A device and method for measuring the gap between agitator blades and a mixing vessel based on light and shadow imaging.

By employing optical imaging technology and image processing methods, the problem of online detection of the gap between the stirring blades and the mixing pot in a vertical mixer was solved, achieving full coverage, accurate measurement, and safety early warning, thereby improving detection accuracy and production efficiency.

CN123111110APending Publication Date: 2026-11-17XIAN AEROSPACE CHEM PROPULTION PLANT
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611046581.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-14
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing technologies cannot achieve online dynamic monitoring of the gap between the mixing blades and the mixing pot of a vertical mixer. The detection range is limited, the anti-interference ability is weak, and the measurement accuracy is insufficient, resulting in safety hazards and low production efficiency.

Method used

A device and method for measuring the gap between the stirring blade and the mixing pot based on light and shadow imaging is adopted. A clear shadow outline is formed by parallel light projection, and edge information is extracted by combining image processing technology to construct a two-dimensional closed geometric model. The gap distribution is calculated in real time and an early warning is given, realizing full-process non-contact online detection.

Benefits of technology

It achieves full-area coverage detection of the gap between the stirring blade and the mixing pot, improves measurement accuracy and detection stability, adapts to complex environments, has a graded early warning function, ensures safe operation of equipment, and improves production efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123111110A_ABST
    Figure CN123111110A_ABST
Patent Text Reader

Abstract

This invention discloses a device and method for measuring the gap between agitator blades and a mixing vessel based on light and shadow imaging, relating to the field of online monitoring technology for solid propellant loading. The method first preprocesses the original light and shadow image using Gaussian filtering, background subtraction, and grayscale enhancement. Then, it extracts the edges of the agitator blades and the inner wall of the mixing vessel using the Canny edge detection algorithm, and reconstructs the complete gap contour region using contour extraction and fitting algorithms to obtain basic contour feature information for gap detection. Next, based on the feature correspondence of the complete gap contour, the original gap geometric parameters are calculated in conjunction with imaging geometric calibration parameters. Simultaneously, equipment rotation speed data is fused to solve for the global gap distribution matrix, minimum gap value and spatial location, average gap, and gap uniformity. Abnormal gap areas are marked, and the gap detection results are output.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of online monitoring technology for solid propellant loading, specifically to a device and method for measuring the gap between the stirring impeller and the mixing pot based on light and shadow imaging. Background Technology

[0002] The gap between the mixing blades and the mixing pot of a vertical mixer is a core parameter that directly determines the quality of material mixing, equipment lifespan, and production safety. In propellant mixing production, if the gap is too small, the blade edges are prone to scraping against the pot wall and bottom, generating localized high temperatures and static electricity. For flammable and explosive materials, this can easily induce combustion and explosion accidents. If the gap is too large, large areas of dead zones will form inside the mixing pot, preventing the materials from being fully kneaded, resulting in particle agglomeration, raw material residue, and other problems, significantly reducing the finished product qualification rate and performance stability.

[0003] Currently, the industry uses manual vacuum sealing to measure the gap between the mixing blades and the mixing pot. This method has significant technical shortcomings and cannot meet the needs of online dynamic monitoring: 1) It only supports measuring the gap at fixed measuring points. Sealing mud is applied to fixed positions inside the mixing pot, and obtaining a large range of blade gap values ​​requires applying a large area of ​​sealing mud, which is time-consuming and labor-intensive. 2) Manual operation has large errors. The thickness, tightness, and pressure of the sealing mud are all controlled manually, resulting in limited sampling points and an inability to reflect the overall gap distribution, leading to low measurement repeatability and accuracy. 3) Operational safety is insufficient. The internal space of the mixing pot is confined, and manual sealing operations pose risks of bumps and knocks. 4) Detection efficiency is low. The entire process of sealing, pressing, demolding, and measuring dimensions is time-consuming, and frequent shutdowns severely disrupt continuous production, reducing the overall utilization rate of the equipment. Optical imaging detection technology has advantages such as real-time imaging, resistance to slight obstruction, high boundary recognition, and adaptability to enclosed environments, but it has not yet been applied to the dynamic detection of mixer blade gaps.

[0004] Therefore, developing a gap measurement device based on the principle of light and shadow imaging can effectively make up for the shortcomings of existing detection technologies and has extremely high practical application value. Summary of the Invention

[0005] To address the shortcomings of the aforementioned background technologies, this invention provides a device and method for measuring the gap between the mixing blade and the mixing pot based on light and shadow imaging. This device utilizes parallel light projection to form a clear shadow outline, capturing the boundary morphology of the mixing blade from all angles, enabling simultaneous calculation of the gap between the blade and the pot wall. The entire process is non-contact online detection, requiring no manual intervention. Measurement deviations are corrected by incorporating operating parameters such as equipment vibration and blade speed, further improving dynamic detection accuracy. Coupled with a graded early warning mechanism, abnormal gap conditions are monitored in real time, ensuring the stable and safe operation of the mixing equipment.

