Device operation data management method and system for laser obstacle removal

Abstract

The present invention discloses a method and system for managing equipment operation data for laser obstacle removal, belonging to the technical field of laser obstacle removal. The system includes a data acquisition module, an intelligent planning module, an equipment management module, and a visualization module; the data acquisition module is used to collect historical logs, as well as the operation parameters of the laser obstacle removal equipment and the image information of the obstacle removal area; the intelligent planning module analyzes the material types of obstacles in the image information, divides a safety area in the image and sets multiple azimuth points, and calculates the maximum adjustable angle of each azimuth point; the equipment management module calculates the angular velocity threshold of each azimuth point, and real-time detects the debugging angular velocity of the laser obstacle removal equipment. When the debugging angular velocity is greater than the angular velocity threshold corresponding to the azimuth point, it controls the laser obstacle removal equipment to stop running; the visualization module real-time displays the changes in the angular velocity threshold of each azimuth point and the working state of the laser obstacle removal equipment through a visualization screen, generates an obstacle removal record and stores it in the historical log.

Description

Technical Field

[0001] The present invention relates to the technical field of laser obstacle removal, and particularly to a method and system for managing operation data of equipment for laser obstacle removal. Background Art

[0002] Obstacles to power cables, such as tree branches, hanging kites, various mooring objects, etc., pose extremely serious hazards to the safety of power transmission. The handling of such obstacles requires very high professionalism, with great risks during the handling process and without affecting the effective operation of power transmission. How to safely, effectively, and quickly dispose of these obstacles has been an important research topic.

[0003] At present, for the problem of cable obstacle removal, laser obstacle removal equipment is usually manually erected, and high-energy lasers are used for long-distance non-contact obstacle removal. This method has some safety problems. For example, during the remote obstacle removal process, it often relies on manual aiming at the obstacle, and the angle is continuously adjusted to achieve burning or cutting of the obstacle. During this process, the debugging angle often changes suddenly due to unstable equipment installation or accidental human touch, causing the high-energy laser to deviate from the original direction and act on non-obstacles, resulting in losses. Moreover, the farther the laser obstacle removal equipment is from the obstacle, the more sensitive the debugging angle is, and the greater the deviation distance caused by accidental touch, further increasing the risk probability. In the prior art, neither the forced limitation of the debugging angle nor the assistance of image recognition can solve the above problems. The forced limitation of the debugging angle cannot flexibly meet the variable debugging angle requirements in the actual situation, and the image recognition assistance has already caused actual losses after obtaining a clear image. Therefore, at present, a more intelligent technical solution for data monitoring and management of laser obstacle removal equipment is needed to solve the above problems. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for managing operation data of equipment for laser obstacle removal to solve the problems raised in the above background art.

[0005] To solve the above technical problems, the present invention provides the following technical solution: A method for managing operation data of equipment for laser obstacle removal, the method comprising the following steps:

[0006] S100. Collect historical logs and operation parameters of the laser obstacle removal equipment during use, and real-time capture image information of the obstacle removal area through a camera installed on the laser obstacle removal equipment.

[0007] S200. Analyze the image information to identify the obstacle, use an intelligent algorithm to analyze the material type of the obstacle, divide a safety area in the image, set azimuth points at the boundary of the safety area, and calculate the maximum adjustable angle of each azimuth point.

[0008] S300. Use a machine learning algorithm to calculate the expected debugging angular velocity based on historical logs, as well as the angular velocity threshold for each azimuth point. Analyze the debugging angular velocity and debugging direction based on the change in the debugging angle of the laser obstacle removal device. When the debugging angular velocity is greater than the angular velocity threshold corresponding to the azimuth point, control the laser obstacle removal device to stop running.

[0009] S400. Display the change in the angular velocity threshold for each azimuth point and the working status of the laser obstacle removal device through a visualization screen. After drawing the operation route, generate an obstacle removal record and store it in the historical log.

[0010] In S100, the historical log refers to the obstacle removal records when the laser obstacle removal device performs different obstacle removal tasks. Each obstacle removal record includes a detection image, the material type of the obstacle, the operation route, and the operation duration. The detection image includes the position information of the obstacle and the cable. The operation route refers to the cutting path of the high-energy laser on the obstacle. The laser obstacle removal device removes obstacles on the cable through high-energy fiber laser technology. The operation parameters refer to the debugging angle and the obstacle distance. The image information refers to the zoom factor of the camera and the video image of the obstacle removal area.

[0011] The debugging angle refers to the irradiation angle when the laser obstacle removal device performs the obstacle removal task, usually including an azimuth movement angle of 360° and a pitch movement angle of 0° - 45°. By adjusting the debugging angle of the laser obstacle removal device, the high-energy laser can move and burn on the object surface to complete cutting and obstacle removal. The obstacle distance refers to the distance between the laser obstacle removal device and the obstacle in reality.

[0012] In S200, the specific steps are as follows:

[0013] S201. When the laser obstacle removal device starts running, the laser focus is focused on the center point of the video image. Use OpenCV technology to decompose the video of the obstacle removal area into single-frame images. Adopt the edge detection method to screen out the blurred single-frame images and select the one with the latest time as the detection image JC now . Use the YOLO object detection algorithm to perform object recognition on the central area of the detection image JC now and regard the object covering the center point of the image as the obstacle.

[0014] S202. Extract the feature information of the obstacle through the convolutional neural network CNN, and combine it with the gray-level co-occurrence matrix GLCM for texture analysis to obtain the material type type of the obstacle a . Screen out all obstacles in the historical log whose material type is type aThe obstacle removal records are used to mark the operation routes in these records. Each marked operation route is mapped to the detection image corresponding to the obstacle removal record. According to different directions, each marked operation route is split into line segments, and the maximum included angle between each line segment and the cable is calculated and used as the cutting angle, and the farthest distance of each line segment in the direction perpendicular to the cable is used as the cutting distance; the maximum cutting angle QG among all marked operation routes is obtained. max and the farthest cutting distance JL max .

