CN118181310A - A robot repair method for turbine top cover based on digital twin - Google Patents

Abstract

A digital twinning-based hydraulic turbine top cover robot repairing method comprises the following steps: s1, establishing a three-dimensional model of the processing equipment and the top cover; s2, constructing a top cover repair processing scene model; s3, acquiring relative position data of the processing equipment and planning a processing track; s4, iteratively optimizing a processing area and a processing track of the processing equipment; s5, outputting position data of the processing equipment after iterative optimization relative to the processing area and a processing program; s6, installing and placing the top cover in an actual field; s7, calibrating the actual position of the processing equipment through a robot; s8, repairing and processing the top cover processing area is executed; s9, after repair processing of the processing area at one position is executed, moving processing equipment to the processing area at the next position. By the method, the operation steps and difficulty of on-site repair of the top cover are reduced, the repair time of defects of the overflow surface of the top cover is saved, and the repair efficiency is improved.

Description

A robot repair method for turbine top cover based on digital twin

Technical Field

The present invention relates to the field of robot processing technology, and in particular to a robot repair method for a turbine top cover based on digital twin.

Background technique

After the top cover of the turbine generator has been in service for a period of time, defects such as cavitation points and pits will appear on the flow surface of the top cover. The existing repair method generally uses manual air planing to remove defects, manual arc welding to fill the removed area, and manual grinding of the arc welding repair area to perform plane shaping, and then complete the repair. The manual repair method has the disadvantages of high labor intensity, low efficiency, poor quality consistency, and certain harm to the health of operators. Therefore, turbine maintenance companies have put forward an urgent need for robot on-site repair. However, to realize the on-site robot on-site additive and subtractive material repair processing of the turbine top cover, high requirements are placed on the design and production of equipment, equipment installation location, commissioning and testing, and process research. In recent years, some equipment manufacturers have developed miniaturized repair equipment for the additive repair link in the process of additive and subtractive material processing and repair of the flow surface of the turbine top cover, but have not optimized the implementation method and repair process of the equipment on site. Operators are still required to perform complex operations on site, which adds new difficulties to the repair process.

Summary of the invention

The technical problem to be solved by the present invention is: to solve the problems existing in the above-mentioned background technology, and to provide a turbine top cover robot repair method based on digital twin. By building a top cover repair processing scene model, a processing placement plan of the processing equipment relative to the top cover flow surface is formed in the model, and the processing area and processing trajectory of the processing equipment are further iterated and optimized, so as to reduce the on-site operation steps and difficulty, and save the time for repairing defects on the top cover flow surface.

In order to solve the above technical problems, the technical solution adopted by the present invention is: a hydraulic turbine top cover robot repair method based on digital twin, comprising the following steps:

S1. Establish 3D model of processing equipment and top cover;

S2. Build a top cover repair processing scenario model, and form a processing placement plan for the processing equipment relative to the top cover flow surface by simulating the placement of the processing equipment;

S3. Obtaining the relative position data of the processing equipment in the top cover repair processing scene model, and planning the processing trajectory of the robot of the processing equipment;

S4. Iteratively optimize the processing area and processing trajectory of the processing equipment by adjusting the installation position of the robot end processing tool and the placement position of the processing equipment relative to the top cover;

S5. Plan the trajectory and processing technology of the processing equipment, and output the position data of the processing equipment relative to the processing area and the processing program after iterative optimization, which are used to guide the placement and processing of the processing equipment in the actual processing process;

S6. Based on the constructed roof repair and processing scene model, install and place the roof on the actual site according to the design of the model;

S7. Based on the output processing equipment processing placement plan, the actual position of the processing equipment is calibrated by the robot, and compared with the top cover repair processing scene model to calibrate the placement position of the processing equipment;

S8. After the processing equipment completes the positioning calibration, the processing program is loaded into the robot's teaching pendant to perform the repair processing of the top cover processing area;

S9. After the processing equipment completes the repair processing of one processing area, the processing equipment is moved to the next processing area, and the top cover repair processing of the next processing area is started.

In S1, a three-dimensional point cloud acquisition device is used to collect point clouds of the flow surface area of the top cover, and point cloud processing technology is used to realize point cloud processing of the flow surface area of the top cover, and point cloud noise points are filtered out. The model of the flow surface of the top cover is reconstructed using reverse modeling technology. Combined with the original drawing data of the top cover, a three-dimensional model of the top cover is established in the three-dimensional software, and then modeling is performed in the three-dimensional software according to the size data of the processing equipment to obtain a three-dimensional model of the processing equipment.

