An automatic measuring device for blade surface defect detection and use method

Abstract

An automated measuring device and method for detecting blade surface defects, the device includes a workbench, a collaborative robot arm and a camera mounted on the workbench, a blade material box for containing blades, the blade material box is placed on the workbench; the collaborative robot arm is connected to the robot arm controller; an end effector is mounted on the collaborative robot arm; the camera is used to collect images of the blades to be measured; the camera and the robot arm controller are connected to a client computer, and the client computer is connected to a server via a switch. The blade material box is placed in the corresponding station, the collaborative robot arm and the end effector coordinate and cooperate, and the path trajectory of the collaborative robot arm is planned to achieve orderly grasping of the blades in the blade material box, the camera is placed in a fixed station, the collaborative robot arm clamps the blade to the corresponding inspection station and moves according to the designed posture, completes the image acquisition of the blade surface quality, and then completes the defect detection through an artificial intelligence algorithm.

Description

Technical Field

[0001] The present invention relates to the technical field of compressor blade surface quality detection, and in particular to an automatic measuring device for blade surface defect detection and a use method thereof. Background Art

[0002] Compressor rotor blades are core components of aircraft engines and gas turbines, mainly used to compress air and gas to perform work. The quality of their products plays a vital role in the performance, safety and reliability of the engine. Moreover, compressor blades are consumable parts of the engine. During the long-term service of the engine, they are affected by the engine operating environment, working airflow and foreign objects, and their surfaces are easily damaged. They need to be returned to the factory for regular inspection and maintenance to ensure the safety of the engine.

[0003] At present, the surface defect inspection of engine compressor blades returned to the factory for inspection and repair mainly relies on conventional manual inspection methods such as manual visual inspection and manual touch. The inspection judgment relies on experience and skills, which is labor-intensive and has low inspection efficiency. The inspection process cannot record the quality file information of the defect location, and it is impossible to track the quality of the inspection process. Summary of the invention

[0004] The main purpose of the present invention is to provide an automated measuring device and a method for use for blade surface defect detection, aiming to solve the above-mentioned technical problems.

[0005] To achieve the above-mentioned purpose, on the one hand, the present invention proposes an automated measuring device for blade surface defect detection, comprising a workbench, and a collaborative robotic arm and a camera installed on the workbench; also comprising a blade material box for containing blades, which is placed on the workbench; the collaborative robotic arm is connected to a robotic arm controller; an end effector is installed on the collaborative robotic arm; the camera is used to capture images of the blades to be measured; the camera and the robotic arm controller are both connected to a client computer, and the client computer is connected to a server via a switch.

[0006] Preferably, a teaching pendant is also included, which is control-connected to the collaborative robot arm.

[0007] Preferably, the joints of the collaborative robot arm include a base, a shoulder, an elbow, a first wrist, a second wrist and a third wrist from bottom to top, forming six joints; arm tubes are used to connect the shoulder and the elbow, and the elbow and the first wrist.

[0008] Preferably, a desk is placed on one side of the workbench, the robotic arm controller is arranged on the desk, a display is placed on the desk, and the display is connected to the client computer; the camera is installed on the workbench through a camera mounting bracket.

[0009] Preferably, an auxiliary light source is provided on the workbench, in order to increase the display effect of the blade surface in the photographed picture, adjust the tone of the picture, determine the contrast of the picture, improve the success rate of identifying defects on the blade surface, and avoid problems such as reflection.

[0010] Preferably, a plurality of spaces are provided in the blade material box for vertically inserting the blades to be tested into the spaces; the plurality of spaces are distributed in a rectangular array.

