grayscale value. Median filtering can effectively remove impulse noise such as salt-and-pepper noise while preserving the edge information of the image. Specifically, this embodiment integrates weighted average grayscale conversion with median filtering noise removal techniques. It utilizes the Sobel operator for edge enhancement and a dynamic weighted template matching algorithm for prior shape constraints and adaptive grayscale weight fusion. An improved Harris corner detection algorithm dynamically introduces curvature constraints to suppress edge point misjudgments, achieving sub-pixel-level corner location accuracy. Furthermore, the corner response function is corrected using the Laplacian operator, and the marker position is determined by calculating weighted cross-correlation similarity using a sliding template. This embodiment addresses the problems of severe noise interference and insufficient corner location accuracy in traditional image processing methods through gradient covariance matrix and curvature constraint mechanisms. It improves the robustness and positioning accuracy of highly reflective infrared marker identification in elevator speed limiter inspection, while meeting high-precision dynamic measurement requirements through rigorous sub-pixel-level positioning algorithms and noise suppression processing.

[0049] The process of converting the pixel displacement of highly reflective infrared marker points into actual physical arc length displacement using a dynamic mapping model, and calculating the phase difference based on the actual physical arc length displacement, includes: The radial and tangential distortion parameters of the camera are obtained by Zhang's calibration method, and the image coordinates are distorted. The distance between the camera and the speed limiter pulley is obtained in real time using a laser ranging module. A dynamic mapping model is established to correct the dynamic conversion coefficient in real time. The pixel displacement of the distortion-corrected high reflective infrared marker point is converted into the actual physical arc length displacement. The instantaneous angular displacement is calculated based on the actual physical arc length displacement and the pulley radius to obtain the phase difference.

[0050] The dynamic mapping model employs a spatiotemporal dual-domain cumulative compensation algorithm to perform moving average processing on phase difference data from multiple consecutive frames to suppress random noise interference. It utilizes the equidistant circular distribution characteristics of the marker points on the speed limiter pulley to establish phase difference constraints. Cross-validation and error compensation are performed using the phase differences of each marker point. The pulley radius is corrected by combining the temperature stress compensation coefficient. A Kalman filter model is introduced to fuse multi-frame rotational speed data. Temperature stress environment parameters are input, and the phase difference after error suppression is output.

[0051] In one specific embodiment, the radial distortion parameters of the camera are obtained using the Zhang calibration method. ,in The radial distortion coefficient is... is the tangential distortion coefficient. For coordinates in the image... The pixel, its distortion-corrected coordinates The following conditions must be met: ; ; ; in, and This represents the original coordinates of a pixel in the image, i.e., the coordinates before distortion; and This represents the pixel coordinates after distortion correction. It is the theoretical position of the pixel in an ideal distortion-free image after eliminating radial and tangential distortion of the camera. It is used for subsequent operations such as accurately extracting marker points and calculating phase differences. It is the radial distance from a pixel to the principal point of the image; and It is the radial distortion coefficient, used to describe the radial distortion of a camera lens. This typically corresponds to low-order radial distortion. For higher-order radial distortion, the values ​​are obtained through camera calibration. Different lenses have different radial distortion coefficients, so they need to be calibrated in advance. and It is the tangential distortion coefficient, used to describe the tangential distortion of a camera lens, caused by mechanical errors such as the lens not being parallel to the image plane.

[0052] Then, the laser ranging module is used to obtain the object distance between the camera and the speed limiter pulley in real time. Combined with the camera focal length, the actual physical arc length corresponding to the pixel displacement can be calculated using the following formula: ; ; In the above formula, yes The dynamic conversion coefficient between pixels and physical size at any given time is used to convert pixel position to actual physical arc length; It means to indicate The actual physical arc length displacement of the time marker is the arc length generated by the actual rotation of the infrared marker on the speed limiter pulley within the interval between adjacent frame image acquisitions; It is the pixel displacement of the marker point, that is, the change in pixel coordinates of the infrared marker point extracted in adjacent frames of the image through the above image recognition algorithm; this model eliminates the systematic error caused by the fixed conversion coefficient, and has been theoretically verified to greatly reduce the error in arc length calculation, thus laying the foundation for accurate phase difference calculation.