In S2, a top cover repair processing scene model is established in the three-dimensional software, and the top cover repair processing scene model and the three-dimensional model of the processing equipment are imported into the three-dimensional simulation software and the offline programming software to simulate the top cover flow surface processing scene, and the placement position of the processing equipment is moved in the constructed top cover repair processing scene model to form a processing placement plan for the processing equipment relative to the top cover flow surface.

In S3, the top cover repair processing scene model laid out in the three-dimensional simulation software is used to obtain the position data of the processing equipment relative to the processing area of the top cover flow surface, and the position data of the processing equipment relative to the processing area is modified in the offline programming software. The trajectory planning function and the end processing tool adjustment function of the offline programming software are used to select the processing area for processing trajectory planning.

In S4, the processing area and the processing trajectory are iteratively optimized by adjusting the installation position of the robot's end processing tool and the placement position of the processing equipment relative to the top cover in the offline programming software; the iterative optimization of the processing trajectory adopts a method of optimizing the installation position data of the processing equipment relative to the processing area, the installation position data of the robot's end processing tool, and the initial joint posture of the robot processing, so as to expand the processing area of the flow surface of the top cover for trajectory planning, and finally obtain multiple processing areas that can cover the entire flow surface of the top cover.

In S5, the end processing tool installation position data of the robot is output to guide the design of the additive and subtractive tool fixture, the safe position point of the robot in the top cover flow surface processing area is output to protect the safety of the robot's initial position, the initial joint angle data of the robot's processing trajectory is output to guide the robot's safe posture during processing, and the processing equipment movement positioning data is output to guide the processing equipment to move to the next processing area for repeated processing.

In S6, several piers are arranged on the actual site according to the design of the model, and the top cover is placed on the piers to restore the top cover repair processing scene in the model.

In S7, after the processing equipment is placed under the top cover, the relative position data of the processing equipment relative to the processing area output by the simulation software is used as a reference, and the actual position of the processing equipment is calibrated by a robot, and compared with the simulation data, and the position of the processing equipment is calibrated and moved; wherein, the positioning method of the processing equipment is to use the visual measurement equipment installed at the end of the robot to visually locate the reference point set on the flow surface of the top cover, calculate the relative position of the reference point relative to the base coordinate of the processing equipment, and combine the relative position data of the reference point relative to the base coordinate of the processing equipment in the simulation software to perform data calibration and comparison, adjust the processing position of the actual processing equipment, and realize the positioning of the end processing tool.

In S8, the robot processing program planned by offline programming is copied to the robot controller via a USB flash drive or a data cable. The processing program is called in the controller to perform an empty walk in the top cover processing area to verify the feasibility of the program. The processing program is executed again in conjunction with the end processing tool control logic to realize the robot top cover additive and subtractive processing and repair.

In S9, after the robot completes the defect repair processing in one processing area, the processing equipment is manually pushed to the next processing area, and the visual measurement equipment at the end of the robot is used to visually locate the reference point set on the flow surface of the top cover, adjust the position of the processing equipment, and compare it with the top cover repair processing scene model to calibrate the placement of the processing equipment and the positioning of the end processing tools.

The present invention has the following beneficial effects:

1. The present invention builds a virtual top cover repair processing scene model, forms a processing placement plan for processing equipment relative to the top cover flow surface in the model, and further iterates and optimizes the processing area and processing trajectory of the processing equipment, so that during the on-site top cover repair processing, the on-site operation steps and difficulty are reduced, thereby saving repair time and improving repair efficiency.

2. The present invention builds a virtual top cover repair processing scene model and a robot processing equipment model, uses offline programming software and virtual simulation software to simulate the processing process, iteratively optimizes the robot end tool installation position and the robot base installation position, reduces the equipment manufacturing iterative optimization in the equipment manufacturing process, thereby reducing the equipment manufacturing time and equipment manufacturing cost, and achieving the purpose of reducing costs and increasing efficiency in equipment manufacturing and equipment debugging.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

FIG1 is a schematic diagram of a top cover repair processing scene model constructed according to the present invention.

FIG. 2 is a schematic diagram showing the state of the processing equipment of the present invention when repairing the flow surface of the top cover.

FIG. 3 is a schematic structural diagram of the processing equipment of the present invention.