[0015] The types of obstacle materials are defined by the management personnel according to the common types of obstacles in the actual obstacle removal process. When performing texture analysis on the gray-level co-occurrence matrix GLCM, all material types included in the historical log are referred to for similarity comparison analysis, and the material type with the highest similarity is selected to output the analysis result.

[0016] S203. Analyze the cable position in the detection image JC now to obtain the overlapping area of the obstacle on the cable, and use the two farthest positions in the overlapping area as endpoints respectively. First, two straight lines intersecting the cable at an angle of QG max and parallel to each other are established with the two endpoints as the midpoints respectively. Two positions with a distance equal to JL max from the cable in the perpendicular direction are found on each straight line as vertices, and a parallelogram area is divided according to the positions of the four vertices as the first area. Then, two straight lines intersecting the cable at an angle of 180° - QG max and parallel to each other are established with the two endpoints as the midpoints respectively. Two positions with a distance equal to JL max from the cable in the perpendicular direction are found on each straight line as vertices, and a parallelogram area is divided according to the positions of the four vertices as the second area.

[0017] S204. Take the area where the obstacle is located in the detection image JC now as the third area. First, the first area and the second area are merged as the fusion area, and then the overlapping area between the third area and the fusion area is used as the safety area safe now . N azimuth points are evenly set on the boundary line of the safety area safe now , and the image distances between each azimuth point and the center point of the detection image JC now are calculated respectively; according to the ratio of the image distance to the real distance under the current zoom factor, the real distances between each azimuth point and the laser focus are calculated; the obstacle distance JL zaw is obtained. According to the relationship between the debugging angle of the laser obstacle removal device and the changing distance of the laser in reality under the distance JL zaw , the maximum adjustable angle AD corresponding to the real distance of each azimuth point is calculated by substitution. max .

[0018] The angular velocity refers to the speed at which the laser obstacle removal device adjusts its angle around a certain fixed direction when performing the obstacle removal task. The higher the angular velocity, the faster the operator adjusts the angle of the laser obstacle removal device, with a shorter adjustment time or a larger adjustment angle.

[0019] In S300, the specific steps are as follows:

[0020] S301. Screen out the obstacle removal records in the historical log where the material type of all obstacles is type a , count the number of these obstacle removal records U, analyze the area of the safe area in the detection image within each obstacle removal record as the independent variable, and use the operation speed obtained by dividing the operation route length by the operation duration in each obstacle removal record as the dependent variable. Package the independent variable and the dependent variable under each obstacle removal record as a sample, and package these U samples as a training set. Establish a linear regression model R, substitute the training set for training to obtain an expression; obtain the area of the safe area safe now and substitute it into the expression to calculate the expected operation speed, and then according to the distance JL zaw Substitute the relationship between the adjustment angle of the laser obstacle removal device and the changing distance of the laser in reality under, and calculate the expected adjustment angular velocity AV exp .

[0021] The specific training process of the linear regression model R is as follows: Use the independent variable within each sample as the input value of R, calculate the difference between the output result and the dependent variable of the corresponding sample, set an error threshold, and when the difference is less than the error threshold, mark the corresponding sample and calculate the next sample; in other cases, adjust the parameters of the regression coefficient and recalculate until all samples are marked; among them, the expression of the linear regression model is:

[0022] R = L0 + L e d e

[0023] In the formula, L e is the e-th regression coefficient, and d e is the e-th independent variable;

[0024] Before taking the area of the safe area in the detection image within each obstacle removal record as the independent variable, it should be considered that due to the difference in the zoom ratio of the detection images in different obstacle removal records, the ratio of the image distance to the real distance is different. Therefore, it is necessary to first calculate the area of the safe area in the detection image within each obstacle removal record at the same zoom speed and then use it as the independent variable. The same applies to the operation route length.

[0025] The expected operation speed is the speed in the image. It is necessary to first convert it to the speed in reality according to the ratio of the image distance to the real distance, and then according to the distance JL zawThe relationship between the debugging angle of the laser obstacle removal equipment and the changing distance of the laser in reality is replaced by the debugging angular velocity.

[0026] S302, calculating the maximum adjustable angle AD of all azimuth points max The average value of AD ave , and then the maximum adjustable angle AD of each azimuth point max Divide by the expected commissioning angular velocity AV exp Get the trigger duration, and select the trigger duration with the smallest value as the reaction duration TC rea , substitute into the formula to calculate the angular velocity threshold AD of each azimuth point pre ; The formula is as follows:

[0027]

[0028] In the formula, α is a constant greater than 1, and β is a constant.

[0029] The safety zone is re-divided as obstacles change, and the angular velocity threshold of each azimuth point changes in real time according to the position of the laser focus in the safety zone. When the laser focus is closer to the boundary line of the safety zone, the angular velocity threshold in that direction is smaller, and the angular velocity threshold in other directions is larger, so as to limit the laser focus from quickly leaving the safety zone. When the laser focus leaves the safety zone, the laser obstacle removal equipment is controlled to stop emitting high-energy laser beams.

[0030] S303, obtaining the position coordinates (X) of the laser focus in real time through the video image. jd , Y jd ), according to the position coordinate change combined with the operating parameters of the laser obstacle removal equipment, the angular velocity AD is analyzed and debugged now and debug direction FX now . Follow the laser focus in the safe area now The angular velocity threshold AD of each azimuth point is calculated in real time within the position pre , with position coordinates (X jd , Y jd ) is the starting point FX now Create a vector for the direction The position coordinates (X jd , Y jd ) is the starting point and each azimuth point position is the direction to establish a trigger vector. When the laser obstacle removal equipment is debugged, the angular velocity AD now Greater than and vector When the angular velocity threshold of the azimuth point corresponding to the trigger vector with the smallest angle is reached, the laser obstacle removal equipment is controlled to stop emitting the high-energy laser beam.