[0053] Traditional phase difference calculations only focus on angular displacement. This embodiment constructs a multi-dimensional solution model of angular displacement, angular velocity, and angular acceleration to comprehensively analyze the dynamic motion characteristics of the speed limiter pulley. Let the radius of the speed limiter pulley be... The time interval between adjacent image frames The formula is shown below: ; Actual arc length obtained based on dynamic mapping model The formulas for instantaneous angular displacement, instantaneous angular velocity, and angular acceleration are shown below: ; ; ; It is an instantaneous angular displacement used to describe the change in angular displacement of the sheave between adjacent images, and the unit is radians (rad). It reflects the angular velocity of the pulley at the current moment, and the unit is radians per second (rad / s). The angular acceleration of the rope pulley is expressed in radians per second squared (rad / s²).

[0054] To reduce the impact of random errors on phase difference calculation, this embodiment proposes a spatiotemporal dual-domain cumulative compensation algorithm based on the above calculations, combining temporal domain moving average and spatial domain circular symmetry constraints. First, a moving average is applied to the phase difference data of N consecutive frames to obtain phase difference data for multiple consecutive frames after the moving average processing. : ; By statistically averaging multiple data points, random noise interference is suppressed, improving the stability of phase difference calculation. Then, the equidistant circular distribution characteristic of the marker points on the speed limiter pulley is utilized (assuming the number of marker points is...). The central angle between adjacent marker points is The phase difference constraint conditions are established as follows: ; ; In the above formula, For the rotation period of the rope wheel, For the first The phase difference of each marker point. Through this constraint, the phase difference of each marker point can be cross-validated and error compensated. The test results show that the random error suppression rate can reach more than 85%, ensuring that the phase difference calculation accuracy is less than or equal to 0.01 radians.

[0055] In the motion analysis of the speed limiter pulley, its rotation follows the laws of circular motion. The angular velocity of the pulley is... The linear velocity is The radius of the rope pulley is Then the angular velocity and linear velocity satisfy the classical relationship: ; angular velocity It reflects the angle through which the pulley rotates per unit time, while the rotational speed... The number of rotations of the rope pulley per unit time is described by the circumferential angle. Establishing a connection yields: ; This formula will change the rotational speed. Convert to angular velocity This lays the foundation for deriving the rotational speed from the phase difference, and the accurate acquisition of the phase difference relies on the high-precision results of the marker point identification mentioned above. Next is the derivation of the rotational speed deviation, based on the frequency of the infrared strobe. The time interval between adjacent frame acquisitions is determined. At the same time, the number of flashes per revolution The infrared stroboscope emits light for each revolution of the rope pulley. The next light pulse. Substituting into the phase difference formula, we obtain the phase difference. : .

[0056] Specifically, this embodiment integrates Zhang's calibration distortion correction technology with a spatiotemporal dual-domain cumulative compensation algorithm. It utilizes a laser ranging module for real-time object distance measurement and a dynamic mapping model for precise conversion of pixel displacement to physical arc length. By combining the equidistant circular distribution characteristics of marker points, it dynamically establishes phase difference constraints to achieve cross-validation and error compensation. Furthermore, it performs real-time correction of the sheave radius based on temperature and stress environmental parameters and fuses multi-frame rotational speed data using a Kalman filter model. This embodiment addresses the insufficient measurement accuracy caused by camera distortion and environmental interference in traditional measurement methods through moving average processing and random noise suppression mechanisms. This improves the accuracy and stability of phase difference calculation in elevator speed governor inspection, while rigorous error compensation and noise suppression meet the requirements of high-precision dynamic measurement.