[0006] The first objective of this invention is to provide a method for measuring the gap between a stirring impeller and a mixing pot based on light and shadow imaging, used for measuring the gap between the stirring impeller and the side wall of the mixing pot inside a vertical mixer, including: Acquire the original image of the light and shadow inside the mixing pot and the timing data of the blade operation, wherein the timing data of the blade operation includes the real-time rotation angle and real-time rotation speed of the blade. The original image of the light and shadow is subjected to noise reduction, stray light removal, and contrast enhancement to obtain a preprocessed grayscale image; Based on the preprocessed grayscale image, the edge set of the stirring blade and the edge set of the mixing pot inner wall are extracted respectively, and the coupling matching relationship between the blade rotation position and the edge point is established. A two-dimensional closed geometric model of the gap between the blade and the mixing pot is constructed based on the blade edge set and the mixing pot inner wall edge set. Based on the two-dimensional closed geometric model of the gap and combined with the imaging pixel-physical size calibration relationship, the Euclidean vertical physical distance between the blade contour point and the pot wall contour point is calculated point by point, and the real-time circumferential rotation angle and axial height of the blade are used as two-dimensional indexes to construct the global gap distribution matrix. The minimum gap value between the impeller and the mixing pot and the spatial location of the minimum gap are obtained under different rotation positions of the impeller based on the global gap distribution matrix.

[0007] In one embodiment, obtaining the minimum gap value between the impeller and the mixing pot and the spatial location of the minimum gap at different rotation positions of the impeller includes: The global minimum gap value and the corresponding blade and boiler wall contour coordinates are obtained by retrieving the global gap distribution matrix. Combined with the radial differential distribution characteristics of the outer edge surface of the double-wing propeller, a dynamic correlation model between the blade rotation position and the minimum gap point is established to track the minimum gap spatial position that is dynamically offset during the blade rotation process.

[0008] In one embodiment, the construction of a two-dimensional closed geometric model of the paddle-mixing pot gap includes: Clustering and filtering are performed on all edge pixels to collect the continuous edge point set of the paddle and the continuous edge point set of the inner wall of the mixing pot, and isolated noise edges and broken invalid edges are removed. A polynomial fitting algorithm is used to smoothly fit the discrete edge point set, repair edge fractures and gaps, and generate a continuous, smooth and closed complete profile of the blade and the complete profile of the inner wall of the mixing pot. Using the complete outline of the paddle and the complete outline of the inner wall of the mixing pot as the inner and outer boundaries, an annular gap detection area is formed, and a two-dimensional closed geometric model of the gap is constructed.

[0009] In one embodiment, establishing the coupling matching relationship between the blade rotation position and the edge point position includes: Gray-level gradient features are extracted from the preprocessed gray-level image to locate the gray-level abrupt boundary at the outer edge of the blade and the inner wall of the mixing pot. Based on the boundary of gray-scale abrupt change in the outer edge of the positioning blade and the inner wall of the mixing pot, the gradient image is subjected to edge thinning and redundancy removal through non-maximum suppression to remove edge shadows and redundant pixels, thus obtaining single-pixel edge lines. Based on the single-pixel edge lines, a high and low dual threshold filtering rule is used to complete the true and false edge filtering, eliminating the small false edges caused by dust and reflection, and retaining continuous and effective target edges; Based on the continuous and effective target edges, and combined with the characteristics of the double-wing helical blade structure matching the real-time rotation angle of the blades, the complete set of stirring blade edges and the set of mixing pot inner wall edges are separated and output, and a dynamic correspondence between the blade rotation position and the two types of edge points is established.

[0010] In one embodiment, obtaining a preprocessed grayscale image includes: A two-dimensional Gaussian kernel function is used to perform Gaussian filtering on the original image of the light and shadow to smooth and reduce noise. The standard deviation of the Gaussian filter is used to control the degree of image smoothing and noise reduction and suppress random noise interference in the mixed pot. Retrieve the pre-stored bladeless unloaded reference background image, perform background difference operation between the Gaussian filtered image and the unloaded reference background image to eliminate metal reflections on the pot body and stray light interference from the fixed optical path, and retain the effective features of the blade shadow to obtain the difference image. Linear grayscale stretching is performed on the difference image to increase the grayscale difference between the blade shadow and the mixed pot background based on the global minimum and global maximum grayscale values ​​of the difference image, thereby enhancing the shadow contrast.

[0011] In one embodiment, the method further includes: comparing the gap values ​​at all points in the entire area with the upper and lower threshold values ​​of the preset safe gap, automatically identifying abnormal areas where the gap is too small and easy to scratch, or where the gap is too large and there is a dead angle for stirring, marking the abnormal contour coordinates, and providing accurate data support for equipment early warning and maintenance positioning.

[0012] The second objective of this invention is to provide a computer program product, including a computer program that, when executed by a processor, implements a method for measuring the gap between the stirring blade and the mixing pot based on light and shadow imaging.

[0013] A third objective of this invention is to provide an electronic device comprising: Processor; and Memory for storing the executable instructions of the processor; The processor is configured to execute a method for measuring the gap between the stirring blade and the mixing pot based on light and shadow imaging by executing the executable instructions.

[0014] A fourth objective of this invention is to provide a system for measuring the gap between a stirring impeller and a mixing pot based on light and shadow imaging, comprising: The image preprocessing module is used to acquire the original light and shadow image inside the mixing pot and the blade operation timing data, wherein the blade operation timing data includes the real-time rotation angle and real-time rotation speed of the blade; and to perform noise reduction, stray light removal and contrast enhancement processing on the original light and shadow image to obtain a preprocessed grayscale image. The edge detection module is used to extract the edge set of the stirring blade and the edge set of the inner wall of the mixing pot based on the preprocessed grayscale image, and to establish a coupling matching relationship between the blade rotation position and the edge point position. The contour extraction and gap region complete contour reconstruction module is used to construct a two-dimensional closed geometric model of the blade-mixing pot gap based on the blade edge set and the mixing pot inner wall edge set. The global gap feature solving and rotation position associated minimum gap positioning module is used to calculate the Euclidean vertical physical distance between the blade contour points and the pot wall contour points point by point based on the two-dimensional closed geometric model of the gap and the imaging pixel-physical size calibration relationship. It constructs a global gap distribution matrix using the real-time circumferential rotation angle and axial height of the blade as two-dimensional indices. Based on the global gap distribution matrix, it obtains the minimum gap value between the blade and the mixing pot under different blade rotation positions and the minimum gap spatial position.