[0031] In S400, the angular velocity thresholds of each azimuth point and the current debugging angular velocity and debugging direction of the laser obstacle clearing device are displayed in real time through a visualization screen. According to the change of the position coordinates of the laser focus, the operation route is drawn on the detection image JC now After the obstacle clearing is completed, the detection image JC now , the material type of the obstacle, the operation route, and the operation duration are jointly stored as the obstacle clearing record in the historical log.

[0032] A device operation data management system for laser obstacle clearing, which includes a data acquisition module, an intelligent planning module, a device management module, and a visualization module.

[0033] The data acquisition module is used to collect the historical log, as well as the operation parameters of the laser obstacle clearing device and the image information of the obstacle clearing area. The intelligent planning module analyzes the material type of the obstacle in the image information, divides the safe area in the image and sets N azimuth points, and calculates the maximum adjustable angle of each azimuth point. The device management module calculates the angular velocity threshold of each azimuth point and real-time detects the debugging angular velocity of the laser obstacle clearing device. When the debugging angular velocity is greater than the angular velocity threshold corresponding to the azimuth point, it controls the laser obstacle clearing device to stop running. The visualization module displays the change of the angular velocity threshold of each azimuth point and the working state of the laser obstacle clearing device in real time through the visualization screen, generates the obstacle clearing record and stores it in the historical log.

[0034] The data acquisition module includes an operation parameter acquisition unit, a historical log acquisition unit, and an image information acquisition unit.

[0035] The operation parameter acquisition unit is used to collect the obstacle distance and the debugging angle of the laser obstacle clearing device.

[0036] The historical log acquisition unit is used to collect the obstacle clearing records when the laser obstacle clearing device performs different obstacle clearing tasks. Each obstacle clearing record includes the detection image, the material type of the obstacle, the operation route, and the operation duration. The detection image includes the position information of the obstacle and the cable, and the operation route refers to the cutting path of the high-energy laser on the obstacle.

[0037] The image information acquisition unit is used to collect the zoom ratio of the camera and collect the video image of the obstacle clearing area through the camera.

[0038] The intelligent planning module includes an image analysis unit and a region planning unit.

[0039] The image analysis unit is used to obtain the detection image and analyze the historical log.

[0040] First, the video image of the obstacle clearing area is frame-decoded, and the single-frame image with the latest time is selected as the detection image JC now , and an intelligent algorithm is used to identify the obstacle and analyze its material type type a。Then, filter out the records of obstacle removal in the historical log where the material type of the obstacle is type a , split the operation routes in these obstacle removal records into line segments, take the maximum included angle between the line segment and the cable as the cutting angle, and take the farthest distance in the direction perpendicular to the cable of the line segment as the cutting distance. Find the cutting angle QG max with the largest angle and the cutting distance JL max with the farthest distance.

[0041] The area planning unit is used to divide the safety area.

[0042] First, take the two positions farthest apart in the overlapping area between the obstacle and the cable in the detection image JC now as the endpoints respectively.

[0043] Secondly, establish two lines that intersect the cable at an angle of QG max and are parallel to each other with the two endpoints as the midpoints respectively. Find two positions on each line where the distance in the direction perpendicular to the cable is equal to JL max as the vertices, and divide a parallelogram area as the first area according to the positions of the four vertices. Then establish two lines that intersect the cable at an angle of 180° - QG max and are parallel to each other with the two endpoints as the midpoints respectively, and divide a parallelogram area as the second area by analogy.

[0044] Finally, merge the first area and the second area as the fusion area, and then take the overlapping area between the area where the obstacle is located in the detection image JC now and the fusion area as the safety area safe now . Uniformly set N azimuth points on the boundary line of the safety area safe now . Obtain the ratio of the image distance to the real distance at the current zoom ratio, and calculate the real distance according to the image distance between each azimuth point and the laser focus. Obtain the obstacle distance JL zaw , and calculate the maximum adjustable angle AD zaw of each azimuth point according to the relationship between the debugging angle of the laser obstacle removal device and the changing distance of the laser in reality at the distance JL max .

[0045] The device management module includes a parameter calculation unit and a dynamic detection unit.

[0046] The parameter calculation unit is used to calculate the angular velocity threshold of each azimuth point.

[0047] First, filter out all records of obstacles in the historical log where the material type is type aFor the obstacle removal records, the area of the safety zone in the detection image within each obstacle removal record is used as the independent variable, and the operation speed obtained by dividing the operation route length by the operation duration is used as the dependent variable; after substituting into the linear regression model and training, an expression is obtained.

[0048] Secondly, substitute the area of the safety zone safe now into the expression to calculate the predicted operation speed, and then according to the distance JL zaw substitute the relationship between the debugging angle of the laser obstacle removal device and the changing distance of the laser in reality to calculate the predicted debugging angular velocity.

[0049] Finally, calculate the average value AD max of the maximum adjustable angles AD ave of all azimuth points, and then divide the maximum adjustable angle AD max of each azimuth point by the predicted debugging angular velocity to obtain the trigger duration, and select the trigger duration with the smallest value as the reaction duration TC rea , and calculate the angular velocity threshold of each azimuth point according to the formula . Among them, α is a constant greater than 1, and β is a constant.

[0050] The dynamic detection unit is used to detect whether to control the laser obstacle removal device to stop running.

[0051] Analyze the debugging angular velocity AD now and the debugging direction FX now according to the operation parameters of the laser obstacle removal device, and establish a vector now with the position coordinates of the laser focus as the starting point FX Establish trigger vectors respectively with the position coordinates of the laser focus as the starting point and the positions of each azimuth point as the directions. When the debugging angular velocity AD now of the laser obstacle removal device is greater than the angular velocity threshold of the azimuth point corresponding to the trigger vector with the smallest included angle with the vector , control the laser obstacle removal device to stop emitting high-energy laser beams.

[0052] The visualization module displays the angular velocity thresholds of each azimuth point, as well as the debugging angular velocity and debugging direction of the laser obstacle removal device in real time, draws the operation route, and after the obstacle removal is completed, stores the detection image JC now , the material type of the obstacle, the operation route, and the operation duration together as the obstacle removal record in the historical log.