In the figure, there are processing equipment 10, a robot 11, an end processing tool 12, a base 13, a transverse platform 14, a top cover 20, a top cover flow surface 21, and a buttress 30.

Detailed ways

Referring to Figures 1-3, a digital twin-based turbine top cover robot repair method includes the following steps:

S1. Establishing a three-dimensional model of the processing equipment 10 and the top cover 20;

S2. Build a top cover repair processing scene model, by simulating the placement of the processing equipment 10, forming a processing placement scheme for the processing equipment 10 relative to the top cover flow surface 21;

S3 obtains the relative position data of the processing equipment 10 in the top cover repair processing scene model, and the processing equipment 10 of the robot 11 performs processing trajectory planning;

S4. By adjusting the installation position of the end processing tool 12 of the robot 11 and the placement position of the processing equipment 10 relative to the top cover 20, iteratively optimizing the processing area and processing trajectory of the processing equipment 10;

S5. The trajectory and processing technology of the processing equipment 10 are planned, and the position data of the processing equipment 10 relative to the processing area and the processing program are output after iterative optimization to guide the actual processing process of the processing equipment 10 to place and process;

S6. According to the built top cover repair processing scene model, in the actual site according to the design of the model on the site to install the top cover 20 placed;

S7. Based on the output processing equipment 10 processing placement program, the actual position of the processing equipment 10 is calibrated by the robot 11, and compared with the top cover repair processing scene model, calibrating the placement position of the processing equipment 10;

S8. After the positioning and calibration of the processing equipment 10 is completed, the processing program is loaded in the teaching pendant of the robot 11, and the repair processing of the top cover 20 processing area is performed;

S9. After the processing equipment 10 completes the repair processing of one processing area, the processing equipment 10 is moved to the next processing area, and the top cover repair processing is started for the next processing area.

By building a top cover repair processing scene model, a processing placement plan for the processing equipment 10 relative to the top cover flow surface 21 is formed in the model, and the processing area and processing trajectory of the processing equipment 10 are further iteratively optimized to reduce the on-site operation steps and difficulty, and save time for repairing top cover flow surface defects.

In S1, a three-dimensional point cloud acquisition device is used to collect point clouds of the top cover flow surface 21 area, and point cloud processing technology is used to realize point cloud processing of the top cover flow surface 21 area, and point cloud noise points are filtered out. The model of the top cover flow surface 21 is reconstructed using reverse modeling technology, and then combined with the original drawing data of the top cover 20, a three-dimensional model of the top cover 20 is established in the three-dimensional software, and then modeling is performed in the three-dimensional software according to the size data of the processing equipment 10 to obtain the three-dimensional model of the processing equipment 10.

Since there are defects such as cavitation points and pits on the top cover flow surface 21, point cloud acquisition is performed on the top cover flow surface 21 area through a three-dimensional point cloud acquisition device to obtain the actual defect characteristics of the top cover flow surface 21. The rest of the top cover 20 does not need to be repaired, so the rest of the top cover 20 is combined with the original drawing data of the top cover 20 to establish a three-dimensional model of the entire top cover 20.

In S2, a top cover repair processing scene model is established in the three-dimensional software, and the top cover repair processing scene model and the three-dimensional model of the processing equipment 10 are imported into the three-dimensional simulation software VisualComponents and the offline programming software SprutCAM to simulate the top cover flow surface processing scene, and the placement position of the processing equipment 10 is moved in the constructed top cover repair processing scene model to form a processing placement plan for the processing equipment 10 relative to the top cover flow surface 21.

The top cover 20 is a large ring structure, so that the top cover flow surface 21 is also a large ring. The processing equipment 10 cannot repair the entire top cover flow surface 21 at one time. Therefore, the top cover flow surface 21 needs to be divided into several processing areas. These processing areas cover the entire top cover flow surface 21. Therefore, it is necessary to determine the placement position of the processing equipment 10. The placement of the top cover flow surface processing area and the processing equipment 10 in the virtual simulation software Visual Components is a feasible placement area position planning based on the actual processing area and the size of the processing equipment 10.

In S3, the top cover repair processing scene model laid out in the three-dimensional simulation software Visual Components is used to obtain the position data of the processing equipment 10 relative to the processing area of the top cover flow surface 21, and the position data of the processing equipment 10 relative to the processing area is modified in the offline programming software SprutCAM. The trajectory planning function and end tool adjustment function of the offline programming software SprutCAM are used to select the processing area for processing trajectory planning.