[0057] The step of calculating the real-time speed limiter based on the phase difference and converting the real-time speed limiter speed into the real-time linear velocity of the speed limiter includes: The real-time speed limiter speed is calculated based on a mathematical model of phase difference and speed deviation. A Kalman filter model is then introduced to fuse multiple frames of speed data for dynamic correction, resulting in the real-time speed limiter speed.

[0058] In one specific embodiment, a state vector and a state transition matrix are defined using a Kalman filter model. Based on the speed calculation model and the observation equation, the speed is dynamically corrected by iteratively updating the state vector through filtering, thus suppressing the influence of random errors on speed calculation. Multiple frames of phase difference data are input, and the real-time speed of the speed limiter after filtering correction is output.

[0059] Based on the law of circular motion, the real-time rotational speed of the speed limiter is converted into the real-time linear velocity of the speed limiter. A temperature stress compensation coefficient is introduced to correct the radius of the rope wheel, and the real-time linear velocity of the speed limiter is obtained.

[0060] In one specific embodiment, considering the material expansion and mechanical stress of the speed limiter sheave caused by temperature changes, the sheave radius is corrected based on the temperature stress compensation coefficient calibrated by material mechanics experiments. The real-time linear velocity of the speed limiter is calculated through the direct correlation between linear velocity and rotational speed, thus solving the system error caused by environmental factors. The temperature stress environmental parameters are input, and the real-time linear velocity of the speed limiter after environmental compensation is output.

[0061] Through formula and formula The formula for speed deviation is derived as follows: ; ; ; Due to the formula This holds true under ideal conditions, therefore it is retained. As a characteristic parameter of the number of flashes per revolution, and Accurate calculations rely on the precise coordinates and phase difference calculation model for marker point identification introduced earlier. The real-time speed limiter... From the theoretical value of rated speed With speed deviation The result of superposition is: ; This model deeply correlates image features, system parameters, and motion parameters, achieving a leap from "visual perception" to "motion quantization." The high-quality data stems from the innovative algorithms and error suppression mechanisms used in marker recognition and phase difference calculation. Traditional phase difference calculation is based on pixel-level displacement; this embodiment introduces a sub-pixel interpolation algorithm to improve accuracy. Let the pixel displacement of the marker point in adjacent frames be... The conversion factor between image pixels and physical size is: Subpixel-level phase difference satisfy: ; In the formula, This sub-pixel level displacement reduces the phase difference calculation error to ±0.01 rad. The sub-pixel corner detection results, based on the aforementioned marker point recognition, ensure high accuracy of the phase difference. Furthermore, to suppress random errors, this embodiment constructs a Kalman filter model to fuse multi-frame rotational speed data. A state vector is defined. and state transition equations As shown below: ; ; ; In the above formula, For a custom state vector, It is a state transition matrix, used in speed limiter detection to describe the system state transition from... arrive The time-shifting pattern; and the observation equation is based on the rotational speed calculation model: ; in, yes The observed value at a given time is a value obtained through actual measurement. Due to factors such as measurement noise, it differs from the actual rotational speed. There is a discrepancy. The noise represents the error introduced during the observation process. By iteratively updating the state vector through this Kalman filter, the rotational speed is dynamically corrected, thereby greatly improving the accuracy of rotational speed measurement.

[0062] This embodiment also considers the effects of material expansion and mechanical stress on the governor sheave due to temperature changes during elevator operation, and introduces a temperature-stress compensation coefficient. The radius of the rope pulley after correction by this coefficient is: ; In the formula, Rated temperature The radius of the lower rope pulley, This is the temperature-stress compensation coefficient, calibrated through materials mechanics experiments. The corrected rotational speed model is shown below: ; In the above formula, The model represents the speed limiter rotational speed after temperature-stress compensation. It resolves systematic errors caused by environmental factors, ensures measurement accuracy under different temperature ranges and varying stress scenarios, and thus guarantees the accuracy of subsequent speed determination for the rope pulley components.