[0053] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0054] 1. Precise safety zone division: By analyzing the operation routes of the same material type in the historical logs, this application obtains the cutting angle and cutting distance, and divides the first zone and the second zone by combining the overlapping areas of obstacles and cables in the current detection image, thereby obtaining the safety zone, which is more precise and efficient compared to the traditional technology that takes the entire obstacle as the safety zone.

[0055] 2. Intelligent angular velocity threshold setting: By setting N azimuth points on the safety zone, this application calculates the maximum adjustable angle in different directions, realizes the limitation of the angular velocity threshold in different directions through the calculation of the expected adjustment angular velocity, and flexibly adjusts the angular velocity threshold in different directions according to the position change of the laser focus, which is more intelligent compared to the fixed angular velocity threshold setting in the traditional technology.

[0056] In summary, compared with the traditional technology, the present invention has the advantages of precise safety zone division and intelligent angular velocity threshold setting, and can improve the data monitoring efficiency of the laser obstacle removal equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention, and do not constitute a limitation to the present invention. In the drawings:

[0058] Figure 1 is a schematic flowchart of the method for managing the operation data of the equipment for laser obstacle removal of the present invention;

[0059] Figure 2 is a schematic structural diagram of the system for managing the operation data of the equipment for laser obstacle removal of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0061] Please refer to Figure 1 , the present invention provides a method for managing the operation data of the equipment for laser obstacle removal, and the method includes the following steps:

[0062] S100. Collect the historical logs and the operation parameters of the laser obstacle removal equipment during use, and capture the image information of the obstacle removal area in real time through the camera installed on the laser obstacle removal equipment.

[0063] S200. Analyze the image information to identify obstacles, use intelligent algorithms to analyze the material types of obstacles, divide the safe area in the image, set azimuth points at the boundaries of the safe area, and calculate the maximum adjustable angle of each azimuth point.

[0064] S300. Use machine learning algorithms to calculate the predicted debugging angular velocity based on historical logs, as well as the angular velocity thresholds for each azimuth point. Analyze the debugging angular velocity and debugging direction according to the change of the debugging angle of the laser obstacle removal device. When the debugging angular velocity is greater than the angular velocity threshold corresponding to the azimuth point, control the laser obstacle removal device to stop running.

[0065] S400. Display the change of the angular velocity threshold of each azimuth point and the working status of the laser obstacle removal device through a visualization screen. After drawing the operation route, generate an obstacle removal record and store it in the historical log.

[0066] In S100, the historical log refers to the obstacle removal records when the laser obstacle removal device performs different obstacle removal tasks. Each obstacle removal record includes the detection image, the material type of the obstacle, the operation route, and the operation duration. The detection image includes the position information of the obstacle and the cable. The operation route refers to the cutting path of the high-energy laser on the obstacle. The laser obstacle removal device removes obstacles on the cable through high-energy fiber laser technology. The operating parameters refer to the debugging angle and the obstacle distance. The image information refers to the zoom ratio of the camera and the video image of the obstacle removal area.

[0067] The debugging angle refers to the irradiation angle of the laser obstacle removal device when performing the obstacle removal task, usually including the azimuth movement angle of 360° and the pitch movement angle of 0° - 45°. By adjusting the debugging angle of the laser obstacle removal device, the high-energy laser can move and burn on the surface of the object to complete cutting and obstacle removal. The obstacle distance refers to the distance between the laser obstacle removal device and the obstacle in reality.

[0068] In S200, the specific steps are as follows:

[0069] S201. When the laser obstacle removal device starts to run, the laser focus is focused on the center point of the video image. Use OpenCV technology to decompose the video of the obstacle removal area into single-frame images, and use the edge detection method to screen out the blurred single-frame images and select the one with the latest time as the detection image JC. now . Use the YOLO object detection algorithm to perform object recognition on the central area of the detection image JC. now Regard the object covering the center point of the image as an obstacle.

[0070] S202. Extract the feature information of the obstacle through the convolutional neural network CNN, and combine the gray-level co-occurrence matrix GLCM for texture analysis to obtain the material type type of the obstacle. a . Screen out all the obstacles in the historical log whose material type is typea The obstacle removal records are used to mark the operation routes in these records. Each marked operation route is mapped to the detection image corresponding to the obstacle removal record. According to different directions, each marked operation route is split into line segments, and the maximum included angle between each line segment and the cable is calculated and used as the cutting angle, and the farthest distance of each line segment in the direction perpendicular to the cable is calculated and used as the cutting distance; the maximum cutting angle QG among all marked operation routes is obtained. max and the farthest cutting distance JL. max .

[0071] The types of obstacle materials are defined by the management personnel according to the common types of obstacles in the actual obstacle removal process. When performing texture analysis on the gray-level co-occurrence matrix GLCM, all material types included in the historical log are referred to for similarity comparison analysis, and the material type with the highest similarity is selected to output the analysis result.

[0072] S203. Analyze the cable position in the detection image JC now to obtain the overlapping area of the obstacle on the cable, and take the two farthest positions in the overlapping area as endpoints respectively. First, two lines intersecting the cable at an angle of QG max and parallel to each other are established with the two endpoints as the midpoints respectively. Two positions with a distance equal to JL max from the cable in the perpendicular direction are found on each line as vertices, and a parallelogram area is divided according to the positions of the four vertices as the first area. Then, two lines intersecting the cable at an angle of 180° - QG max and parallel to each other are established with the two endpoints as the midpoints respectively. Two positions with a distance equal to JL max from the cable in the perpendicular direction are found on each line as vertices, and a parallelogram area is divided according to the positions of the four vertices as the second area.