[0015] The fifth objective of this invention is to provide an apparatus for measuring the gap between a stirring impeller and a mixing vessel based on light and shadow imaging. This apparatus is installed in the intermediate cover of a vertical mixer and is encapsulated for explosion-proof and dust-proof protection of the overall gap. It includes: The directional lighting module is used to provide a stable standard light source for imaging, and includes a parallel light source generator, an angle adjustment bracket, a brightness adjustment drive unit, and a dustproof and light-transmitting lens. The light and shadow imaging acquisition module is used to acquire the original light and shadow image inside the mixing pot and the timing data of the blade operation. The light and shadow imaging acquisition module consists of a high-definition industrial camera, a camera mounting bracket, and an image synchronization acquisition unit; the acquired light and shadow image and the timing data of the blade operation are transmitted to the data processing controller. The data processing controller is used to run the method for measuring the gap between the stirring blade and the mixing pot based on light and shadow imaging, and outputs the detection results of the blade-pot wall gap. The data storage and early warning module is used to record the detection time, corresponding gap value, and operating parameters of each group of data processing controllers in real time, automatically generate gap change curves, and support historical data retrieval and traceability.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention provides a device and method for measuring the gap between a stirring impeller and a mixing pot based on light and shadow imaging. The method first preprocesses the original light and shadow image by performing Gaussian filtering, background subtraction, and grayscale enhancement. Then, it extracts the edges of the stirring impeller and the inner wall of the mixing pot using the Canny edge detection algorithm, and reconstructs the complete gap contour region using contour extraction and fitting algorithms to obtain basic contour feature information for gap detection. Next, based on the feature correspondence of the complete gap contour, the original gap geometric parameters are calculated in conjunction with imaging geometric calibration parameters. Simultaneously, equipment rotation speed data is fused to solve for the global gap distribution matrix, minimum gap value and spatial location, average gap, and gap uniformity. Abnormal gap areas are marked, and the gap detection results are output.

[0017] This invention employs Gaussian filtering preprocessing combined with the Canny dual-target edge detection scheme to accurately separate the blade edge from the pot wall edge. It obtains the complete gap contour through contour fitting and reconstruction technology, and solves the gap geometric parameters based on contour feature matching. This invention is fully adapted to the complex detection environment inside the mixed pot, and has strong anti-interference ability and high detection stability.

[0018] This invention has the capability to quantify and calculate complete gap parameters, and can simultaneously output multi-dimensional parameters such as minimum gap, average gap, and gap uniformity, accurately quantifying the degree of gap distribution between the impeller and the mixing pot, with comprehensive detection dimensions; This invention solves the gap problem by combining the global gap matrix with the minimum gap coordinates. It not only outputs the gap size, but also accurately locates the gap anomaly points, thus overcoming the shortcomings of traditional single-point ranging which lacks spatial positioning capabilities. This invention adds a new dual-wing propeller blade rotation position coupling detection logic, which is adapted to the characteristics of the helical curved surface structure. It can capture the dynamic offset law of the gap position during the entire blade rotation process, establish the correlation feature between the rotation angle and the minimum gap point, avoid the problems of missed detection and false detection caused by fixed point detection, and comprehensively cover the gap state of the stirring dynamic process. Attached Figure Description

[0019] Figure 1 A flowchart of the method for measuring the gap between the stirring blade and the mixing pot based on light and shadow imaging provided by the present invention; Figure 2 This is a schematic diagram of a vertical mixer. Figure 3 A schematic diagram of the area below the hollow blade in the screenshot of the blade-boiler wall clearance measurement. Figure 4 A schematic diagram of the lower part of the hollow blade in the screenshot of the blade-boiler wall clearance measurement; Figure 5 A schematic diagram of the upper part of the hollow blade in the screenshot of the blade-boiler wall clearance measurement; Figure 6 This is a schematic diagram of the area above the hollow blade in the screenshot of the blade-boiler wall clearance measurement. Detailed Implementation

[0020] To enable those skilled in the art to better understand and implement the technical solutions of the present invention, the present invention will be further described below in conjunction with specific embodiments and accompanying drawings. However, the embodiments described are not intended to limit the present invention.

[0021] This invention addresses the problems of existing mixer gap detection methods, such as inability to monitor gaps online, limited detection range, weak anti-interference ability, and insufficient measurement accuracy. It provides a device and method for measuring the gap between the stirring impeller and the mixing pot based on light and shadow imaging. By using parallel light projection to form a clear shadow outline, the device captures the boundary morphology of the stirring impeller from all angles, enabling simultaneous calculation of the gap between the impeller and the pot wall. The entire process is non-contact online detection, requiring no manual intervention. Measurement deviations are corrected by incorporating operating parameters such as equipment vibration and impeller speed, further improving dynamic detection accuracy. Coupled with a graded early warning mechanism, abnormal gap conditions are monitored in real time, ensuring the stable and safe operation of the mixing equipment.

[0022] To implement a method for measuring the gap between agitator blades and a mixing vessel based on light and shadow imaging, this invention provides a light and shadow measuring device for the agitator gap. This device is installed in the intermediate cover of a vertical mixer and features an explosion-proof and dust-proof encapsulation for the entire gap measurement. The measuring device includes: The directional lighting module is used to provide a stable standard light source for imaging. It includes a parallel light source generator, an angle adjustment support, a brightness adjustment drive unit, and a dustproof and light-transmitting lens. The light source adopts a highly collimated narrow-beam parallel light structure, which can project a range of beams vertically downwards. The beam hits the bottom of the mixing pot longitudinally. During the operation of the propeller, the light path is blocked, forming a shadow outline with a clear contrast between light and dark on the bottom surface of the pot, thereby defining the actual boundary position of the propeller.