[0063] Specifically, this embodiment integrates a mathematical model of phase difference and rotational speed deviation with a Kalman filter dynamic correction algorithm. It utilizes sub-pixel interpolation for ±0.01 rad phase difference calculation and iterative state vector updates to fuse multi-frame rotational speed data. Combined with a temperature stress compensation coefficient, it dynamically corrects the sheave radius to compensate for systemic errors caused by environmental factors. Furthermore, it performs real-time linear velocity conversion of the governor's rotational speed based on the laws of circular motion. A compensation model calibrated through materials mechanics experiments ensures measurement accuracy under varying temperature and stress scenarios. This embodiment addresses the problems of severe random error interference and systemic errors caused by environmental factors in traditional rotational speed measurement methods through filtering iteration and environmental compensation mechanisms, improving the accuracy and environmental adaptability of rotational speed calculation in elevator governor inspection.

[0064] The step of comparing the real-time linear velocity of the speed limiter with a preset speed threshold to obtain the speed limiter's operating state determination result includes: Based on elevator safety standards, electrical and mechanical speed thresholds are determined, and a three-level intelligent speed range determination rule is constructed. The speed range determination result is then combined with the real-time linear velocity output of the speed governor.

[0065] In one specific embodiment, the electrical action speed threshold is determined based on the elevator's rated speed, satisfying the range from 1.15 times the rated speed to the mechanical action speed threshold. The mechanical action speed threshold follows the standard of not exceeding 1.5 times the rated speed or the rated speed plus 0.5 meters per second, depending on the rated speed. Kalman filtering is used to track speed fluctuations and identify potential anomalies. The elevator's rated speed is input, and standardized electrical and mechanical action speed thresholds are output.

[0066] Based on the speed range determination result, the time integration algorithm is used to verify the speed duration. Combined with the dual threshold verification, the current speed is compared with the historical speed average to obtain the speed limiter action status determination result.

[0067] In one specific embodiment, when the speed is determined to be in an electrical action range, the speed duration is verified to be no less than 0.2 seconds using a time integration algorithm. When the speed is determined to be in a mechanical action range, a dual threshold verification is introduced to compare the current speed with the average speed of the past three frames and the duration is no less than 0.1 seconds. After confirming the overspeed trend, the corresponding braking command is output, historical speed data is input, and the speed limiter action status determination result after time verification is output.

[0068] The direct relationship between linear velocity and rotational speed is shown below: ; Substituting the formula from the previous text into the above equation, we obtain the fused phase difference. flicker frequency Number of flashes per revolution With the radius of the rope wheel To further refine the accuracy of the linear velocity and cover environmental and measurement errors, the final linear velocity solution model is shown below: ; in, This refers to the linear velocity of the speed limiter pulley component under normal conditions. This involves solving for the final linear velocity after introducing multiple modified models. The real-time linear velocity of the speed limiter is calculated based on the above formula. Subsequently, automated speed determination requires deep integration of elevator safety standards, speed governor operating mechanisms, and intelligent algorithms to construct a complete "perception-decision-execution" closed loop. Firstly, based on standards such as GB7588, and in conjunction with the elevator's rated speed... Determine the baseline threshold and the electrical action speed threshold. Must meet Mechanical motion speed threshold Follow according to different rated speeds or .

[0069] Next, a three-level intelligent speed range determination rule is constructed: when If the system is deemed to be operating normally, it uses a Kalman filter to track speed fluctuations and identify potential anomalies; when... The electrical action is triggered by a time integration algorithm. ; Verify the speed duration, if it meets the requirements The system determines that the electrical triggering mechanism has activated and outputs a braking request to the elevator control system. The mechanical action is triggered, and a dual threshold check is introduced. The current speed is compared with the average speed of the past 3 frames. If the condition is met and continues... Upon confirming the overspeed trend, a mechanical braking command is output to drive the speed limiter in conjunction.