[0073] S204. Take the area where the obstacle is located in the detection image JC now as the third area. First, the first area and the second area are merged as the fusion area, and then the overlapping area between the third area and the fusion area is used as the safety area safe now . N azimuth points are evenly set on the boundary line of the safety area safe now , and the image distances between each azimuth point and the center point of the detection image JC now are calculated respectively; according to the ratio of the image distance to the real distance under the current zoom factor, the real distances between each azimuth point and the laser focus are calculated; the obstacle distance JL zaw is obtained. According to the relationship between the debugging angle of the laser obstacle removal equipment and the changing distance of the laser in reality under the distance JL zaw , the maximum adjustable angle AD corresponding to the real distance of each azimuth point is calculated by substitution.max 。

[0074] The angular velocity refers to the speed at which the laser obstacle clearing device adjusts its angle around a certain fixed direction when performing the obstacle clearing task. The higher the angular velocity, the faster the operator adjusts the angle of the laser obstacle clearing device during operation, with a shorter adjustment time or a larger adjustment angle.

[0075] In S300, the specific steps are as follows:

[0076] S301. Screen out the obstacle clearing records in the historical log where the material type of all obstacles is type a , count the number of these obstacle clearing records U, analyze the area of the safe zone in the detection image within each obstacle clearing record as the independent variable, and use the operation speed obtained by dividing the operation route length by the operation duration in each obstacle clearing record as the dependent variable. Package the independent variable and the dependent variable under each obstacle clearing record as a sample, and package these U samples as a training set. Establish a linear regression model R, substitute the training set for training to obtain an expression; obtain the area of the safe zone safe now and substitute it into the expression to calculate the predicted operation speed. Then, according to the relationship between the angle adjustment of the laser obstacle clearing device and the changing distance of the laser in reality under the distance JL zaw , substitute and calculate to obtain the predicted adjustment angular velocity AV exp 。

[0077] The specific training process of the linear regression model R is as follows: Use the independent variable within each sample as the input value of R, calculate the difference between the output result and the dependent variable of the corresponding sample, set an error threshold, and when the difference is less than the error threshold, mark the corresponding sample and calculate the next sample; in other cases, adjust the parameters of the regression coefficient and recalculate until all samples are marked; among them, the expression of the linear regression model is:

[0078] R = L0 + L e d e

[0079] In the formula, L e is the e-th regression coefficient, and d e is the e-th independent variable;

[0080] Before using the area of the safe zone in the detection image within each obstacle clearing record as the independent variable, it should be considered that due to the difference in the zoom ratio of the detection images in different obstacle clearing records, the ratio of the image distance to the real distance is different. Therefore, it is necessary to first calculate the area of the safe zone in the detection image within each obstacle clearing record at the same zoom speed and then use it as the independent variable. The same applies to the operation route length.

[0081] The predicted operation speed is the speed in the image. It is necessary to first convert it to the speed in reality according to the ratio of the image distance to the real distance, and then according to the distance JL zawThe relationship between the debugging angle of the laser obstacle removal equipment and the changing distance of the laser in reality is replaced by the debugging angular velocity.

[0082] S302, calculating the maximum adjustable angle AD of all azimuth points max The average value of AD ave , and then the maximum adjustable angle AD of each azimuth point max Divide by the expected commissioning angular velocity AV exp Get the trigger duration, and select the trigger duration with the smallest value as the reaction duration TC rea , substitute into the formula to calculate the angular velocity threshold AD of each azimuth point pre ; The formula is as follows:

[0083]

[0084] In the formula, α is a constant greater than 1, and β is a constant.

[0085] The safety zone is re-divided as obstacles change, and the angular velocity threshold of each azimuth point changes in real time according to the position of the laser focus in the safety zone. When the laser focus is closer to the boundary line of the safety zone, the angular velocity threshold in that direction is smaller, and the angular velocity threshold in other directions is larger, so as to limit the laser focus from quickly leaving the safety zone. When the laser focus leaves the safety zone, the laser obstacle removal equipment is controlled to stop emitting high-energy laser beams.

[0086] S303, obtaining the position coordinates (X) of the laser focus in real time through the video image. jd , Y jd ), according to the position coordinate change combined with the operating parameters of the laser obstacle removal equipment, the angular velocity AD is analyzed and debugged now and debug direction FX now . Follow the laser focus in the safe area now The angular velocity threshold AD of each azimuth point is calculated in real time within the position pre , with position coordinates (X jd , Y jd ) is the starting point FX now Create vector for direction The position coordinates (X jd , Y jd ) is the starting point and each azimuth point position is the direction to establish a trigger vector. When the laser obstacle removal equipment is debugged, the angular velocity AD now Greater than and vector When the angular velocity threshold of the azimuth point corresponding to the trigger vector with the smallest angle is reached, the laser obstacle removal equipment is controlled to stop emitting the high-energy laser beam.

[0087] In S400, the angular velocity thresholds of each azimuth point and the current debugging angular velocity and debugging direction of the laser obstacle clearing device are displayed in real time through the visualization screen. According to the change of the position coordinates of the laser focus, the operation route is drawn on the detection image JC now After the obstacle clearing is completed, the detection image JC now , the material type of the obstacle, the operation route, and the operation duration are jointly stored as the obstacle clearing record in the historical log.

[0088] Please refer to Figure 2 , the present invention provides an equipment operation data management system for laser obstacle clearing. The system includes a data acquisition module, an intelligent planning module, an equipment management module, and a visualization module.

[0089] The data acquisition module is used to collect the historical log, as well as the operation parameters of the laser obstacle clearing device and the image information of the obstacle clearing area. The intelligent planning module analyzes the material type of the obstacle in the image information, divides the safe area in the image and sets N azimuth points, and calculates the maximum adjustable angle of each azimuth point. The equipment management module calculates the angular velocity threshold of each azimuth point and real-time detects the debugging angular velocity of the laser obstacle clearing device. When the debugging angular velocity is greater than the angular velocity threshold corresponding to the azimuth point, it controls the laser obstacle clearing device to stop running. The visualization module displays the change of the angular velocity threshold of each azimuth point and the working state of the laser obstacle clearing device in real time through the visualization screen, generates the obstacle clearing record and stores it in the historical log.

