High-speed rail head surface coating robot self-adaptive grinding and polishing machining method and system

Abstract

The invention belongs to the technical field of grinding control, and particularly discloses a high-speed rail head surface coating robot self-adaptive grinding and polishing machining method and system. Comprising the following steps: reconstructing an actual three-dimensional model according to a three-dimensional theoretical model of a high-speed train head and three-dimensional shape data obtained by measurement; on the basis of comparison of the actual three-dimensional model and the theoretical three-dimensional model, obtaining a distribution topological graph of putty coating material removal allowance on the surfaces of different parts; constructing a contact force physical model of the robot tail end grinding disc and the high-speed rail head workpiece; according to the distribution of the removed margins and the size and shape of the high-speed train head, robot grinding and polishing machining areas are divided, and an active-disturbance-rejection fuzzy variable resistance control method is adopted to achieve local area constant force grinding accurate control; and a high-speed rail head putty material accurate removal model is established, and robot grinding and polishing process parameters of corresponding parts are reversely obtained based on the surface putty coating material removal allowance of different parts. The overall efficiency and quality of processing of the putty coating on the surface of the high-speed train head are remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of grinding control, and more specifically, relates to a high-speed railway head surface coating robot adaptive grinding and polishing processing method and system. Background Art

[0002] In recent years, the mileage of high-speed rail construction in my country has continued to grow rapidly, and the production of high-speed rail locomotives has also put forward a more vigorous demand. In the production process of the locomotive, in order to meet the requirements of the vehicle surface beauty and flatness as well as the acid rain, exposure and other environmental requirements during service, the requirements for the coating operation of the locomotive surface are very high. In order to meet the surface quality requirements of the locomotive, it is necessary to scrape putty on the surface of the locomotive several times to level it, and then grind, apply putty, grind, spray primer, grind, spray middle paint, grind, and spray topcoat. During the operation, grinding operations need to be performed many times, and the quality of grinding not only affects the flatness and beauty of the locomotive surface, but also has a direct impact on the adhesion of the next coating and the quality and life of the entire coating. After the high-speed rail locomotive is coated with white putty, its profile almost universally needs to be polished, so the quality of grinding directly determines the service life of the high-speed rail locomotive.

[0003] At present, the front painting process of high-speed rail enterprises in my country is still mainly done manually, especially the grinding operation is basically all done manually. The grinding process is labor-intensive, inefficient, and in a poor environment. It is easily affected by subjective factors such as the operator's mood and physical strength. The operation is arbitrary, the grinding removal amount is difficult to stably control, and poor grinding areas are prone to occur. At the same time, a large amount of dust is generated during the grinding process of putty, and there are also serious dust pollution and health hazards. In addition, the front of the high-speed rail presents a streamlined shape. It is the most important large and complex component on the high-speed rail vehicle. It is composed of nearly 100 pieces of skins of different shapes. It has the characteristics of complex curved surface modeling and variable / weak structural rigidity, resulting in a high product scrap rate. Intelligent manufacturing represented by robot technology provides a direction for achieving high-quality automated grinding and polishing of such large curvature and variable curvature components. Combining the advantages of robots such as high flexibility, large working space, and flexible configuration of multiple sensors with grinding and polishing processing technology is expected to solve the disadvantages of traditional processing methods in processing accuracy, product consistency, processing efficiency, and working environment, thereby improving the surface processing quality and efficiency of workpieces. Summary of the invention

[0004] In view of the above defects or improvement needs of the prior art, the present invention provides a method and system for the adaptive grinding and polishing of the surface coating of the high-speed railway head by a robot, which integrates structured light measurement and photometric stereo vision technology, and combines the improved K-means clustering algorithm to process three-dimensional morphological data, thereby realizing the accurate three-dimensional measurement and model reconstruction of the putty coating on the surface of the high-speed railway head. At the same time, by establishing an accurate contact force model and a material removal theoretical model, adaptive grinding force control of different areas of the high-speed railway head is realized, the grinding accuracy and surface quality are improved, and the stability and consistency of the processing process are ensured, thereby significantly improving the overall efficiency and quality of the high-speed railway head coating processing.

[0005] To achieve the above object, according to one aspect of the present invention, a method for adaptive grinding and polishing of a high-speed railway head surface coating by a robot is proposed, comprising the following steps:

[0006] Step 1: reconstructing the actual 3D model of the high-speed train head according to the 3D theoretical model of the high-speed train head and the 3D shape data obtained by measurement;

[0007] Step 2: Based on the actual three-dimensional model and the three-dimensional theoretical model, a distribution topological map of the surface putty coating material removal residue at different parts is obtained;

[0008] Step 3: Construct a contact force model between the grinding disc at the end of the robot and the workpiece at the front of the high-speed railway;

[0009] Step 4: According to the distribution of the removed excess and the size and shape of the high-speed rail head, the robot grinding and polishing processing area is divided, and a fuzzy variable resistance control method with self-disturbance rejection is used to achieve precise control of constant force grinding in local areas;

[0010] Step five: Establish a precise removal model for putty materials on the front of a high-speed railway train, and reversely obtain the robot grinding and polishing process parameters of the corresponding parts based on the removal amount of surface putty coating materials in different parts.