[0070] The judgment result is not isolated but deeply integrated with the system: when an electrical action is triggered, a pre-alarm is activated, outputting a low-frequency audible and visual warning and uploading data; when a mechanical action is triggered, an emergency alarm is triggered, outputting a high-frequency audible and visual warning and uploading data; if the speed continues to rise after electrical braking, the fault-tolerant deceleration mode is automatically switched, the traction machine frequency is adjusted to pull back to the safe range, and the curve is recorded; after judgment, the triggering mechanism self-check is initiated, injecting virtual speed to verify the response and effectiveness, forming a "detection-judgment-verification" closed loop. This judgment logic originates from the speed governor's graded braking mechanism and the state machine model of control theory. It simulates the physical triggering process through interval division and uses a compensation model and verification to suppress the influence of environment and noise. In engineering applications, combined with sub-pixel phase difference and dynamic compensation, the error is controlled within ±0.03m / s, the judgment time is ≤5ms, and machine learning is integrated to reduce the false alarm rate to below 0.1%, achieving accurate, real-time, and intelligent judgment. This provides core decision support for elevator overspeed protection and becomes a key hub for the intelligent detection system from "perception" to "execution," strengthening the elevator safety operation defense line.

[0071] Specifically, this embodiment integrates a three-level speed range intelligent judgment rule with a time integration algorithm verification mechanism. It utilizes Kalman filtering for speed fluctuation tracking and identification of potential anomalies, and dual-threshold verification to compare the current speed with the historical speed average. Combined with elevator safety standards, it dynamically determines electrical and mechanical action speed thresholds to achieve standardized braking judgment. Furthermore, based on a perception-decision-execution closed-loop mechanism, it performs real-time judgment of the speed governor's action state, ensuring high-precision real-time response through ±0.03m / s error control and a judgment time of ≤5ms. This embodiment solves the problems of insufficient speed judgment accuracy and response time delay in traditional speed governor inspection methods through a graded braking mechanism and a state machine model, improving the accuracy and real-time performance of overspeed protection judgment in elevator speed governor inspection. Simultaneously, it meets the stringent technical requirements of elevator safety standards through a false alarm rate control of less than 0.1% and a rigorous time verification algorithm.

[0072] Please see Figure 2 The present invention also provides an elevator speed governor testing device based on infrared stroboscopic and phase difference testing, the device comprising: The marker point setting module is used to paste highly reflective infrared marker points at equal intervals on the edge of the speed limiter rope wheel to obtain the finished speed limiter rope wheel, and calculate the initial emission frequency of the infrared stroboscope based on the rated speed of the speed limiter. The illumination acquisition module is used to use an infrared stroboscope to irradiate the speed limiter rope wheel with pulsed infrared light according to the initial emission frequency of the infrared stroboscope, and simultaneously use a high-precision camera to acquire dynamic image sequences of the speed limiter rope wheel. The image processing module is used to process the dynamic image sequence of the speed limiter rope wheel to obtain the processed grayscale image sequence. The dynamic weighted template matching algorithm is used to identify the position of the highly reflective infrared marker in the adjacent frame image of the grayscale image sequence and obtain the pixel coordinates of the highly reflective infrared marker. The phase difference calculation module is used to convert the pixel displacement of highly reflective infrared markers into actual physical arc length displacement through a dynamic mapping model, and calculate the phase difference based on the actual physical arc length displacement. The speed calculation module is used to calculate the real-time speed limiter based on the phase difference and convert the real-time speed limiter speed into the real-time linear speed limiter. The status determination module is used to compare the real-time linear velocity of the speed limiter with the preset speed threshold to determine the speed limiter's action status.