[0090] The data acquisition module includes an operation parameter acquisition unit, a historical log acquisition unit, and an image information acquisition unit.

[0091] The operation parameter acquisition unit is used to collect the obstacle distance and the debugging angle of the laser obstacle clearing device.

[0092] The historical log acquisition unit is used to collect the obstacle clearing records when the laser obstacle clearing device performs different obstacle clearing tasks. Each obstacle clearing record includes the detection image, the material type of the obstacle, the operation route, and the operation duration. The detection image includes the position information of the obstacle and the cable, and the operation route refers to the cutting path of the high-energy laser on the obstacle.

[0093] The image information acquisition unit is used to collect the zoom ratio of the camera and collect the video image of the obstacle clearing area through the camera.

[0094] The intelligent planning module includes an image analysis unit and a region planning unit.

[0095] The image analysis unit is used to obtain the detection image and analyze the historical log.

[0096] First, the video image of the obstacle clearing area is frame-decoded, and the single-frame image with the latest time is selected as the detection image JC now, use an intelligent algorithm to identify obstacles and analyze their material types type a . Then, filter out the obstacle removal records in the historical log where the material type of the obstacle is type a . Split the operation routes in these obstacle removal records into line segments, take the maximum included angle between the line segments and the cable as the cutting angle, and take the farthest distance in the direction perpendicular to the cable of the line segment as the cutting distance. Find the cutting angle QG with the largest angle max and the cutting distance JL with the farthest distance max .

[0097] The area planning unit is used to divide the safety area.

[0098] First, take the two positions farthest apart in the overlapping area between the obstacles and the cable in the detection image JC now as the endpoints respectively.

[0099] Secondly, establish two straight lines intersecting the cable at an angle of QG max and parallel to each other with the two endpoints as the midpoints respectively. Find two positions on each straight line where the distance in the direction perpendicular to the cable is equal to JL max as the vertices. Divide a parallelogram area as the first area according to the positions of the four vertices. Then, establish two straight lines intersecting the cable at an angle of 180° - QG max and parallel to each other with the two endpoints as the midpoints respectively, and divide another parallelogram area as the second area by analogy.

[0100] Finally, merge the first area and the second area as the fusion area, and then take the overlapping area between the area where the obstacle is located in the detection image JC now and the fusion area as the safety area safe now . Uniformly set N azimuth points on the boundary line of the safety area safe now . Obtain the ratio of the image distance to the real distance at the current zoom ratio, and calculate the real distance according to the image distance between each azimuth point and the laser focus. Obtain the obstacle distance JL zaw . Calculate the maximum adjustable angle AD of each azimuth point according to the relationship between the debugging angle of the laser obstacle removal device and the changing distance of the laser in reality at the distance JL zaw . max .

[0101] The device management module includes a parameter calculation unit and a dynamic detection unit.

[0102] The parameter calculation unit is used to calculate the angular velocity threshold of each azimuth point.

[0103] First, filter out all the obstacle removal records in the historical log where the material type of the obstacle is type aFor the obstacle removal records, the area of the safety zone in the detected image in each obstacle removal record is used as the independent variable, and the operation speed obtained by dividing the operation route length by the operation duration is used as the dependent variable; after substituting into the linear regression model and training, an expression is obtained.

[0104] Secondly, substitute the area of the safety zone safe now into the expression to calculate the predicted operation speed, and then according to the distance JL zaw substitute the relationship between the debugging angle of the laser obstacle removal device and the changing distance of the laser in reality to calculate the predicted debugging angular velocity.

[0105] Finally, calculate the average value AD max of the maximum adjustable angles AD ave of all azimuth points, and then divide the maximum adjustable angle AD max of each azimuth point by the predicted debugging angular velocity to obtain the trigger duration, and select the trigger duration with the smallest value as the reaction duration TC rea , and calculate the angular velocity threshold of each azimuth point according to the formula . Among them, α is a constant greater than 1, and β is a constant.

[0106] The dynamic detection unit is used to detect whether to control the laser obstacle removal device to stop running.

[0107] Analyze the debugging angular velocity AD now and the debugging direction FX now according to the operation parameters of the laser obstacle removal device, and establish a vector now with the position coordinates of the laser focus as the starting point FX Establish trigger vectors respectively with the position coordinates of the laser focus as the starting point and the position of each azimuth point as the direction. When the debugging angular velocity AD now of the laser obstacle removal device is greater than the angular velocity threshold of the azimuth point corresponding to the trigger vector with the smallest included angle with the vector , control the laser obstacle removal device to stop emitting high-energy laser beams.

[0108] The visualization module displays the angular velocity thresholds of each azimuth point, as well as the debugging angular velocity and debugging direction of the laser obstacle removal device in real time, draws the operation route, and after the obstacle removal is completed, saves the detected image JC now , the material type of the obstacle, the operation route, and the operation duration together as the obstacle removal record into the historical log.

[0109] Example 1:

[0110] Suppose there are a total of two azimuth points, H1 and H2, on the boundary of the safety zone, and their maximum adjustable angles are 6° and 12° respectively, and the predicted debugging angular velocity is 2° / s; when the constant α is 1.5 and the constant β is 1, substitute into the formula to calculate the angular velocity thresholds of these two azimuth points respectively:

[0111] H1:

[0112] H1:

[0113] Then the angular velocity thresholds of the azimuth points H1 and H2 are 2.86° / s and 7.71° / s respectively.

[0114] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device.

