Method for detecting skew of coal-taking grab bucket of cement plant

Abstract

The invention discloses a cement plant coal taking grab bucket skew detection method, which relates to the technical field of visual inspection, and comprises the following steps: S1, obtaining material distribution information of a grab bucket operation area; s2, based on the material distribution information, a grab bucket is controlled to align at the highest point of the materials to descend to grab the materials; s3, acquiring a grab bucket working video, and performing distortion correction on each frame of image of the grab bucket video; s4, inputting the corrected grab bucket video into the target detection model to obtain the rotation angle of the grab bucket platform; s5, according to the current working stage of the grab bucket, whether the rotation angle of the grab bucket platform exceeds a preset threshold value of the corresponding stage or not is judged, and if yes, alarm information is pushed. Materials are scanned through the radar level gage, the grab bucket is guided to be aligned to the highest point, charge level interference is eliminated, and pixel-level distortion correction is conducted on the image based on internal parameters of a camera; and angle calculation is not influenced by lens distortion, so that the output skew angle truly reflects the mechanical state of the equipment, the false alarm rate is greatly reduced, and a reliable data basis is provided for accurate diagnosis.

Description

Technical Field

[0001] This invention relates to the field of visual inspection technology, and in particular to a method for detecting the tilt of a coal grab bucket in a cement plant. Background Technology

[0002] In cement plant industrial settings, grab cranes are key equipment for loading and unloading bulk materials such as coal and clinker. During operation, due to mechanical wear and tear, control system malfunctions, and other reasons, the grab platform is prone to tilting, leading to the following problems: Uneven material grabbing: A tilted grab bucket will cause the amount of coal grabbed by different parts of the grab bucket to be inconsistent. Some bucket segments may grab too much coal and some too little coal, which will affect the loading and unloading efficiency. It may also cause the grab bucket to be unbalanced in force, making it easy to collide and grab materials in a loose manner. Equipment damage: Misalignment will cause the mechanical structure of the grab bucket (such as hinge points, connecting rods, cylinders, etc.) to bear additional off-center stress. Over time, this will accelerate the wear, deformation, or even breakage of these components, reduce the service life of the grab bucket, and increase equipment maintenance costs. Impact on subsequent operations: After the grab bucket picks up the tilted coal, there may be incomplete unloading and uneven material accumulation during unloading, which will affect the efficiency of subsequent transportation or storage operations.

[0003] Currently, existing technologies for detecting grab bucket tilt mainly involve: directly extracting the grab bucket's edge or feature points from fixed-image grab bucket data to calculate the tilt angle, or indirectly inferring the tilt state by constructing a three-dimensional motion trajectory model based on a multi-frame image sequence during the grab bucket's descent. However, these methods have the following limitations: (1) Uncorrected camera distortion: Wide-angle or telephoto lenses will produce significant radial and eccentric distortion when shooting the grab at close range. If not corrected, the tilt angle calculated directly from the image will have systematic errors, causing the detection result to deviate from the real physical angle. (2) Ignoring material distribution interference: In actual operation, the surface of the material pile is uneven. When the grab bucket falls, it may cause unreal posture deflection due to local high points or slopes of the material pile. It is impossible to distinguish between "temporary tilt caused by uneven material surface" and "continuous tilt caused by equipment failure", which can easily generate false alarms. Therefore, although existing detection methods can make a rough judgment on skewness, they have low detection accuracy and high false alarm rate, which makes it difficult to meet the needs of industrial sites for accurate perception of equipment status and fault early warning. To this end, a skewness detection method for coal grab buckets in cement plants is proposed. Summary of the Invention

[0004] The purpose of this invention is to solve the problems in the prior art by proposing a method for detecting the tilt of a coal grab bucket in a cement plant.

[0005] A method for detecting the tilt of a coal grab bucket in a cement plant includes the following steps: S1. Obtain material distribution information in the grab bucket operating area; S2. Based on the material distribution information, control the grab bucket to descend and grab the material at the highest point of the material. S3. Acquire the working video of the grab bucket and perform distortion correction on each frame of the grab bucket video; S4. Input the corrected grab video into the target detection model to obtain the rotation angle of the grab platform; S5. Based on the current working stage of the grab bucket, determine whether the rotation angle of the grab bucket platform exceeds the preset threshold of the corresponding stage. If it does, push an alarm message.