[0011] On the other hand, the present invention also provides a method for using the above-mentioned automated measuring device, comprising the following process:

[0012] Spatial trajectory planning: collect the position information, size information and safety distance of the objects placed on the workbench, and input them into the control system of the collaborative robot arm, and determine the object interference area and each plane information when the collaborative robot arm moves;

[0013] Database construction: Build a different database for each blade to be tested, including the size of the blade material box, the spacing of the internal space of the blade material box, and the surface attribute information of the blade to be tested; establish a lightweight database for the information of the motion trajectory of the collaborative robot arm, and match each blade to be tested with the corresponding trajectory information one by one;

[0014] Blade defect detection: The detection process includes image acquisition of the blade to be inspected, image processing, feature extraction, and algorithm judgment of the acquired images; using the CNN convolutional neural network and RPN network structure, an algorithm model is established for blade defects. Through updating and iteration, a blade real-time monitoring model is constructed to generate detection results of whether there are defects on the blade surface and the specific defect type;

[0015] Result feedback: After comparing with the blade model to be inspected, the status of the blade during the inspection process is fed back in real time, and the error type of defects on the blade surface is prompted.

[0016] Preferably, when judging the interference zone of an object during the movement of a collaborative robot arm, if the collaborative robot arm enters the interference zone during the two-point motion posture, a new posture is automatically added to avoid the interference zone; if the collaborative robot arm enters the interference zone during the two-posture motion on the same plane, a new position point is added so that the collaborative robot arm reaches the target posture in a progressive manner; if the collaborative robot arm enters the interference zone during the two-posture motion on different planes, it first moves to the two-plane overlapping angle position outside the interference zone, and then performs posture judgment on the same plane, and finally reaches the target posture to complete obstacle avoidance.

[0017] Preferably, when building a lightweight database for information on the motion trajectory of the collaborative robot arm, if the collaborative robot arm supports a dragging teaching function, the current position and posture are recorded by dragging, assisted by an automatic obstacle avoidance function.

[0018] Preferably, the specific operation of drag teaching is: the collaborative robot arm starts from the initial posture, moves to the first blade grasping posture, sends a signal to the end effector, the end effector clamps the blade, and the collaborative robot arm moves to the next posture, completes the image data collection of the blade surface quality in the shooting area of ​​the camera, and then puts the blade back to the original empty position of the original blade material box, the end effector releases the blade, and the collaborative robot arm grasps and shoots the next blade until the surface data of all blades on the material tray are obtained, and the collaborative robot arm returns to the set end posture.

[0019] Due to the adoption of the above technical solution, compared with the manual visual method in the prior art, the automatic measuring device has the following beneficial effects:

[0020] (1) By adopting the automated measuring device provided by the present invention, the blade material box is placed in the corresponding workstation, the collaborative robot arm and the end effector coordinate and cooperate, and the path trajectory of the collaborative robot arm is planned to achieve orderly grasping of the blades in the blade material box. The camera is placed in a fixed workstation, and the collaborative robot arm clamps the blade to the corresponding inspection workstation and moves according to the designed posture to complete the image acquisition of the blade surface quality, and then complete the defect detection through the artificial intelligence algorithm.

[0021] (2) The automated measuring device provided by the present invention has a simple structure, a high degree of automated operation, and high measuring efficiency. It has high detection efficiency, reliable detection results, a wide range of applications, and can measure various types of blades, thus having a good application prospect.

[0022] (3) The automated measuring device provided by the present invention can quickly, efficiently and accurately detect the location of surface defects, and record and save them in real time in the system;

[0023] (4) The automated measuring device provided by the present invention has high versatility and can be extended to the detection of surface defects of other parts by intelligently learning the surface defects of other parts.

[0024] (5) The automated measuring device provided by the present invention can realize non-contact detection of micro-defects on the surface of a blade with complex curved surfaces, and can realize three-dimensional detection of micro-defects on the surface of a blade. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0026] Figure 1A three-dimensional diagram of an automated measuring device for blade surface defect detection provided by the present invention;

[0027] Figure 2 A schematic diagram of the hardware connection of the automated measuring device provided by the present invention;

[0028] Figure 3 Schematic diagram of the joints of the collaborative robot arm in the present invention;

[0029] Figure 4 Schematic diagram of the working space of the collaborative robot arm in the present invention Figure 1 ;

[0030] Figure 5 Schematic diagram of the working space of the collaborative robot arm in the present invention Figure 2 ;

[0031] Figure 6 The automatic obstacle avoidance function principle of the collaborative robot arm in the present invention when performing space trajectory planning;

[0032] Figure 7 This is a flowchart of the motion trajectory planning process of the collaborative robot arm in the present invention.