The installation position data of the robot designed in the virtual simulation software Visual Components relative to the top cover flow surface processing area is used to guide the robot installation position in the offline programming software SprutCAM, saving the time for adjusting the robot posture in the offline programming software SprutCAM.

In S4, the processing area and the processing trajectory are iteratively optimized by adjusting the installation position of the end processing tool 12 of the robot 11 and the placement position of the processing equipment 10 relative to the top cover 20 in the offline programming software SprutCAM; wherein, the iterative optimization of the processing trajectory adopts a method of optimizing the installation position data of the processing equipment 10 relative to the processing area, the installation position data of the end processing tool 12 of the robot 11, and the initial joint posture of the robot 11 processing, so as to expand the processing area of the top cover flow surface 21 for trajectory planning, and finally obtain multiple processing areas that can cover the entire top cover flow surface 21.

Referring to Figs. 2 and 3, the processing equipment 10 includes a base 13, on which a lateral platform 14 is mounted, and the lateral platform 14 is moved by a servo motor driving a ball screw slide, and the robot 11 is mounted on the lateral platform 14, so that the robot 11 can move a distance left and right, thereby expanding the processing area. After expanding the processing area of the processing equipment 10 relative to the flow surface 21 of the top cover, the number of processing areas can be further reduced, and the number of movements of the processing equipment 10 can be reduced, thereby saving repair time and improving repair efficiency.

In S5, the installation position data of the end processing tool 12 of the robot 11 is output to guide the design of the additive and subtractive tool fixture, the safe position point of the robot 11 in the processing area of the top cover flow surface 21 is output to protect the safety of the robot's initial position, the initial joint angle data of the robot 11 processing trajectory is output to guide the robot's safe posture during processing, and the moving positioning data of the processing equipment 10 is output to guide the processing equipment 10 to move to the next processing area for repeated processing.

The output robot offline simulation result data, after final verification of feasibility, submits the iteratively optimized processing equipment 10 installation position data, the robot 11 end processing tool 12 installation position data, the robot safety point data, the robot processing initial joint data, and the reference data of the robot moving to the next processing area to form the operation instructions for the implementation plan of the robot in-situ additive and subtractive material processing of the flow surface of the turbine top cover, which is used to guide the placement and processing of the processing equipment 10 in the actual processing process.

In S6 , referring to FIG. 1 , a plurality of buttresses 30 are arranged on the site in accordance with the design of the model in the actual site, and the top cover 20 is placed on the buttresses 30 to restore the top cover repair processing scene in the model.

As shown in Figure 1, it is both a constructed roof repair processing scene model and a scene of repair in the actual site. The processing equipment 10 enters the site using the layout data of the roof repair processing scene model constructed by simulation, and relies on manual execution of equipment transfer and equipment layout installation, where the equipment installation position is determined by the simulation results.

In S7, referring to Figs. 1 and 2, after placing the processing equipment 10 under the top cover 20, the actual position of the processing equipment 10 is calibrated by the robot 11 with reference to the relative position data of the processing equipment 10 relative to the processing area output by the simulation software, and compared with the simulation data, and the position of the processing equipment 10 is calibrated and moved. Among them, the positioning method of the processing equipment 10 is to use the visual measurement equipment installed at the end of the robot 11 to visually locate the reference point set on the flow surface 21 of the top cover, calculate the relative position of the reference point relative to the base coordinate of the processing equipment 10, and combine the relative position data of the reference point relative to the base coordinate of the processing equipment 10 in the simulation software to perform data calibration and comparison, adjust the processing position of the actual processing equipment 10, and realize the positioning of the end processing tool 12.

In S8, the robot processing program planned by offline programming is copied to the controller of the robot 11 via a USB flash drive or a data cable. The processing program is called in the controller to perform an empty walk in the top cover processing area to verify the feasibility of the program. The processing program is executed again in conjunction with the control logic of the end processing tool 12 to realize the robot top cover additive and subtractive processing and repair.

In S9, after the robot 11 completes the defect repair processing in one processing area, the processing equipment 10 is manually pushed to the next processing area, and then the visual measurement equipment at the end of the robot 11 is used to visually locate the reference point set on the top cover flow surface 21, adjust the position of the processing equipment 10, and compare it with the top cover repair processing scene model to calibrate the placement of the processing equipment 10 and the positioning of the end processing tool 12.