[0023] The light and shadow imaging acquisition module is used to acquire the original light and shadow image inside the mixing pot and the timing data of the blade operation. The light and shadow imaging acquisition module consists of a high-definition industrial camera, a camera mounting bracket, and an image synchronization acquisition unit; the acquired light and shadow image and the timing data of the blade operation are transmitted to the data processing controller. The light and shadow imaging acquisition module is used to capture dynamic light and shadow images in real time. The camera and the directional supplementary lighting module are installed side by side to constantly detect the position of the light source at the bottom of the mixing pot. The camera is equipped with a high frame rate image sensor, and the shooting frame rate can be synchronously matched with the rotation speed of the stirring blade to ensure that every rotation phase of the blade can be fully imaged. The lens is equipped with anti-fog and anti-material adhesion protective lenses to prevent moisture and splashed materials in the cavity from contaminating the lens and affecting the image quality, and to fully capture the shadow patterns of the sides and bottom edges of the blades and the relative positions between layers.

[0024] The data processing controller is used to run the following method for measuring the gap between the stirring blade and the mixing pot based on light and shadow imaging, and outputs the detection results of the blade-pot wall gap. The data processing controller is the core of the device's operation, incorporating image denoising, contour edge extraction algorithms, spatial coordinate conversion models, and multi-parameter error correction programs. The controller receives raw light and shadow images from the camera, first performing filtering and noise reduction, and removing invalid interference pixels caused by dust and fine materials. It then accurately identifies and delineates the outer contour lines of the stirring blades, the inner wall contour lines of the mixing pot, and the reference contour lines of the pot bottom. Based on the imaging spatial coordinate relationship, it automatically calculates the vertical gap between the blades and the pot wall, combines this with built-in blade attitude data to determine the minimum gap position, and outputs the precise blade-pot wall gap detection results.

[0025] The data storage and early warning module is used to record the detection time, corresponding gap value, and operating parameters of each group of data processing controllers in real time, automatically generate gap change curves, and support historical data retrieval and traceability.

[0026] The data storage and early warning module includes a large-capacity storage unit, a human-machine interface terminal, and an audible and visual early warning alarm. The human-machine interface terminal displays the real-time distribution of gaps across the entire area. Safety gap thresholds for each level are pre-programmed. When the measured gap deviates from the acceptable range, the module triggers a tiered early warning. Minor deviations trigger audible and visual alarms, while severe gap anomalies simultaneously output control signals, triggering the mixing equipment to slow down or stop, thus mitigating the safety risk of equipment scratches.

[0027] This invention, based on a light and shadow measurement device for stirring gaps, can define component positions by relying on shadow boundaries, overcoming the limitations of single-point detection and achieving full coverage detection of gaps throughout the cavity, completely eliminating blind spots in deep cavity structures. Through adaptive adjustment of the light source angle and brightness, it effectively suppresses interference from strong metallic reflections, providing high imaging clarity in low-light, sealed cavities and exhibiting strong adaptability to complex and harsh working conditions. Combined with impeller attitude data, it can determine the location of the minimum gap, thereby obtaining the distribution of the minimum gap between the impeller and the pot wall along the mixing pot axis, significantly improving the gap measurement range under dynamic impeller operation. This invention enables fully non-contact online detection without downtime. It features graded early warning and equipment linkage protection functions, allowing real-time monitoring of gap wear and deformation changes, providing precise data support for stirring process adjustments and equipment maintenance.

[0028] This invention, based on the provided light and shadow measurement device for measuring the gap between the stirring blades and the mixing pot, firstly, the device is powered on and completes a self-test, sequentially checking the light source's emission status, camera imaging function, and sensor signal transmission for normality; the projection angle and brightness of the light source are adjusted, the camera's focal length and imaging coordinate reference are calibrated, and a reference image of the cavity background without blade obstruction is recorded to eliminate initial installation errors. Secondly, the mixer starts no-load stirring operation, and the directional supplementary lighting module continuously and constantly projects a parallel beam of light, which hits the bottom of the mixing pot; the rotating stirring blades continuously block the light path, forming a continuously changing dynamic shadow contour at the bottom of the mixing pot. Meanwhile, the light and shadow imaging acquisition module continuously captures light and shadow images at a synchronized frame rate, and the images are acquired synchronously; the image data is uploaded to the controller in real time. The controller extracts the actual contour boundaries of the stirring blades and the pot wall, locking the minimum blade gap in the actual spatial position within the mixing pot. Thirdly, the edge detection algorithm extracts the blade contour edges, and combined with the blade posture, determines the relationship between the location of the minimum blade-pot wall gap and the position of the mixing pot, outputting the final accurate gap measurement result and the location of the minimum gap. In addition, the measured gap value is compared with the preset safety threshold. When the gap is within the acceptable range, the data is stored and continuously monitored in a loop. If the gap exceeds the standard, an audible and visual warning is immediately activated. If the deviation is serious, the stirring system is linked to perform speed reduction and shutdown operations. After all gap measurements are completed, the device automatically stops the light source projection and image acquisition, summarizes and archives the gap detection data throughout the process, and resets all components to standby mode.