[0033] Explanation of the accompanying drawings: 1. Client computer; 2. Switch; 3. Server; 4. Robotic arm controller; 5. Camera; 6. Blade material box; 7. Collaborative robot arm; 701. Base; 702. Shoulder; 703. Elbow; 704. First wrist; 705. Second wrist; 706. Third wrist; 8. Teach pendant; 9. End effector; 10. Robotic arm main control chassis; 11. Office desk; 12. Workbench; 13. Camera mounting bracket. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0036] In addition, the descriptions of "first", "second", etc. in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0037] Combination Figure 1 , Figure 2 As shown, an automated measuring device for blade surface defect detection is characterized in that it includes a workbench 12, and a collaborative robot arm 7 and a camera 5 installed on the workbench 12; it also includes a blade material box 6 for containing blades, and the blade material box 6 is placed on the workbench 12; the collaborative robot arm 7 is connected to the robot arm controller 4; an end effector 9 is installed on the collaborative robot arm 7; the camera 5 is used to collect images of the blade to be measured; the camera 5 and the robot arm controller 4 are both connected to the client computer 1, and the client computer 1 is connected to the server 3 through the switch 2. Specifically, the robot arm controller 4 is connected to the client computer 1 through a network cable, and the camera 5 is connected to the client computer 1 through a USB interface. Finally, the control of the collaborative robot arm 7 and the camera 5 is realized through the client computer 1. The end effector 9 at the end of the collaborative robot arm 7 sequentially performs three operations on the blade: grasping, surface quality acquisition and identification, and blade return. The corresponding position calibration matching is completed according to the blade material box 6, thereby realizing the automation of the solution.

[0038] In this embodiment, the automated measuring device further includes a teaching pendant 8, which is control-connected to the collaborative robot arm 7. Through the teaching pendant 8 operation interface or drag teaching, the user can control the rotation of each joint of the collaborative robot arm 7 to ensure that the trajectory movement of each angle posture is completed in the motion space.

[0039] Combination Figure 3 As shown, the joints of the collaborative robot arm 7 include a base 701, a shoulder 702, an elbow 703, a first wrist 704, a second wrist 705 and a third wrist 706 from bottom to top, forming six joints; the shoulder 702 and the elbow 703, as well as the elbow 703 and the first wrist 704 are connected by arm tubes.

[0040] Combination Figure 1As shown, a desk 11 is placed on one side of the workbench 12, the robotic arm controller 4 is arranged on the desk 11, a display is placed on the desk 11, and the display is connected to the client computer 1; the camera 5 is installed on the workbench 12 through the camera mounting bracket 13, the robotic arm main control chassis 10 is placed at the bottom of the workbench 12, and the host of the client computer 1 is placed below the desk 11.

[0041] In this embodiment, an auxiliary light source (not shown in the figure) is provided on the workbench 12. The auxiliary light source, such as angled white light or ultraviolet light, aims to increase the display effect of the blade surface in the captured picture, adjust the tone of the picture, determine the contrast of the picture, improve the success rate of identifying defects on the blade surface, and avoid problems such as reflection.

[0042] A plurality of spaces are provided in the blade material box 6 for vertically inserting the blades to be tested into the spaces; the plurality of spaces are distributed in a rectangular array, and a matrix distribution is adopted to facilitate trajectory planning of the collaborative robot arm 7 .