After the processing equipment 10 completes processing in the current processing area, it is manually pushed to the approximate position of the next processing area, and the top cover processing area is identified using a visual measurement device. The current position of the processing equipment 10 is calculated and the posture is adjusted to ensure that the position of the robot processing device is consistent with the data of the robot processing device output in the virtual simulation software.

When implementing the repair method of this application, the specific steps are as follows:

(1) Processing equipment 10 brought in and installed: The processing equipment 10 brought in and installed mainly includes the mechanical motion system brought in and installed, the electrical control system brought in and installed, the robot additive and subtractive processing system brought in and installed, the 3D visual measurement system hardware brought in and installed, and the processing process online monitoring and adaptive control system hardware brought in and installed. The details are as follows:

1) Each component of the processing equipment 10 is manually transported to the site, and the processing equipment 10 used for additive and subtractive material processing is manually moved to the equipment storage area specified in the virtual simulation software VisualComponents in the gap between the piers 30 of the turbine top cover flow surface maintenance site. The additive and subtractive material simulation scene built by the virtual simulation software guides the storage of equipment in the actual processing process.

2) Each control cabinet of the electrical control system completes the wiring in the designated storage area of the turbine top cover inspection area, and the control cabinet display panel is directly attached to the main control device. All external wiring of the control cabinet is connected with aviation plugs or other foolproof plugs to facilitate user installation and disassembly. The compressed air, cooling water, hydraulic oil and other pipelines involved in the system are connected with industry-specific pipe joints.

3) The milling and grinding system of the processing equipment 10 mainly includes a robot 11, an electric spindle, a floating spindle, a quick-change system, and mounting flanges of various components. The robot 11 is mounted on the corresponding mounting position on the lateral platform 14 of the base 13 through the base bolts. The robot side of the quick-change system is mounted on the robot end flange through the connecting flange, and the electric spindle, floating spindle, etc. are respectively mounted on the tool side of different quick-change systems through the connecting flange. The connected electric spindle and floating spindle are placed at the designated position of the tool for quick tool replacement.

4) The hardware in the 3D vision measurement and machining trajectory generation system mainly includes 3D vision measurement equipment, industrial PC, and connecting cables, which can be directly brought into the site manually. The 3D vision measurement equipment is installed on the tool side of the quick-change system through a flange, and the 3D measurement equipment and industrial PC are connected through a network cable to complete the installation.

5) The hardware of the online monitoring system for the machining process mainly includes current sensors, vibration sensors, force sensors, and data acquisition gateways, which can be brought into the site manually. The current sensor is installed on the power supply line of the electric spindle according to the sensor requirements, the vibration sensor is pasted on the end flange of the robot, and the force sensor is installed between the end flange of the robot and the robot side of the quick-change system through the flange. The output line of each sensor is connected to the corresponding input interface of the data acquisition gateway, and then the industrial PC and the data acquisition gateway are connected through the network cable to complete the installation.

(2) Robot hand-eye calibration: The purpose of robot hand-eye calibration is to obtain the transformation relationship between the visual measurement coordinate system and the robot end flange coordinate system, so that the measurement point cloud data and the processing trajectory can be expressed in the same coordinate system later. It is an important part of the measurement system construction and directly affects the measurement accuracy.

(3) Positioning of processing equipment 10: After the processing equipment 10 is manually transported into the site, the robot processing equipment is first transported to the initial processing area. The robot 11 is placed on the base 13 using the additional seventh axis to achieve horizontal linear motion of the robot 11. The teaching pendant is operated to control the joint angle of the mobile robot to a suitable position. The visual system is controlled to take photos to identify the current position of the robot. Using the reference feature data of the identified processing area, the processing equipment 10 is manually moved to the top cover processing area and the robot positioning calibration is achieved after slight adjustments. In order to facilitate movement, casters with brakes are set at the bottom of the processing equipment 10.

(4) Three-dimensional visual measurement and identification of features to be processed: After the processing equipment 10 is positioned, the industrial PC controls the visual measurement equipment to measure the processing area by communicating with the robot and the electrical control system to obtain the robot's position on the flow surface of the turbine top cover and the robot's own posture. The sector point cloud of the defective area of the flow surface of the top cover is scanned by the visual measurement equipment, and the flow surface point cloud data is obtained and uploaded to the industrial PC. The point cloud processing software installed on the industrial PC is used for feature recognition, and the robot base coordinates are calculated using the feature recognition function.