[0115] Finally, it should be noted that the above are only preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for managing equipment operation data for laser obstacle removal, characterized in that: The method comprises the following steps: S100, collecting historical logs and operating parameters of the laser obstacle removal device during use, and taking real-time image information of the obstacle removal area through a camera installed on the laser obstacle removal device; S200, analyzing image information to identify obstacles, using intelligent algorithms to analyze the material types of obstacles and divide safety zones in the image, setting azimuth points at the boundaries of the safety zones and calculating the maximum adjustable angle of each azimuth point; S300, using a machine learning algorithm to calculate the expected debugging angular velocity and the angular velocity threshold of each azimuth point according to the historical log, analyzing the debugging angular velocity and debugging direction according to the change of the debugging angle of the laser obstacle removal device, and controlling the laser obstacle removal device to stop running when the debugging angular velocity is greater than the angular velocity threshold of the corresponding azimuth point; S400, displaying the changes in the angular velocity thresholds of various positions and the working status of the laser obstacle removal equipment on a visual screen, drawing the operation route and generating obstacle removal records to store in the history log; In S100, the historical log refers to the obstacle clearance record when the laser obstacle clearance equipment performs different obstacle clearance tasks. Each obstacle clearance record includes the detection image, the material type of the obstacle, the operation route and the operation duration; the detection image includes the location information of the obstacle and the cable, and the operation route refers to the cutting path of the high-energy laser on the obstacle; the laser obstacle clearance equipment removes obstacles on the cable through high-energy fiber laser technology; the operating parameters refer to the debugging angle and the obstacle distance; the image information refers to the zoom ratio of the camera and the video image of the obstacle clearance area; In S300, the specific steps are as follows: S301. Filter out all obstacles in the history log whose material type is type a Obstacle clearance records are collected, and the number of these obstacle clearance records U is counted. The area of ​​the safety zone in the detection image in each obstacle clearance record is analyzed and used as the independent variable. The operation speed obtained by dividing the operation route length by the operation time in each obstacle clearance record is used as the dependent variable. The independent variables and dependent variables under each obstacle clearance record are packaged as samples, and these U samples are packaged as a training set. A linear regression model R is established, and the expression is obtained after substituting the training set into the training set. The safety zone safety is obtained. now The area is substituted into the expression to calculate the expected operation speed, and then according to the distance JL zaw The relationship between the debugging angle of the laser obstacle removal equipment and the changing distance of the laser in reality is substituted into the calculation to obtain the expected debugging angular velocity AV exp ; S302, calculating the maximum adjustable angle AD of all azimuth points max The average value of AD ave , and then the maximum adjustable angle AD of each azimuth point max Divide by the expected commissioning angular velocity AV exp Get the trigger duration, and select the trigger duration with the smallest value as the reaction duration TC rea , substitute into the formula to calculate the angular velocity threshold AD of each azimuth point pre ; The formula is as follows: In the formula, α is a constant greater than 1, and β is a constant; S303, obtaining the position coordinates (X) of the laser focus in real time through the video image. jd , Y jd ), according to the position coordinate change combined with the operating parameters of the laser obstacle removal equipment, the angular velocity AD is analyzed and debugged now and debug direction FX now ; According to the laser focus in the safety zone now The angular velocity threshold AD of each azimuth point is calculated in real time within the position pre , with position coordinates (X jd , Y jd ) is the starting point FX now Create a vector for the direction The position coordinates (X jd , Y jd ) is the starting point and each azimuth point position is the direction to establish a trigger vector. When the laser obstacle removal equipment is debugged, the angular velocity AD now Greater than and vector When the angular velocity threshold of the azimuth point corresponding to the trigger vector with the smallest angle is reached, the laser obstacle removal equipment is controlled to stop emitting the high-energy laser beam.

2. The method for managing equipment operation data for laser obstacle removal according to claim 1, characterized in that: In S200, the specific steps are as follows: S201. When the laser obstacle removal equipment starts running, the laser focus is focused on the center point of the video image. The OpenCV technology is used to decompose the video of the obstacle removal area into single-frame images. The edge detection method is used to filter out the blurred single-frame images and select the latest one as the detection image JC. now ; Use YOLO target detection algorithm to detect image JC now The target is identified in the central area of ​​the image, and the target covering the center point of the image is regarded as an obstacle; S202, extract the characteristic information of the obstacle through the convolutional neural network CNN, and combine the gray level co-occurrence matrix GLCM to perform texture analysis to obtain the material type of the obstacle a ; Filter all obstacles in the history log by material type a Obstacle clearance records are obtained, and the operation routes in these obstacle clearance records are marked. Each marked operation route is mapped to the detection image of the corresponding obstacle clearance record, and each marked operation route is segmented according to different directions. The maximum angle of intersection between each line segment and the cable is calculated as the cutting angle, and the farthest distance of each line segment in the direction perpendicular to the cable is calculated as the cutting distance; the largest cutting angle QG among all marked operation routes is obtained max And the longest cutting distance JL max ; S203, analyzing and detecting image JC now The cable position in the image is obtained, and the overlapping area of ​​the obstacle on the cable is obtained. The two positions farthest apart in the overlapping area are respectively used as endpoints; first, two intersection angles QG are established with the two endpoints as the midpoints. max On each straight line, find two lines that are perpendicular to the cable and have a distance equal to JL. max As the vertices, a parallelogram area is divided as the first area according to the positions of the four vertices; then two intersection angles of 180°-QG are established with the two end points as the midpoints. max On each straight line, find two lines that are perpendicular to the cable and have a distance equal to JL. max The positions of the four vertices are taken as vertices, and a parallelogram area is divided according to the positions of the four vertices as the second area; S204, the detection image JC now The area where the obstacle is located is taken as the third area. The first area and the second area are first merged as the fusion area, and then the overlapping area between the third area and the fusion area is taken as the safety area. now ; In the safe zone now N points are evenly set on the boundary line, and the JC between each point and the detection image is calculated separately. now Image distance between center points; based on the ratio of image distance to actual distance at the current zoom factor, calculate the actual distance between each azimuth point and the laser focus; Get obstacle distance JL zaw , according to the distance JL zaw The relationship between the debugging angle of the laser obstacle removal equipment and the changing distance of the laser in reality is substituted into the calculation of the maximum adjustable angle AD corresponding to the actual distance of each azimuth point. max .