[0043] On the other hand, this embodiment also provides a method for using the above-mentioned automated measuring device, comprising the following process:

[0044] Space trajectory planning: collect the position information, object size information and safety distance of the objects placed on the workbench 12, and input them into the control system of the collaborative robot arm 7, and determine the object interference area and each plane information when the collaborative robot arm 7 moves;

[0045] Database construction: A different database is constructed for each blade to be tested, including the size of the blade material box 6, the internal spacing of the blade material box 6, and the surface attribute information of the blade to be tested; a lightweight database is established for the information of the motion trajectory of the collaborative robot arm 7, and each blade to be tested is matched with the corresponding trajectory information one by one; after completing the trajectory planning, the user only needs to select the corresponding blade model to complete the surface quality inspection of the corresponding type of blade, which is convenient and fast;

[0046] Blade defect detection: The detection process includes image acquisition of the blade to be inspected, image processing, feature extraction, and algorithm judgment of the acquired images; CNN (Convolutional Neural Networks) convolutional neural network and RPN (Region Proposal Network) network structure, establish an algorithm model for blade defects, and continuously improve the detection accuracy and generalization of the algorithm through updates and iterations, build a blade real-time monitoring model, and generate detection results of whether there are defects on the blade surface and the specific defect type;

[0047] Result feedback: After comparing with the blade model to be inspected, the status of the blade during the inspection process is fed back in real time, and the error type of defects on the blade surface is prompted. The system can summarize multiple inspection results, and through data analysis and comparison, it will eventually feed back to the process to complete the closed loop of the inspection process.

[0048] Combination Figures 4 to 6 As shown, when judging the interference zone of the object during the movement of the collaborative robot arm 7, when the collaborative robot arm 7 enters the interference zone during the two-point motion posture, a new posture is automatically added to avoid the interference zone; when the two postures on the same plane enter the interference zone during the motion, a new position point is added to enable the collaborative robot arm 7 to reach the target posture in a progressive manner; when the two postures on different planes enter the interference zone during the motion, first move to the two plane overlapping angle positions outside the interference zone, and then perform posture judgment on the same plane, and finally reach the target posture to complete obstacle avoidance.

[0049] In this embodiment, when a lightweight database is established for the information of the motion trajectory of the collaborative robot arm 7, if the collaborative robot arm 7 supports the dragging teaching function, the current position and posture are recorded by dragging, assisted by the automatic obstacle avoidance function.

[0050] Combination Figure 7 As shown, the collaborative robot arm 7 starts from the initial position and moves to the first blade grabbing position, sending a signal to the end effector 9, the end effector 9 clamps the blade, and the collaborative robot arm 7 moves to the next position, completes the image data collection of the blade surface quality in the shooting area of ​​the camera 5, and then puts the blade back to the original empty position of the original blade material box 6. The end effector 9 releases the blade, and the collaborative robot arm 7 grabs and shoots the next blade until the surface data of all the blades on the material tray are obtained, and the collaborative robot arm 7 returns to the set end position.

[0051] The working principle of the present invention is:

[0052] By using the automated measuring device provided by the present invention, the blade material box 6 is placed in the corresponding workstation, the collaborative robot arm 7 and the end effector 9 coordinate and cooperate, and the path trajectory of the collaborative robot arm 7 is planned to achieve orderly grasping of the blades in the blade material box 6. The camera 5 is placed in a fixed workstation, the collaborative robot arm 6 clamps the blades to the corresponding inspection workstation and moves according to the designed posture to complete the image acquisition of the blade surface quality, and then complete the defect detection through the artificial intelligence algorithm.

[0053] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. All equivalent structural changes made by using the contents of the present invention specification and drawings under the inventive concept of the present invention, or directly / indirectly applied in other related technical fields are included in the patent protection scope of the present invention.