(5) Trajectory planning of defective areas on the flow surface of the turbine top cover

The point cloud of the defective area on the flow surface of the turbine top cover is reverse modeled, and a robot simulation processing platform is built in the offline programming software SprutCAM. The robot processing process trajectory route is designed according to the relevant processes of additive and subtractive materials. The post-processing function module attached to the offline programming software is used to realize the robot processing trajectory processing. The motion simulation function is used to simulate the post-processed robot processing trajectory to ensure the reliability of the robot processing trajectory. The verified robot processing trajectory program is downloaded to the robot controller, and the robot base coordinates and workpiece coordinates are obtained after point cloud feature recognition to realize the visual guidance of the robot processing process. In addition, in the top cover in-situ additive and subtractive material repair scene built in the virtual simulation software, the adjacent repair processing areas of the top cover are marked with features. In the actual processing application, the visual measurement system is used to accurately obtain the position transformation data during the robot movement process. The matching positioning of the robot in the top cover processing scene is completed by feature matching, and the robot is assisted to quickly locate to the adjacent processing area to facilitate the robot to quickly identify and position.

(6) Processing: According to the different processes required for the processing area and the processing area, a laser cladding head, electric spindle milling or floating spindle material addition and subtraction, and grinding tools are selected to perform repair processing on the defective area of the turbine top cover. In view of the repair process flow of the defective area of the turbine top cover, the present invention adopts a process method of first removing the defects by air gouging, then milling and grinding the welding area in the defective area after air gouging, and then using laser cladding to enhance the strength of the top cover for repair. Specifically, it includes: firstly, a large-scale defective area is quickly gouged manually, and a robot is connected to the electric spindle to install a milling cutter for fine shaping and milling to remove a large amount of residual height generated by air gouging. After milling, the remaining excess of about 0.2 mm is removed by connecting the robot to the floating spindle to install a louver or sandpaper and other grinding tools. The switching between the electric spindle and the floating spindle is realized by a quick change device. After grinding, the defective area of the turbine top cover is welded and regularized by the robot to repair the flow surface of the top cover, and the robot is loaded with a grinding tool to shape the welding surface to lay the foundation for the robot laser cladding to strengthen the flow surface.

(7) Online monitoring of the machining process: During the process of adding or subtracting materials by the turbine top cover robot, factors such as system positioning error and measurement error may cause uneven machining allowance, resulting in poor machining consistency, machining vibration, tool breakage, etc. In order to avoid the above situation or take timely control measures after the above situation occurs, the present invention adopts real-time online monitoring of the machining process, monitors the cutting force, vibration, and electric spindle current of the machining process in real time, and generates corresponding strategies for control. The specific process is as follows:

Robot machining cutting force monitoring and force-position hybrid control: A force sensor is installed on the end flange of the robot. Since the coordinate system where the cutting force is located is inconsistent with the coordinate system where the force vector detected by the sensor is located, the coordinate conversion relationship between the measured force/torque and the cutting force is obtained using the principle of virtual work. A decoupling model between the measured force/torque and the cutting force is established to achieve accurate measurement of the grinding force. A data acquisition gateway is used to collect cutting force data in real time and upload it to an industrial PC for processing. The industrial PC uses the independently developed robot machining online monitoring and control software to visually monitor the cutting force data. At the same time, the real-time cutting force data is compared with the set cutting force threshold, and a feedback control strategy is generated (adjusting the robot posture). Instructions are issued to adjust the robot's real-time posture to ensure that the cutting force is constant during the machining process and achieve force-position hybrid control.

(8) The processing equipment 10 moves to the next processing area to perform robot processing device positioning and processing. After the processing equipment 10 completes the processing of the current area of the top cover, the robot is first controlled to return to a safe position, and then the processing equipment 10 is manually moved to an adjacent processing area to compare the device placement posture output by the virtual simulation software to adjust the robot processing device posture. Finally, the robot vision system is used to obtain the top cover processing area data and combine the robot processing posture of the virtual simulation to determine the robot processing status, and the robot top cover repair is restarted.

The robot top cover additive and subtractive material processing and the turbine top cover flow surface defective area additive and subtractive material repair processing of the present invention can achieve the robot's rapid repair accuracy and efficiency requirements for the turbine top cover defective area.

The processing equipment 10 of the present invention can realize horizontal linear motion, and the whole includes a base 13, a lateral platform 14 and a robot 11. The weight of a single part does not exceed 50 kg. For easy movement, casters with brakes are installed under the base 13, as shown in Figure 3. The processing equipment 10 is convenient for on-site transportation, installation and debugging of additive and subtractive material repair.