[0006] Preferably, in step S1, the material distribution information is obtained by scanning with a radar level gauge, specifically including the average material level value and the highest point extreme value.

[0007] Preferably, in step S3, the distortion correction for each frame of the grab video specifically involves: Calculate the distortion error of each pixel in the image, and correct the pixel coordinates based on the error; any point on the surface The distortion error can be corrected by the following methods:

[0008]

[0009] in Image points after distortion correction coordinates, This represents the geometric error of a pixel caused by distortion. The calculation formula is as follows:

[0010]

[0011] in, The principal point of the camera lens; The effective focal length of the lens; Focal length error refers to the difference between the actual focal length and the theoretical focal length. , and Radial distortion coefficient is a lens optical parameter. and This is the eccentricity distortion coefficient, caused by the misalignment of the lens's optical center during mounting. , for , The scale factor of direction; and Indicates the plane where the camera's image sensor is located relative to... Axial direction and Theoretical tilt angle in the axial direction.

[0012] Preferably, in step S4, the target detection model is the YOLO-OBB model.

[0013] Preferably, in step S4, the rotation angle of the grab platform is the angle between the long or short side of the grab platform and the horizontal direction, and the deflection direction is marked according to the detection results. The specific marking method is as follows: (1) If the angle is the angle between the long side and the horizontal plane, the rotation angle of the grab platform is the angle output by the model, marked as right deviation; (2) If the angle is the angle between the short side and the horizontal plane, the rotation angle of the grab platform is 90°-angle, and it is marked as left deviation.

[0014] Preferably, in step S5, the determination of the current working stage of the grab bucket is based on the relationship between the grab bucket height and the height of the highest point of the material, specifically including:

[0015] in, This refers to the height distance between the radar level gauge and the highest point of the material surface. To lower the grab bucket's steel cable to a lower height, For grab bucket base height data when When the value is greater than or equal to 0, it is determined to be in the descent phase; when When the value is less than 0, it is determined to be in the grabbing phase.

[0016] Preferably, in step S5, based on the current working stage of the grab bucket, it is determined whether the rotation angle of the grab bucket platform exceeds a preset threshold for the corresponding stage. If it does, an alarm message is pushed, specifically: During the descent phase, if the rotation angle of the grab platform is detected to exceed the first threshold... θ If 1 is selected, an alarm message will be sent. During the grabbing phase, if the rotation angle of the grab platform continuously exceeds the second threshold, θ 2. Then an alarm message will be sent. Second threshold θ 2 > First threshold θ 1.

[0017] Preferably, the alarm information pushed during the descent phase indicates a potential malfunction in the wire rope or pulley; the alarm information pushed during the grasping phase indicates a potential malfunction in the opening / closing rope or pulley block.

[0018] Preferably, the radar level gauge is installed on the vertical beam of the grab bucket.

[0019] Compared with existing technologies, the advantages of this invention are: 1. This invention uses a radar level gauge to scan materials, guides the grab bucket to align with the highest point, eliminates material surface interference, and performs pixel-level distortion correction on the image based on camera intrinsic parameters to ensure that the angle calculation is not affected by lens distortion. This allows the output tilt angle to truly reflect the mechanical state of the equipment, significantly reducing the false alarm rate and providing a reliable data foundation for accurate diagnosis.

[0020] 2. This invention divides the grab bucket's operating cycle into two stages: descent and grabbing, and sets differentiated thresholds. During the descent stage, the threshold is strict, and even slight tilting will indicate a potential problem with the lifting mechanism. During the working stage, the threshold is lenient, and an alarm is only triggered if the limit is exceeded continuously. This filters out material interference, and the alarm information is directly associated with potentially faulty components, providing clear guidance for maintenance, reducing unplanned downtime, and lowering operation and maintenance costs. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the process of the present invention.

[0022] Figure 2 This is a diagram showing the result of the YOLO-OBB model processing the output rotation angle of the grab bucket working video in this invention.

[0023] Figure 3 This is a schematic diagram of the grab bucket platform tilted to the right in this invention.