[0073] In one specific embodiment, please refer to Figure 3 The elevator speed governor testing device based on infrared stroboscopic and phase difference testing includes: 1. The support frame is the core component of the speed governor, ensuring accurate relative positioning of all parts, reducing the impact of elevator vibration on the speed governor, guaranteeing the stability and reliability of the mechanical transmission, and thus preventing false triggering or component wear. 2. The switch is the elevator's operating speed. When the elevator speed exceeds 115% of the rated speed (as stipulated by national standards), the switch triggers the electrical safety circuit, cutting off the elevator's power supply and stopping the elevator. 3. The infrared stroboscope emits adjustable-frequency infrared pulses to illuminate the elevator speed governor wheel. Working in conjunction with a high-speed camera, it captures images of the wheel and uses the stroboscope effect to generate a phase difference, providing a key light source and dynamic characteristic basis for subsequent calculation of the speed governor's real-time speed based on the phase difference, achieving automated and accurate detection. 4. The sheave is connected to the elevator car or counterweight via the speed governor rope and rotates with the elevator. Its rotation speed directly reflects the elevator speed, providing speed signals for the switch and braking components. 5. The lighting is the sufficient illumination environment for the camera to obtain clearer and more stable images. 6 is a high-precision camera, its function is to acquire clear video of the speed governor rope pulley component and extract a large number of image frames for subsequent processing and calculation. 7 is a three-axis motion platform, its purpose is to increase the range of images acquired by the camera, thereby increasing the accuracy of the speed governor inspection. 8 is a rope-pressing component; when the elevator overspeeds and the switch is triggered, the rope-pressing component uses mechanical force to press the speed governor rope tightly onto the pulley, generating friction to forcibly decelerate or brake the car. 9 is a manual reset device; when the speed governor is triggered due to overspeed, the reset device must be manually operated to return all components to their initial state, release the brakes, and restore the electrical circuit so that the elevator can resume operation. 10 is an extension bracket, its function is to support and fix the infrared stroboscope, ensuring the coordination of the equipment structure and function, and adapting to testing requirements. 11 is a base, which serves as the basic support structure for the speed governor, fixing the entire speed governor device and ensuring its stability during elevator operation, preventing displacement due to vibration or external forces.

[0074] Specifically, this embodiment integrates modular functional design and an integrated hardware architecture, utilizing six core modules for fully automated inspection of the entire process, including marker point setting, illumination acquisition, image processing, phase difference calculation, velocity calculation, and state determination. It combines the synchronous coordination of an infrared stroboscope and a high-precision camera with a three-axis motion platform to dynamically adjust the detection range, achieving multi-dimensional precise measurement. Furthermore, the overall device's stability is ensured by a support base system, and the consistency and reliability of the inspection environment are guaranteed through illumination compensation and extended supports. This embodiment addresses the problems of limited functionality, insufficient accuracy, and poor environmental adaptability in traditional speed governor inspection devices through modular division of labor and hardware integration optimization, thereby improving the automation level and detection accuracy of elevator speed governor inspection devices.

[0075] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for elevator overspeed governor inspection based on infrared stroboscopic and phase difference velocity measurement, characterized in that, Includes the following steps: Highly reflective infrared markers are equidistantly distributed along the edge of the speed limiter rope wheel to obtain the completed speed limiter rope wheel. The initial emission frequency of the infrared stroboscope is calculated based on the rated speed of the speed limiter. An infrared stroboscope is used to irradiate the speed limiter rope wheel with pulsed infrared light according to the initial emission frequency of the infrared stroboscope, and a high-precision camera is used to collect dynamic image sequences of the speed limiter rope wheel simultaneously. Image processing is performed on the dynamic image sequence of the speed limiter rope wheel to obtain a processed grayscale image sequence. The dynamic weighted template matching algorithm is used to identify the position of the highly reflective infrared marker in the adjacent frame image of the grayscale image sequence, and the pixel coordinates of the highly reflective infrared marker are obtained. The pixel displacement of highly reflective infrared marker points is converted into actual physical arc length displacement through a dynamic mapping model, and the phase difference is calculated based on the actual physical arc length displacement. Calculate the real-time speed limiter speed based on the phase difference, and convert the real-time speed limiter speed into the real-time linear speed limiter. The real-time linear velocity of the speed limiter is compared with the preset speed threshold to determine the speed limiter's operating status.