[0029] To achieve the above objectives, see Figure 1 As shown, a method for measuring the gap between a stirring impeller and a mixing pot based on light and shadow imaging is characterized in that, for measuring the gap between the stirring impeller and the side wall of the mixing pot inside a vertical mixer, it includes: S1. Obtain the original image of the light and shadow inside the mixing pot and the blade operation timing data, wherein the blade operation timing data includes the real-time rotation angle and real-time rotation speed of the blade. The original image of the light and shadow is subjected to noise reduction, stray light removal, and contrast enhancement to obtain a preprocessed grayscale image; The preprocessed grayscale image is obtained, including: A two-dimensional Gaussian kernel function is used to perform Gaussian filtering on the original image of the light and shadow to smooth and reduce noise. The standard deviation of the Gaussian filter is used to control the degree of image smoothing and noise reduction and suppress random noise interference in the mixed pot. Retrieve the pre-stored bladeless unloaded reference background image, perform background difference operation between the Gaussian filtered image and the unloaded reference background image to eliminate metal reflections on the pot body and stray light interference from the fixed optical path, and retain the effective features of the blade shadow to obtain the difference image. Linear grayscale stretching is performed on the difference image to increase the grayscale difference between the blade shadow and the mixed pot background based on the global minimum and global maximum grayscale values ​​of the difference image, thereby enhancing the shadow contrast.

[0030] For example, a two-dimensional Gaussian kernel function is used to perform global smoothing on the image, suppressing random noise interference within the mixed pan. The algorithm has high stability and strong real-time performance, making it suitable for industrial online high-speed inspection scenarios. Uniform noise reduction is achieved through weighted averaging of neighboring pixels.

[0031] In the formula, This represents the standard deviation of the Gaussian filter, which is a preset calibration noise reduction coefficient that controls the smoothness of the filter. These represent the horizontal and vertical pixel coordinates in the image pixel coordinate system, respectively. This represents the grayscale value of a pixel at coordinates (x, y) after weighted smoothing and noise reduction using a two-dimensional Gaussian kernel function.

[0032] Call the device's pre-stored bladeless no-load reference background image Differential elimination of stray light such as metallic reflection from the pot body:

[0033] In the formula, This is the grayscale matrix of the difference image after background subtraction processing; This is the grayscale matrix of the real-time acquired image after Gaussian filtering and noise reduction. The grayscale matrix of the pre-stored bladeless unloaded reference background image; This is an absolute value operation used to eliminate interference from fixed optical paths and retain the effective features of blade shadows.

[0034] Linear grayscale stretching is applied to the difference image to enhance the grayscale difference between the blade shadow and the background inside the mixing pot, thus adapting to the environment inside the mixing pot.

[0035] In the formula, These are the global minimum and maximum grayscale values ​​of the difference image; This represents the final output grayscale value of the pixel at coordinates (x, y) after linear grayscale stretching and contrast enhancement.

[0036] S2. Based on the preprocessed grayscale image, extract the set of stirring blade edges and the set of mixing pot inner wall edges respectively, and establish the coupling matching relationship between the blade rotation position and the edge point position; Establishing a coupling matching relationship between the blade rotation position and the edge point position includes: Gray-level gradient features are extracted from the preprocessed gray-level image to locate the gray-level abrupt boundary at the outer edge of the blade and the inner wall of the mixing pot. Based on the boundary of gray-scale abrupt change in the outer edge of the positioning blade and the inner wall of the mixing pot, the gradient image is subjected to edge thinning and redundancy removal through non-maximum suppression to remove edge shadows and redundant pixels, thus obtaining single-pixel edge lines. Based on the single-pixel edge lines, a high and low dual threshold filtering rule is used to complete the true and false edge filtering, eliminating the small false edges caused by dust and reflection, and retaining continuous and effective target edges; Based on the continuous and effective target edges, and combined with the characteristics of the double-wing helical blade structure matching the real-time rotation angle of the blades, the complete set of stirring blade edges and the set of mixing pot inner wall edges are separated and output, and a dynamic correspondence between the blade rotation position and the two types of edge points is established.

[0037] For example, after image preprocessing, a simplified Canny edge detection algorithm is used to extract the edges of the two targets, specifically identifying and separating the edges of the stirring blades and the inner wall of the mixing pot. The algorithm locates the target boundaries based on the image's grayscale abrupt change features, eliminating the need for complex threshold segmentation. It effectively resists interference from dust, stray light, and local reflections within the cavity, stably outputting clear and realistic component edge contours. Simultaneously, by combining the equipment's real-time rotation angle parameters and matching the structural characteristics of the dual-wing helical blades, the algorithm captures the edge correspondence between the blades and the pot wall at different rotation positions, providing reliable basic edge information for subsequent dynamic gap contour reconstruction and minimum gap calculation at variable positions.

[0038] Gradient feature extraction: grayscale gradient calculation is performed on the preprocessed image to capture the locations of abrupt changes in light and shadow, and to initially locate the target edges of the outer edge of the blade and the inner wall of the pot. Edge refinement and redundancy removal: Non-maximum suppression is used to remove edge shadows and redundant pixels, resulting in fine single-pixel edge lines; True and False Edge Filtering: A dual threshold filtering rule with high and low thresholds is used to retain continuous and valid target edges and remove fragmented false edges; Dual-edge output and rotational position coupling matching: Finally, complete sets of stirring blade edges and mixing pot inner wall edges are extracted respectively. Combined with the real-time rotation angle of the double-wing propeller blades, the correspondence between blade rotation position and edge point is established. Since the blade is a double-wing propeller structure, the effective gap area and outer contour projection position of the blade continuously change under different rotation positions. The minimum gap corresponds to the point on the inner wall of the pot, which dynamically shifts with the blade rotation, thus completing the accurate dynamic separation and matching of the two types of target edges.