[0024] Figure 4 This is a schematic diagram of the grab bucket platform tilting to the left in this invention. Detailed Implementation

[0025] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0026] Reference Figure 1 As shown, a method for detecting the tilt of a coal grab bucket in a cement plant includes the following steps: S1. Obtain material distribution information in the grab bucket operating area. The material distribution information is obtained by scanning the radar level gauge, specifically including the average material level value and the highest point extreme value. S2. Based on the material distribution information, control the grab bucket to align with the highest point of the material and lower it to grab the material, so as to avoid the grab bucket tilting due to uneven material. S3. Acquire the working video of the grab bucket, perform distortion correction on each frame of the grab bucket video, and use an industrial camera to shoot the working video of the grab bucket directly. In step S3, the distortion correction for each frame of the grab video specifically involves: Calculate the distortion error of each pixel in the image, and correct the pixel coordinates based on the error; any point on the surface The distortion error can be corrected by the following methods:

[0027]

[0028] in Image points after distortion correction coordinates, This represents the geometric error of a pixel caused by distortion. The calculation formula is as follows:

[0029]

[0030] in, The principal point of the camera lens; The effective focal length of the lens; Focal length error refers to the difference between the actual focal length and the theoretical focal length. , and Radial distortion coefficient is a lens optical parameter. and This is the eccentricity distortion coefficient, caused by the misalignment of the lens's optical center during mounting. , for , The scale factor of direction; and Indicates the plane where the camera's image sensor is located relative to... Axial direction and Theoretical tilt angle in the axial direction.

[0031] S4. Input the corrected grab video into the target detection model to obtain the rotation angle of the grab platform; The target detection model is the YOLO-OBB model. The corrected grab video is input into the YOLO-OBB model, and the output results are as follows: Figure 2 As shown.

[0032] In step S4, the rotation angle of the grab platform is the angle between the long or short side of the grab platform and the horizontal direction, and the deflection direction is marked according to the detection results. The specific marking method is as follows: (1) If the angle is the angle between the long side and the horizontal plane, the rotation angle of the grab platform is the angle output by the model, marked as rightward deviation, such as Figure 3 As shown; (2) If the angle is the angle between the shorter side and the horizontal plane, the rotation angle of the grab platform is 90°-angle, and it is marked as leftward deviation, such as Figure 4 As shown.

[0033] S5. Based on the current working stage of the grab bucket, determine whether the rotation angle of the grab bucket platform exceeds the preset threshold of the corresponding stage. If it does, push an alarm message.

[0034] The determination of the current working stage of the grab bucket is based on the relationship between the grab bucket height and the height of the highest point of the material, specifically including:

[0035] The radar level gauge is installed on the vertical beam of the grab bucket. This refers to the height distance between the radar level gauge and the highest point of the material surface (i.e., the vertical distance between the crossbeam and the highest point of the material surface). To lower the grab bucket's steel cable to a lower height, The grab bucket base height is determined by data such as the grab bucket's own dimensions and the dimensions of the grab bucket's moving seat on the crossbeam.

[0036] Specifically, the vertical position is: when the grab bucket cable descends to the following height... Furthermore, when the lowest and highest points of the grab bucket are aligned with the material surface, the distance between the radar level gauge and the highest point of the material surface is... minus .

[0037] when When the value is greater than or equal to 0, it is determined to be in the descent phase; when When the value is less than 0, it is determined to be in the grabbing phase.

[0038] Based on the current working stage of the grab bucket, determine whether the rotation angle of the grab bucket platform exceeds the preset threshold for the corresponding stage. If it does, an alarm message is sent, specifically: During the descent phase, if the rotation angle of the grab platform is detected to exceed the first threshold... θ If 1 (5%), an alarm message will be sent because this stage is the process of the grab bucket descending. At this time, there should not be a large tilt. If there is a tilt, it is considered that there is a potential fault in the lifting and lowering control parts such as the wire rope and pulley of the equipment. During the grabbing phase, if the rotation angle of the grab platform continuously exceeds the second threshold, θ If 2 (15%), an alarm message will be pushed. At this time, it is the grabbing stage. Due to the influence of material stacking, there may be a large angle of deflection. When the angle of deflection in a certain direction exceeds the specified threshold continuously, an alarm will be pushed to reduce false alarms. If the grab is tilted at this stage, it is considered that the grab is not opened evenly, indicating that there may be equipment hazards in the opening and closing rope and pulley group.