2. The elevator overspeed governor inspection method based on infrared stroboscopic and phase difference velocity measurement according to claim 1, characterized in that, The process involves using an infrared stroboscope to emit pulsed infrared light onto the speed limiter pulley at its initial emission frequency, while simultaneously using a high-precision camera to capture a sequence of dynamic images of the speed limiter pulley, including: An infrared stroboscope emits adjustable frequency infrared light pulses to illuminate the speed limiter rope wheel, and a high-precision camera externally triggers the synchronization signal of the infrared stroboscope. Based on the change in the speed limiter sheave rotation speed, an adaptive frequency adjustment algorithm is used to adjust the emission frequency of the infrared stroboscope in real time, so that the position of the highly reflective infrared marker points in adjacent frames remains relatively stable or produces a measurable phase shift, thus obtaining a dynamic image sequence of the speed limiter sheave.

3. The elevator overspeed governor inspection method based on infrared stroboscopic and phase difference velocity measurement according to claim 2, characterized in that, The adaptive frequency adjustment algorithm adopts a real-time rotation speed monitoring mode, setting the frame rate of the high-precision camera to no less than 1000 frames per second. The rotational speed fluctuation is calculated by initially acquiring the image sequence. When the stroboscopic frequency matches the rotational speed, the position of the highly reflective infrared marker point remains stable in adjacent frame images. When the rotational speed fluctuation causes the frequency to mismatch, the highly reflective infrared marker point generates a phase shift in adjacent frame images. The timing stability is verified by cross-correlation algorithm to ensure that the synchronization error meets the requirements for no less than 1000 consecutive frames, and the adjusted infrared stroboscopic transmitter emission frequency is obtained.

4. The elevator overspeed governor inspection method based on infrared stroboscopic and phase difference velocity measurement according to claim 1, characterized in that, The process of converting the pixel displacement of highly reflective infrared marker points into actual physical arc length displacement using a dynamic mapping model, and calculating the phase difference based on the actual physical arc length displacement, includes: The radial and tangential distortion parameters of the camera are obtained by Zhang's calibration method, and the image coordinates are distorted. The distance between the camera and the speed limiter pulley is obtained in real time using a laser ranging module. A dynamic mapping model is established to correct the dynamic conversion coefficient in real time. The pixel displacement of the distortion-corrected high reflective infrared marker point is converted into the actual physical arc length displacement. The instantaneous angular displacement is calculated based on the actual physical arc length displacement and the pulley radius to obtain the phase difference.

5. The elevator overspeed governor inspection method based on infrared stroboscopic and phase difference velocity measurement according to claim 4, characterized in that, The dynamic mapping model employs a spatiotemporal dual-domain cumulative compensation algorithm to perform moving average processing on phase difference data from multiple consecutive frames to suppress random noise interference. It utilizes the equidistant circular distribution characteristics of the marker points on the speed limiter pulley to establish phase difference constraints. Cross-validation and error compensation are performed using the phase differences of each marker point. The pulley radius is corrected by combining the temperature stress compensation coefficient. A Kalman filter model is introduced to fuse multi-frame rotational speed data. Temperature stress environment parameters are input, and the phase difference after error suppression is output.

6. The elevator overspeed governor inspection method based on infrared stroboscopic and phase difference velocity measurement according to claim 1, characterized in that, The process of attaching highly reflective infrared markers at equal intervals along the edge of the speed limiter rope wheel to obtain the completed speed limiter rope wheel, and calculating the initial emission frequency of the infrared stroboscope based on the rated speed of the speed limiter, includes: High-purity infrared reflective film material is used to make square-shaped highly reflective infrared markers, which are then matched with the pixel resolution of a high-precision camera to obtain the designed highly reflective infrared markers. The designed highly reflective infrared markers are pasted on the edge of the speed limiter rope wheel according to the principle of equidistant circumference, resulting in the finished speed limiter rope wheel.