[0039] S3. Construct a two-dimensional closed geometric model of the gap between the blade and the mixing pot based on the blade edge set and the mixing pot inner wall edge set; The construction of a two-dimensional closed geometric model of the impeller-mixing pot gap includes: Clustering and filtering are performed on all edge pixels to collect the continuous edge point set of the paddle and the continuous edge point set of the inner wall of the mixing pot, and isolated noise edges and broken invalid edges are removed. A polynomial fitting algorithm is used to smoothly fit the discrete edge point set, repair edge fractures and gaps, and generate a continuous, smooth and closed complete profile of the blade and the complete profile of the inner wall of the mixing pot. Using the complete outline of the paddle and the complete outline of the inner wall of the mixing pot as the inner and outer boundaries, an annular gap detection area is formed, and a two-dimensional closed geometric model of the gap is constructed.

[0040] For example, based on the acquired edge information of the blade edge and the pot wall edge, a contour extraction algorithm is used to complete the edge point set fitting and contour closure reconstruction, extract the complete gap boundary contour, construct the gap detection area between the blade and the pot wall, and realize accurate modeling of the gap's full-domain contour.

[0041] Edge point set filtering and clustering: Cluster matching is performed on all detected edge pixels to collect continuous point sets of the blade edge and continuous point sets of the inner wall of the pot, and isolated noise edges and broken invalid edges are removed. Contour curve fitting and reconstruction: A polynomial fitting algorithm is used to smoothly fit the discrete edge point set, repair minor edge breaks and gaps, and generate a continuous, smooth, and closed complete profile of the blade edge and the complete profile of the mixing pot inner wall. Gap region boundary construction: Using the blade profile and the inner wall profile of the mixing pot as double boundaries, a complete annular gap detection profile region is formed, and a two-dimensional closed geometric model of the blade-pot wall gap is established. Contour feature parameter calibration: Extract feature parameters such as boundary curvature, boundary orientation, and contour correspondence of the gap contour to provide complete contour data support for subsequent calculation of gap geometric parameters, minimum gap location, and gap uniformity solution.

[0042] S4. Based on the two-dimensional closed geometric model of the gap and combined with the imaging pixel-physical size calibration relationship, calculate the Euclidean vertical physical distance between the blade contour point and the pot wall contour point point by point, and construct the global gap distribution matrix with the real-time circumferential rotation angle and axial height of the blade as two-dimensional indexes; obtain the minimum gap value and minimum gap spatial position between the blade and the mixing pot under different rotation positions of the blade according to the global gap distribution matrix.

[0043] Obtain the minimum clearance value and spatial location of the minimum clearance between the impeller and the mixing pot at different rotation positions of the impeller, including: The global minimum gap value and the corresponding blade and boiler wall contour coordinates are obtained by retrieving the global gap distribution matrix. Combined with the radial differential distribution characteristics of the outer edge surface of the double-wing propeller, a dynamic correlation model between the blade rotation position and the minimum gap point is established to track the minimum gap spatial position that is dynamically offset during the blade rotation process.

[0044] For example, based on the reconstructed complete gap contours of the blades and the pot wall, and combined with the imaging pixel-physical size calibration relationship, a global traversal matching is performed on the two types of discrete point sets of contours. The Euclidean vertical physical distance between the blade contour points and the pot wall contour points is calculated point by point to obtain the gap values ​​at all positions. A global gap distribution matrix is ​​constructed using the real-time circumferential rotation angle and axial height of the blades as two-dimensional indices. Based on this, the global minimum gap value and its corresponding blade and pot wall contour coordinates are obtained by retrieving the matrix. Combining the structural characteristics of the biplane propeller blade, a dynamic correlation model between the rotation position and the minimum gap point is established. Because the radial distance of the outer edge surface of the propeller blade is distributed differently, the gap position of the blade and the inner wall of the pot continuously switches during the blade rotation, causing the pot wall position corresponding to the minimum gap to dynamically shift with the rotation angle. This can accurately capture the real minimum gap spatial position under different rotation positions.

[0045] The invention also includes: comparing the gap values ​​at all points in the entire area with the preset upper and lower thresholds of the safety gap, automatically identifying abnormal areas where the gap is too small and easy to scratch, or where the gap is too large and there is a dead angle for stirring, marking the abnormal contour coordinates, and providing accurate data support for equipment early warning and maintenance positioning.

[0046] For example, based on the global contour matching results, the key geometric parameters of the gap are quantitatively solved, specifically including: Global single-point gap calculation: Traverse all pairs of matching points between the blade and the boiler wall profile, calculate the actual gap value at each position point by point using the Euclidean distance formula, and construct a complete global gap distribution matrix.

[0047] Average clearance calculation: The overall average clearance between the impeller and the boiler wall is obtained by averaging the clearance values ​​across the entire region. , The number of gap contour samples. This represents the real-time interval value.

[0048] Gap uniformity calculation: The uniformity of gap distribution is characterized by the standard deviation of the gap over the entire region; the smaller the value, the better the uniformity. .

[0049] Excessive Gap Extraction: By comparing the gap values ​​at all points across the entire area with the preset upper and lower thresholds for safe gaps, abnormal areas are automatically identified, such as gaps that are too small and prone to scratching, or gaps that are too large and create dead zones for stirring. The abnormal contour coordinates are marked to provide accurate data support for equipment early warning and maintenance positioning.

[0050] To further illustrate the method for measuring the gap between the stirring blade and the mixing pot based on light and shadow imaging provided by the present invention, specific examples are used for explanation.