The three-dimensional visual measurement and processing trajectory generation system included in the present invention mainly realizes the contour measurement of the repair area of the flow surface of the turbine top cover, the feature recognition and processing trajectory visual guidance of the area to be repaired, and the robot movement and positioning visual guidance of the processing area adjacent to the top cover. In terms of hardware, a binocular structured light camera is selected as a three-dimensional measurement device to perform three-dimensional measurement on the processing area to obtain surface point cloud data of the defective area. The binocular structured light camera is small in size and can be directly installed on the robot body. The robot hand-eye calibration function module is used to realize the precise calibration of the robot eye on the hand. The workpiece point cloud obtained by the camera is converted to the robot base coordinates through the space coordinate conversion matrix, which can be intuitively displayed in the software for operator observation and robot visual guidance.

The robot additive and subtractive processing included in the present invention mainly realizes the execution of robot additive and subtractive processing instructions in terms of function, including robot tool replacement, robot milling, robot tool parameter setting, etc. in the robot subtractive link. The robot milling process mainly relies on the electric spindle device installed at the end of the robot to clamp the milling tool, and the robot grinding and polishing process mainly relies on the floating electric spindle installed at the end of the robot to cooperate with the robot grinding and polishing tools (rotary files, louvers, grinding wheels, sandpaper, etc.) to realize the grinding and polishing of the workpiece surface. The robot tool changing function is generally realized by a quick-change mechanism. The robot welding and cladding work in the robot additive link is carried out with the help of the quick-change mechanism installed at the end of the robot, and the robot quick-change mechanism is used to realize the rapid replacement of robot processing tools, so as to quickly realize the change of robot processing technology.

The robot processing online monitoring system included in the present invention mainly realizes the equipment status monitoring during the robot additive and subtractive material processing. The operator can obtain the robot processing status in real time through the equipment status monitoring system and deal with emergencies in time. The hardware of the robot equipment status monitoring system mainly includes force sensors, current sensors, vibration sensors, data acquisition gateways and other data acquisition and transmission equipment. The software included in the robot additive and subtractive material processing system uses the independently developed robot processing online monitoring and adaptive control software. The corresponding system software runs on an industrial PC, analyzes the collected data, and generates adaptive control strategies (such as adjusting the robot posture, adjusting the robot terminal movement speed, etc.) and sends them to the robot for execution.

Claims (10)