7. The elevator overspeed governor inspection method based on infrared stroboscopic and phase difference velocity measurement according to claim 1, characterized in that, The image processing of the dynamic image sequence of the speed limiter rope wheel yields a processed grayscale image sequence. A dynamic weighted template matching algorithm is then used to identify the position of highly reflective infrared markers in adjacent frames of the grayscale image sequence, obtaining the pixel coordinates of the highly reflective infrared markers. This includes: The dynamic image sequence of the speed limiter rope wheel is converted to grayscale, median filtering is used to remove noise and edge enhancement, and the processed grayscale image sequence is output. A dynamic weighted template matching algorithm combined with an improved Harris corner detection algorithm is used to identify the position of highly reflective infrared markers in adjacent frames of a grayscale image sequence, and obtain the pixel coordinates of the highly reflective infrared markers.

8. The elevator speed governor testing method based on infrared stroboscopic and phase difference speed measurement as described in claim 1, characterized in that, The step of calculating the real-time speed limiter based on the phase difference and converting the real-time speed limiter speed into the real-time linear velocity of the speed limiter includes: The real-time speed limiter speed is calculated based on the mathematical model of phase difference and speed deviation. A Kalman filter model is introduced to fuse multiple frames of speed data for dynamic correction to obtain the real-time speed limiter speed. Based on the law of circular motion, the real-time rotational speed of the speed limiter is converted into the real-time linear velocity of the speed limiter. A temperature stress compensation coefficient is introduced to correct the radius of the rope wheel, and the real-time linear velocity of the speed limiter is obtained.

9. The elevator speed governor testing method based on infrared stroboscopic and phase difference speed measurement as described in claim 1, characterized in that, The step of comparing the real-time linear velocity of the speed limiter with a preset speed threshold to obtain the speed limiter's operating state determination result includes: Based on elevator safety standards, electrical and mechanical speed thresholds are determined, and a three-level intelligent speed range determination rule is constructed. The speed range determination result is then combined with the real-time linear velocity output of the speed limiter. Based on the speed range determination result, the time integration algorithm is used to verify the speed duration. Combined with the dual threshold verification, the current speed is compared with the historical speed average to obtain the speed limiter action status determination result.

10. An elevator speed governor testing device based on infrared stroboscopic and phase difference speed measurement, used to implement the elevator speed governor testing method based on infrared stroboscopic and phase difference speed measurement as described in any one of claims 1-9, characterized in that, The device includes: The marker point setting module is used to paste highly reflective infrared marker points at equal intervals on the edge of the speed limiter rope wheel to obtain the finished speed limiter rope wheel, and calculate the initial emission frequency of the infrared stroboscope based on the rated speed of the speed limiter. The illumination acquisition module is used to use an infrared stroboscope to irradiate the speed limiter rope wheel with pulsed infrared light according to the initial emission frequency of the infrared stroboscope, and simultaneously use a high-precision camera to acquire dynamic image sequences of the speed limiter rope wheel. The image processing module is used to process the dynamic image sequence of the speed limiter rope wheel to obtain the processed grayscale image sequence. The dynamic weighted template matching algorithm is used to identify the position of the highly reflective infrared marker in the adjacent frame image of the grayscale image sequence and obtain the pixel coordinates of the highly reflective infrared marker. The phase difference calculation module is used to convert the pixel displacement of highly reflective infrared markers into actual physical arc length displacement through a dynamic mapping model, and calculate the phase difference based on the actual physical arc length displacement. The speed calculation module is used to calculate the real-time speed limiter based on the phase difference and convert the real-time speed limiter speed into the real-time linear speed limiter. The status determination module is used to compare the real-time linear velocity of the speed limiter with the preset speed threshold to determine the speed limiter's action status.

Images