[0051] This embodiment describes the detection of the impeller-boiler wall gap in a 100L vertical mixer. (See [link to documentation]). Figure 2 As shown in the schematic diagram of the vertical mixer, the mixer includes a mixing pot 5, the top of which is covered with an intermediate cover. Inside the mixing pot 5, there are vertical solid blades 2 and hollow blades 3. A detection probe 1 is installed at the intermediate cover.

[0052] The stirring gap light and shadow measurement device was installed on a 100L vertical mixer. After powering on and testing, all modules worked normally. The mixing pot was raised under no-load, the mixer speed was set to 24rpm, and the stirring gap light and shadow measurement device was turned on. The following figure records the images of the blade passing under the detection probe at four different times.

[0053] The device detects the output, and the position of the minimum clearance between the impeller and the pot wall appears as shown in [reference]. Figure 3 Below the hollow blade shown, see Figure 4 The hollow blade shown is located in the lower part, see below. Figure 5 The hollow blade shown above, see Figure 6 Above the hollow blades shown, the minimum clearance values ​​are 4.2 mm, 4.1 mm, 4.0 mm, and 4.0 mm, respectively.

[0054] Preset safety gap threshold 4 The minimum clearance is 1mm. An audible and visual warning is immediately triggered when the minimum clearance is below the lower limit or above the upper limit. In this embodiment, the clearance value measured under the unloaded state of the blade did not exceed the safe clearance threshold, therefore the blade-boiler wall clearance is normal.

[0055] The present invention provides a computer program product, including a computer program that, when executed by a processor, implements a method for measuring the gap between the stirring blade and the mixing pot based on light and shadow imaging.

[0056] This invention provides an electronic device, comprising: Processor; and Memory for storing the executable instructions of the processor; The processor is configured to execute a method for measuring the gap between the stirring blade and the mixing pot based on light and shadow imaging by executing the executable instructions.

[0057] This invention provides a system for measuring the gap between a stirring impeller and a mixing pot based on light and shadow imaging, comprising: The image preprocessing module is used to acquire the original light and shadow image inside the mixing pot and the blade operation timing data, wherein the blade operation timing data includes the real-time rotation angle and real-time rotation speed of the blade; and to perform noise reduction, stray light removal and contrast enhancement processing on the original light and shadow image to obtain a preprocessed grayscale image. The edge detection module is used to extract the edge set of the stirring blade and the edge set of the inner wall of the mixing pot based on the preprocessed grayscale image, and to establish a coupling matching relationship between the blade rotation position and the edge point position. The contour extraction and gap region complete contour reconstruction module is used to construct a two-dimensional closed geometric model of the blade-mixing pot gap based on the blade edge set and the mixing pot inner wall edge set. The global gap feature solving and rotation position associated minimum gap positioning module is used to calculate the Euclidean vertical physical distance between the blade contour points and the pot wall contour points point by point based on the two-dimensional closed geometric model of the gap and the imaging pixel-physical size calibration relationship. It constructs a global gap distribution matrix using the real-time circumferential rotation angle and axial height of the blade as two-dimensional indices. Based on the global gap distribution matrix, it obtains the minimum gap value between the blade and the mixing pot under different blade rotation positions and the minimum gap spatial position.

[0058] The above description of the embodiments is only for the purpose of helping to understand the method and core idea of ​​the present invention; at the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the idea of ​​the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for measuring the gap between a stirring impeller and a mixing pot based on light and shadow imaging, characterized in that, Measurement of the gap between the internal stirring blades and the side wall of the mixing pot in a vertical mixer, including: Acquire the original image of the light and shadow inside the mixing pot and the timing data of the blade operation, wherein the timing data of the blade operation includes the real-time rotation angle and real-time rotation speed of the blade. The original image of the light and shadow is subjected to noise reduction, stray light removal, and contrast enhancement to obtain a preprocessed grayscale image; Based on the preprocessed grayscale image, the edge set of the stirring blade and the edge set of the mixing pot inner wall are extracted respectively, and the coupling matching relationship between the blade rotation position and the edge point is established. A two-dimensional closed geometric model of the gap between the blade and the mixing pot is constructed based on the blade edge set and the mixing pot inner wall edge set. Based on the two-dimensional closed geometric model of the gap and combined with the imaging pixel-physical size calibration relationship, the Euclidean vertical physical distance between the blade contour point and the pot wall contour point is calculated point by point, and the real-time circumferential rotation angle and axial height of the blade are used as two-dimensional indexes to construct the global gap distribution matrix. The minimum gap value between the impeller and the mixing pot and the spatial location of the minimum gap are obtained under different rotation positions of the impeller based on the global gap distribution matrix.

2. The method for measuring the gap between the stirring impeller and the mixing pot based on light and shadow imaging according to claim 1, characterized in that, Obtain the minimum clearance value and spatial location of the minimum clearance between the impeller and the mixing pot at different rotation positions of the impeller, including: The global minimum gap value and the corresponding blade and boiler wall contour coordinates are obtained by retrieving the global gap distribution matrix. Combined with the radial differential distribution characteristics of the outer edge surface of the double-wing propeller, a dynamic correlation model between the blade rotation position and the minimum gap point is established to track the minimum gap spatial position that is dynamically offset during the blade rotation process.

3. The method for measuring the gap between the stirring impeller and the mixing pot based on light and shadow imaging according to claim 1, characterized in that, The construction of a two-dimensional closed geometric model of the impeller-mixing pot gap includes: Clustering and filtering are performed on all edge pixels to collect the continuous edge point set of the paddle and the continuous edge point set of the inner wall of the mixing pot, and isolated noise edges and broken invalid edges are removed. A polynomial fitting algorithm is used to smoothly fit the discrete edge point set, repair edge fractures and gaps, and generate a continuous, smooth and closed complete profile of the blade and the complete profile of the inner wall of the mixing pot. Using the complete outline of the paddle and the complete outline of the inner wall of the mixing pot as the inner and outer boundaries, an annular gap detection area is formed, and a two-dimensional closed geometric model of the gap is constructed.