3. The method for managing equipment operation data for laser obstacle removal according to claim 1, characterized in that: In S400, the angular velocity threshold changes of various positions and the current debugging angular velocity and debugging direction of the laser obstacle removal equipment are displayed in real time on the visual screen, and the detection image JC is displayed according to the position coordinate changes of the laser focus. now Draw the operation route on the top, and after the obstacle is cleared, the image JC will be detected now , the material type of the obstacle, the operation route and the operation duration are stored in the history log as obstacle clearance records.

4. Equipment operation data management system for laser obstacle removal, characterized by: The system includes a data acquisition module, an intelligent planning module, an equipment management module and a visualization module; The data acquisition module is used to collect historical logs, operating parameters of the laser obstacle removal equipment and image information of the obstacle removal area; The intelligent planning module analyzes the material type of obstacles in the image information, divides the safety zone in the image and sets N azimuth points, and calculates the maximum adjustable angle of each azimuth point; The equipment management module calculates the angular velocity threshold of each azimuth point and detects the debugging angular velocity of the laser obstacle removal equipment in real time. When the debugging angular velocity is greater than the angular velocity threshold of the corresponding azimuth point, the laser obstacle removal equipment is controlled to stop running. The visualization module displays the changes in the angular velocity threshold of each azimuth point and the working status of the laser obstacle removal equipment in real time on the visualization screen, generates obstacle removal records and stores them in the historical log. The data acquisition module includes an operation parameter acquisition unit, a history log acquisition unit and an image information acquisition unit; The operating parameter acquisition unit is used to collect the obstacle distance and the debugging angle of the laser obstacle removal equipment; The historical log collection unit is used to collect obstacle clearance records when the laser obstacle clearance equipment performs different obstacle clearance tasks. Each obstacle clearance record includes the detection image, the material type of the obstacle, the operation route and the operation duration; the detection image includes the location information of the obstacle and the cable, and the operation route refers to the cutting path of the high-energy laser on the obstacle; The image information acquisition unit is used to acquire the zoom multiple of the camera and to acquire the video image of the obstacle clearance area through the camera; The equipment management module includes a parameter calculation unit and a dynamic detection unit; The parameter calculation unit is used to calculate the angular velocity threshold of each azimuth point; First, filter out all obstacles in the history log whose material type is type a Obstacle clearance records, the area of ​​the safe zone in the detection image in each obstacle clearance record is taken as the independent variable, and the operation speed obtained by dividing the operation route length by the operation time is taken as the dependent variable; the expression is obtained after substituting it into the linear regression model for training; Secondly, the safe zone now Substitute the area into the expression to calculate the expected operation speed, and then according to the distance JL zaw The relationship between the debugging angle of the laser obstacle removal equipment and the changing distance of the laser in reality is substituted into the calculation to obtain the expected debugging angular velocity; Finally, calculate the maximum adjustable angle AD of all azimuth points max The average value of AD ave , and then the maximum adjustable angle AD of each azimuth point max Divide by the expected debugging angular velocity to get the trigger duration, and select the trigger duration with the smallest value as the reaction time TC rea , according to the formula Calculate the angular velocity threshold of each azimuth point; where α is a constant greater than 1, and β is a constant; The dynamic detection unit is used to detect whether to control the laser obstacle removal equipment to stop running; Analyze and debug angular velocity AD according to the operating parameters of laser obstacle removal equipment now and debug direction FX now FX is the starting point, with the laser focus position coordinates now Create a vector for the direction With the position coordinates of the laser focus as the starting point and each azimuth point as the direction, a trigger vector is established. When the laser obstacle removal equipment is debugged, the angular velocity AD now Greater than and vector When the angular velocity threshold of the azimuth point corresponding to the trigger vector with the smallest angle is reached, the laser obstacle removal equipment is controlled to stop emitting the high-energy laser beam.

5. The equipment operation data management system for laser obstacle removal according to claim 4, characterized in that: The intelligent planning module includes an image analysis unit and a regional planning unit; The image analysis unit is used to obtain the detection image and analyze the historical log; First, decode the video image of the obstacle clearance area and select the latest single frame image as the detection image JC now , using intelligent algorithms to identify obstacles and analyze their material type a ; Then filter out the material type of the obstacle in the historical log as type a Obstacle clearance records are obtained, and the operation routes in these obstacle clearance records are divided into line segments. The maximum angle between the line segment and the cable is taken as the cutting angle, and the farthest distance of the line segment in the direction perpendicular to the cable is taken as the cutting distance. The maximum cutting angle QG is found. max And the longest cutting distance JL max ; Regional planning units are used to divide safety zones; First, the detection image JC now The two positions farthest apart in the overlapping area between the obstacle and the cable are taken as endpoints; Secondly, two lines with an angle of QG intersecting the cable are established with the two endpoints as the midpoints. max On each straight line, find two lines that are perpendicular to the cable and have a distance equal to JL. max As the vertices, a parallelogram area is divided as the first area according to the positions of the four vertices; then two intersection angles of 180°-QG are established with the two end points as the midpoints. max And the straight lines are parallel to each other, and so on, a parallelogram area is divided again as the second area; Finally, the first area and the second area are combined as the fusion area, and then the detection image JC now The overlapping area between the obstacle area and the fusion area is regarded as the safety zone. now ; In the safe zone now N azimuth points are evenly set on the boundary line, the ratio of the image distance to the actual distance under the current zoom factor is obtained, and the actual distance is calculated according to the image distance between each azimuth point and the laser focus; Get obstacle distance JL zaw , according to the distance JL zaw The relationship between the laser obstacle removal equipment debugging angle and the laser's changing distance in reality is used to calculate the maximum adjustable angle AD of each azimuth point. max .

6. The equipment operation data management system for laser obstacle removal according to claim 4, characterized in that: The visualization module displays the angular velocity thresholds of various locations and the debugging angular velocity and debugging direction of the laser obstacle removal equipment in real time, draws the operation route and displays the detection image JC after the obstacle removal is completed. now , the material type of the obstacle, the operation route and the operation duration are stored in the history log as obstacle clearance records.

Images