Claims

1. An automated measuring device for blade surface defect detection, characterized in that: It comprises a workbench (12), and a collaborative robot arm (7) and a camera (5) mounted on the workbench (12); it also comprises a blade material box (6) for containing blades, and the blade material box (6) is placed on the workbench (12); The collaborative robot arm (7) is connected to the robot arm controller (4); an end effector (9) is installed on the collaborative robot arm (7); the camera (5) is used to collect images of the blade to be tested; the camera (5) and the robot arm controller (4) are both connected to the client computer (1), and the client computer (1) is connected to the server (3) via a switch (2); The automated measuring device is capable of determining the interference zone of an object when the collaborative robot arm (7) moves. When the collaborative robot arm (7) enters the interference zone during the motion of two points, a new posture is automatically added to avoid the interference zone. When the collaborative robot arm (7) enters the interference zone during the motion of two postures on the same plane, a new position point is added so that the collaborative robot arm (7) reaches the target posture in a progressive manner. When the collaborative robot arm (7) enters the interference zone during the motion of two postures on different planes, the collaborative robot arm first moves to the position of the two planes' overlapping angles outside the interference zone, and then determines the posture of the same plane, and finally reaches the target posture to complete obstacle avoidance.

2. An automated measuring device for blade surface defect detection according to claim 1, characterized in that: It also includes a teaching pendant (8), which is control-connected to the collaborative robot arm (7).

3. The automatic measuring device for blade surface defect detection according to claim 1, characterized in that: The joints of the collaborative robot arm (7) include, from bottom to top, a base (701), a shoulder (702), an elbow (703), a first wrist (704), a second wrist (705) and a third wrist (706), forming six joints; arm tubes are used to connect the shoulder (702) and the elbow (703), and the elbow (703) and the first wrist (704).

4. The automatic measuring device for blade surface defect detection according to claim 1, characterized in that: A desk (11) is placed on one side of the workbench (12), the robotic arm controller (4) is arranged on the desk (11), a display is placed on the desk (11), and the display is connected to the client computer (1); the camera (5) is mounted on the workbench (12) via a camera mounting frame (13).

5. The automatic measuring device for blade surface defect detection according to claim 1, characterized in that: An auxiliary light source is arranged on the workbench (12).

6. The automatic measuring device for blade surface defect detection according to claim 1, characterized in that: A plurality of spaces are arranged in the blade material box (6) for vertically inserting the blades to be tested into the spaces; the plurality of spaces are distributed in a rectangular array.

7. A method for using the automated measuring device according to any one of claims 1 to 6, characterized in that: The process includes the following: Spatial trajectory planning: collecting the position information, object size information and safety distance of the object placed on the workbench (12), and inputting them into the control system of the collaborative robot arm (7), and determining the object interference area and each plane information when the collaborative robot arm (7) moves; Database construction: construct a different database for each blade to be tested, including the size of the blade material box (6), the internal space interval of the blade material box (6), and the surface attribute information of the blade to be tested; establish a lightweight database for the information of the motion trajectory of the collaborative robot arm (7), and perform a one-to-one matching between each blade to be tested and the corresponding trajectory information; Blade defect detection: The detection process includes image acquisition of the blade to be inspected, image processing, feature extraction, and algorithm judgment of the acquired images; using the CNN convolutional neural network and RPN network structure, an algorithm model is established for blade defects. Through updating and iteration, a blade real-time monitoring model is constructed to generate detection results of whether there are defects on the blade surface and the specific defect type; Result feedback: After comparing with the blade model to be inspected, the status of the blade during the inspection process is fed back in real time, and the error type of defects on the blade surface is prompted.

8. The method for using the automated measuring device according to claim 7, characterized in that: When establishing a lightweight database for information on the motion trajectory of the collaborative robot arm (7), if the collaborative robot arm (7) supports a dragging teaching function, the current position and posture are recorded by dragging, assisted by an automatic obstacle avoidance function.

9. The method for using the automated measuring device according to claim 8, characterized in that: The specific operation of the drag teaching is as follows: the collaborative robot arm (7) starts from the initial posture and moves to the first blade grasping posture, sends a signal to the end effector (9), the end effector (9) clamps the blade, and the collaborative robot arm (7) moves to the next posture, completes the image data collection of the blade surface quality in the shooting area of ​​the camera (5), and then puts the blade back to the original empty position of the original blade material box (6), the end effector (9) releases the blade, and the collaborative robot arm (7) grasps and shoots the next blade until the surface data of all the blades on the material tray are obtained, and the collaborative robot arm (7) returns to the set end posture.

Images