4. The method for measuring the gap between the stirring impeller and the mixing pot based on light and shadow imaging according to claim 1, characterized in that, Establishing a coupling matching relationship between the blade rotation position and the edge point position includes: Gray-level gradient features are extracted from the preprocessed gray-level image to locate the gray-level abrupt boundary at the outer edge of the blade and the inner wall of the mixing pot. Based on the boundary of gray-scale abrupt change in the outer edge of the positioning blade and the inner wall of the mixing pot, the gradient image is subjected to edge thinning and redundancy removal through non-maximum suppression to remove edge shadows and redundant pixels, thus obtaining single-pixel edge lines. Based on the single-pixel edge lines, a high and low dual threshold filtering rule is used to complete the true and false edge filtering, eliminating the small false edges caused by dust and reflection, and retaining continuous and effective target edges; Based on the continuous and effective target edges, and combined with the characteristics of the double-wing helical blade structure matching the real-time rotation angle of the blades, the complete set of stirring blade edges and the set of mixing pot inner wall edges are separated and output, and a dynamic correspondence between the blade rotation position and the two types of edge points is established.

5. The method for measuring the gap between the stirring impeller and the mixing pot based on light and shadow imaging according to claim 1, characterized in that, The preprocessed grayscale image is obtained, including: A two-dimensional Gaussian kernel function is used to perform Gaussian filtering on the original image of the light and shadow to smooth and reduce noise. The standard deviation of the Gaussian filter is used to control the degree of image smoothing and noise reduction and suppress random noise interference in the mixed pot. Retrieve the pre-stored bladeless unloaded reference background image, perform background difference operation between the Gaussian filtered image and the unloaded reference background image to eliminate metal reflections on the pot body and stray light interference from the fixed optical path, and retain the effective features of the blade shadow to obtain the difference image. Linear grayscale stretching is performed on the difference image to increase the grayscale difference between the blade shadow and the mixed pot background based on the global minimum and global maximum grayscale values ​​of the difference image, thereby enhancing the shadow contrast.

6. The method for measuring the gap between the stirring impeller and the mixing pot based on light and shadow imaging according to claim 1, characterized in that, Also includes: By comparing the gap values ​​at all points across the entire area with the preset upper and lower thresholds for safe gaps, abnormal areas are automatically identified, such as gaps that are too small and prone to scratching, or gaps that are too large and create dead zones for stirring. The abnormal contour coordinates are marked, providing accurate data support for equipment early warning and maintenance location.

7. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for measuring the gap between the stirring blade and the mixing pot based on light and shadow imaging as described in any one of claims 1 to 6.

8. An electronic device, characterized in that, include: processor; as well as Memory for storing the executable instructions of the processor; The processor is configured to execute the method for measuring the gap between the stirring blade and the mixing pot based on light and shadow imaging, as described in any one of claims 1 to 6, by executing the executable instructions.

9. A system for measuring the gap between the stirring impeller and the mixing pot based on light and shadow imaging as described in claim 1, characterized in that, include: The image preprocessing module is used to acquire the original image of the light and shadow inside the mixing pot and the timing data of the blade operation. The timing data of the blade operation includes the real-time rotation angle and real-time rotation speed of the blade. The original image of the light and shadow is subjected to noise reduction, stray light removal, and contrast enhancement to obtain a preprocessed grayscale image; The edge detection module is used to extract the edge set of the stirring blade and the edge set of the inner wall of the mixing pot based on the preprocessed grayscale image, and to establish a coupling matching relationship between the blade rotation position and the edge point position. The contour extraction and gap region complete contour reconstruction module is used to construct a two-dimensional closed geometric model of the blade-mixing pot gap based on the blade edge set and the mixing pot inner wall edge set. The global gap feature solving and rotation position associated minimum gap positioning module is used to calculate the Euclidean vertical physical distance between the blade contour points and the pot wall contour points point by point based on the two-dimensional closed geometric model of the gap and the imaging pixel-physical size calibration relationship. It constructs a global gap distribution matrix using the real-time circumferential rotation angle and axial height of the blade as two-dimensional indices. Based on the global gap distribution matrix, it obtains the minimum gap value between the blade and the mixing pot under different blade rotation positions and the minimum gap spatial position.

10. An apparatus for measuring the gap between a stirring impeller and a mixing pot based on light and shadow imaging as described in claim 1, characterized in that, The device is installed in the middle cover of the vertical mixer, and is encapsulated with overall gaps for explosion-proof and dustproof protection, including: The directional lighting module is used to provide a stable standard light source for imaging, and includes a parallel light source generator, an angle adjustment bracket, a brightness adjustment drive unit, and a dustproof and light-transmitting lens. The light and shadow imaging acquisition module is used to acquire the original light and shadow image inside the mixing pot and the timing data of the blade operation. The light and shadow imaging acquisition module consists of a high-definition industrial camera, a camera mounting bracket, and an image synchronization acquisition unit; the acquired light and shadow image and the timing data of the blade operation are transmitted to the data processing controller. A data processing controller is used to run the method for measuring the gap between the stirring blade and the mixing pot based on light and shadow imaging as described in any one of claims 1 to 6, and output the detection result of the blade-pot wall gap. The data storage and early warning module is used to record the detection time, corresponding gap value, and operating parameters of each group of data processing controllers in real time, automatically generate gap change curves, and support historical data retrieval and traceability.