[0039] As is known from common technical knowledge, this invention can be implemented through other embodiments that do not depart from its spirit or essential characteristics. Therefore, the disclosed embodiments described above are merely illustrative in all respects and are not the only ones. All modifications within the scope of this invention or its equivalents are included in this invention.

Claims

1. A method for detecting the tilt of a coal grab bucket in a cement plant, characterized in that: Includes the following steps: S1. Obtain material distribution information in the grab bucket operating area; S2. Based on the material distribution information, control the grab bucket to descend and grab the material at the highest point of the material. S3. Acquire the working video of the grab bucket and perform distortion correction on each frame of the grab bucket video; S4. Input the corrected grab video into the target detection model to obtain the rotation angle of the grab platform; S5. Based on the current working stage of the grab bucket, determine whether the rotation angle of the grab bucket platform exceeds the preset threshold of the corresponding stage. If it does, push an alarm message.

2. The method for detecting the tilt of a coal grab bucket in a cement plant according to claim 1, characterized in that: In step S1, the material distribution information is obtained by scanning with a radar level gauge, specifically including the average material level value and the highest point extreme value.

3. The method for detecting the tilt of a coal grab bucket in a cement plant according to claim 1, characterized in that: In step S3, the distortion correction for each frame of the grab video specifically involves: Calculate the distortion error of each pixel in the image, and correct the pixel coordinates based on the error; any point on the surface The distortion error can be corrected by the following methods: in Image points after distortion correction coordinates, This represents the geometric error of a pixel caused by distortion. The calculation formula is as follows: in, The principal point of the camera lens; The effective focal length of the lens; Focal length error refers to the difference between the actual focal length and the theoretical focal length. , and Radial distortion coefficient is a lens optical parameter. and This is the eccentric distortion coefficient, caused by the mounting offset of the lens's optical center; , for , The scale factor of direction; and Indicates the plane where the camera's image sensor is located relative to... Axial direction and Theoretical tilt angle in the axial direction.

4. The method for detecting the tilt of a coal grab bucket in a cement plant according to claim 1, characterized in that: In step S4, the target detection model is the YOLO-OBB model.

5. The method for detecting the tilt of a coal grab bucket in a cement plant according to claim 1, characterized in that: In step S4, the rotation angle of the grab platform is the angle between the long or short side of the grab platform and the horizontal direction, and the deflection direction is marked according to the detection results. The specific marking method is as follows: (1) If the angle is the angle between the long side and the horizontal plane, the rotation angle of the grab platform is the angle output by the model, marked as right deviation; (2) If the angle is the angle between the short side and the horizontal plane, the rotation angle of the grab platform is 90°-angle, and it is marked as left deviation.

6. The method for detecting the tilt of a coal grab bucket in a cement plant according to claim 2, characterized in that: In step S5, the determination of the current working stage of the grab bucket is based on the relationship between the grab bucket height and the height of the highest point of the material, specifically including: in, This refers to the height distance between the radar level gauge and the highest point of the material surface. To lower the grab bucket's steel cable to a lower height, For grab bucket base height data when When the value is greater than or equal to 0, it is determined to be in the descent phase; when When the value is less than 0, it is determined to be in the grabbing phase.

7. The method for detecting the tilt of a coal grab bucket in a cement plant according to claim 6, characterized in that: In step S5, based on the current working stage of the grab bucket, it is determined whether the rotation angle of the grab bucket platform exceeds the preset threshold for the corresponding stage. If it does, an alarm message is pushed out. Specifically: During the descent phase, if the rotation angle of the grab platform is detected to exceed the first threshold... θ If 1 is selected, an alarm message will be sent. During the grabbing phase, if the rotation angle of the grab platform continuously exceeds the second threshold, θ 2. If an alarm message is sent, it will be pushed to the user. Second threshold θ 2 > First threshold θ 1.

8. The method for detecting the tilt of a coal grab bucket in a cement plant according to claim 7, characterized in that: The alarm message pushed during the descent phase indicates a potential malfunction in the wire rope or pulley; the alarm message pushed during the grabbing phase indicates a potential malfunction in the opening / closing rope or pulley block.

9. A method for detecting the tilt of a coal grab bucket in a cement plant according to claim 6, characterized in that: The radar level gauge is installed on the vertical beam of the grab bucket.

Images