1. A hydraulic turbine top cover robot repair method based on digital twin, characterized in that it includes the following steps: S1. Establishing a three-dimensional model of the processing equipment (10) and the top cover (20); S2. Building a top cover repair processing scene model, by simulating the placement of the processing equipment (10), forming a processing placement scheme for the processing equipment (10) relative to the top cover flow surface (21); S3. Obtaining the relative position data of the processing equipment (10) in the top cover repair processing scene model, and planning the processing trajectory of the robot (11) of the processing equipment (10); S4. Iteratively optimizing the processing area and processing trajectory of the processing equipment (10) by adjusting the installation position of the end processing tool (12) of the robot (11) and the placement position of the processing equipment (10) relative to the top cover (20); S5. Planning the trajectory and processing technology of the processing equipment (10), outputting the position data of the processing equipment (10) relative to the processing area and the processing program after iterative optimization, which are used to guide the position placement and processing of the processing equipment (10) in the actual processing process; S6. According to the built top cover repair processing scene model, the top cover (20) is installed and placed on the site in accordance with the design of the model in the actual site; S7. Based on the output processing equipment (10) processing placement plan, the actual position of the processing equipment (10) is calibrated by the robot (11), and compared with the top cover repair processing scene model to calibrate the placement position of the processing equipment (10); S8. After the processing equipment (10) completes the positioning calibration, the processing program is loaded in the teaching pendant of the robot (11) to perform the repair processing of the top cover (20) processing area; S9. After the processing equipment (10) completes the repair processing of one processing area, the processing equipment (10) is moved to the next processing area, and the top cover repair processing is started for the next processing area. 2. The method for repairing a turbine top cover by a robot based on digital twinning according to claim 1 is characterized in that: in S1, a three-dimensional point cloud acquisition device is used to collect point clouds of the top cover flow surface (21) area, point cloud processing technology is used to realize point cloud processing of the top cover flow surface (21) area, point cloud noise points are filtered out, and a model of the top cover flow surface (21) is reconstructed using reverse modeling technology, and then a three-dimensional model of the top cover (20) is established in a three-dimensional software in combination with the original drawing data of the top cover (20), and then modeling is performed in the three-dimensional software according to the size data of the processing equipment (10) to obtain a three-dimensional model of the processing equipment (10). 3. The method for repairing a turbine top cover robot based on digital twin according to claim 1 is characterized in that: in S2, a top cover repair processing scene model is established in a three-dimensional software, the top cover repair processing scene model and the three-dimensional model of the processing equipment (10) are imported into the three-dimensional simulation software and the offline programming software, the top cover flow surface processing scene is simulated, and the placement position of the processing equipment (10) is moved in the constructed top cover repair processing scene model to form a processing placement plan for the processing equipment (10) relative to the top cover flow surface (21). 4. The method for repairing a turbine top cover robot based on digital twin according to claim 1 is characterized in that: in S3, the top cover repair processing scene model laid out in the three-dimensional simulation software is used to obtain the position data of the processing equipment (10) relative to the processing area of the top cover flow surface (21), the position data of the processing equipment (10) relative to the processing area is modified in the offline programming software, and the trajectory planning function and the end processing tool adjustment function of the offline programming software are used to select the processing area for processing trajectory planning. 5. The digital twin-based turbine top cover robot repair method according to claim 1, characterized in that: in S4, by adjusting the installation position of the end processing tool (12) of the robot (11) and the placement position of the processing equipment (10) relative to the top cover (20) in the offline programming software, the processing area and the processing trajectory are iteratively optimized; wherein the iterative optimization of the processing trajectory adopts a method of optimizing the installation position data of the processing equipment (10) relative to the processing area, the installation position data of the end processing tool (12) of the robot (11), and the initial joint posture of the robot (11) processing, so as to expand the processing area of the top cover flow surface (21) for trajectory planning, and finally obtain multiple processing areas that can cover the entire top cover flow surface (21). 6. The digital twin-based turbine top cover robot repair method according to claim 1 is characterized in that: in S5, the installation position data of the end processing tool (12) of the robot (11) is output to guide the design of the additive and subtractive tool fixture, the safe position point of the robot (11) in the processing area of the top cover flow surface (21) is output to protect the safety of the robot's initial position, the initial joint angle data of the robot (11) processing trajectory is output to guide the robot's safe posture during the processing, and the mobile positioning data of the processing equipment (10) is output to guide the processing equipment (10) to move to the next processing area for repeated processing. 7. The digital twin-based turbine top cover robot repair method according to claim 1 is characterized in that: in S6, a plurality of piers (30) are first arranged on the site according to the design of the model, and the top cover (20) is placed on the piers (30) to restore the top cover repair processing scene in the model. 8. The method for repairing a turbine top cover based on digital twin according to claim 1, characterized in that: in S7, after placing the processing equipment (10) below the top cover (20), the actual position of the processing equipment (10) is calibrated by using the robot (11) with reference to the relative position data of the processing equipment (10) relative to the processing area output by the simulation software, and the actual position of the processing equipment (10) is compared with the simulation data to calibrate and move the position of the processing equipment (10); wherein the positioning method of the processing equipment (10) is to use a visual measurement device installed at the end of the robot (11) to visually locate a reference point set on the flow surface (21) of the top cover, calculate the relative position of the reference point relative to the base coordinate of the processing equipment (10), and perform data calibration and comparison in combination with the relative position data of the reference point relative to the base coordinate of the processing equipment (10) in the simulation software, adjust the processing position of the actual processing equipment (10), and realize the positioning of the end processing tool (12). 9. The digital twin-based turbine top cover robot repair method according to claim 1 is characterized in that: in S8, the robot processing program planned by offline programming is copied to the controller of the robot (11) through a USB flash drive or a data cable transmission method, the processing program is called in the controller to perform an empty walk in the top cover processing area to verify the feasibility of the program, and the processing program is executed again in conjunction with the control logic of the end processing tool (12) to realize the robot top cover additive and subtractive processing and repair. 10. The digital twin-based turbine top cover robot repair method according to claim 1 is characterized in that: in S9, after the robot (11) completes the defect repair processing in one processing area, the processing equipment (10) is manually pushed to the next processing area, and then the visual measurement equipment at the end of the robot (11) is used to visually locate the reference point set on the top cover flow surface (21), adjust the position of the processing equipment (10), and compare it with the top cover repair processing scene model, calibrate the placement of the processing equipment (10), and calibrate the positioning of the end processing tool (12).