[0011] As further preferred, step one comprises the following steps:

[0012] (11) Import the three-dimensional theoretical model of the high-speed rail head and build a robot measurement workstation for the high-speed rail head in the simulation software to generate the robot's scanning path;

[0013] (22) obtaining time-synchronized three-dimensional shape information of the high-speed train head and depth information and photometric information of the putty layer;

[0014] (23) The three-dimensional shape information and the depth information of the putty layer are fused with the photometric information. During the fusion process, the improved K-means clustering algorithm is used to establish the spatial topological relationship of the point cloud. Then, the noise points of each type of point cloud after clustering are identified and removed. The data is then denoised, streamlined, fitted, spliced, and smoothed to obtain a point cloud model of the putty layer on the surface of the high-speed rail head.

[0015] (24) Based on the point cloud model, gradient model, photometric information, and depth information of the putty layer on the surface of the high-speed train head, the actual three-dimensional model of the high-speed train head is reconstructed in reverse.

[0016] As further preferred, step 2 comprises the following steps:

[0017] Based on the Preston wear law, a theoretical model of material removal is established. A microelement of dx*dy is taken on the surface to be polished. In this theoretical model of material removal, the relative sliding velocity v a Approximately equal to the linear velocity v of the abrasive s (x,y), let D l is the removal depth per unit contact length between the polishing disc and the rubber surface along the polishing path, and can be obtained as:

[0018]

[0019] Among them, v f is the feed rate, β is the Preston wear coefficient;

[0020] Based on the constructed actual 3D model and 3D theoretical model, the distribution topology of the surface putty coating material removal residue in different parts is accurately obtained.

[0021] As a further preferred embodiment, in step three, constructing the contact force model between the grinding head and the workpiece includes:

[0022] (31) Using the Galerkin method, assuming that the left and right boundaries of the circular thin plate are in the shape of a quadratic curve after contact deformation, the displacement of each point on the thin plate is calculated to satisfy:

[0023]

[0024] Among them, h corresponds to the thickness of the circular thin plate after deformation, and d corresponds to the thickness of the circular thin plate compressed in the z-axis direction;

[0025] (32) Assuming that all mechanical energy is converted into shear potential energy, the shear potential energy, i.e., the total energy, is derived from the shear strain:

[0026]

[0027] Among them, G is the shear modulus, and R corresponds to the radius of the bottom surface of the circular thin plate before deformation;

[0028] (33) According to the relationship between energy and force, the contact force on the grinding disc can be obtained:

[0029]

[0030] Among them, F N is the contact force on the grinding disc, and U is the total energy.

[0031] As a further preferred embodiment, step three further includes: analyzing the stress distribution of the grinding disc, specifically:

[0032] Using the infinitesimal method, take an infinitesimal element with a length of Δr for the portion where r=r0, and calculate the stress distribution at the bottom of the grinding disc according to the calculation method of the shear potential energy and energy in step (32) and the calculation method of the contact force on the grinding disc in step (33):

[0033]

[0034] Among them, σ(r,0) is the stress of one microelement at the bottom of the grinding disk.

[0035] As a further preferred embodiment, step three further comprises:

[0036] The circular thin plate model includes the sponge pad and sandpaper on the grinding head. According to the deformation relationship between the sponge pad and the sandpaper, the comprehensive shear modulus of the grinding disc is derived as follows:

[0037]

[0038] Among them, E1 and E2 correspond to the elastic modulus of the sponge pad and sandpaper, d1 and d2 correspond to the original thickness of the sponge pad and sandpaper, and v1 and v2 correspond to the Poisson's ratio of the sponge pad and sandpaper.

[0039] As a further preference, in step four, the fuzzy variable resistance control method with self-disturbance rejection includes a fuzzy variable impedance controller and a self-disturbance rejection polishing controller, wherein the fuzzy variable impedance controller is used for the inner loop and the self-disturbance rejection polishing controller is used for the outer loop, and the two constitute a closed-loop control circuit.

[0040] As a further preferred embodiment, the fuzzy variable resistance control method with self-disturbance rejection to achieve precise control of constant force grinding includes:

[0041] (41) Given the desired position, the anti-disturbance grinding controller estimates the grinding model and robot disturbance state online and compensates the signal based on the current working position of the grinding tool fed back by the position sensor;

[0042] (42) The output value of the anti-disturbance grinding controller is used as the input of the impedance model. The relationship between the telescopic displacement of the end effector and the contact force is established through the model. The output force correction value compensates the force to make the force deviation close to zero, thereby preventing over-cutting caused by overshoot of the contact force.

[0043] (43) The impedance parameters of the impedance model are fuzzy controlled and combined with the force feedback information considering the environmental stiffness compensation to optimize the output.

[0044] According to another aspect of the present invention, there is also provided a high-speed railway head surface coating robot adaptive grinding and polishing system, comprising:

[0045] Actual 3D model reconstruction module: used to reconstruct the actual 3D model of the high-speed train head according to the 3D theoretical model of the high-speed train head and the 3D shape data obtained by measurement;

[0046] Material removal amount calculation module: used to obtain a distribution topological map of the surface putty coating material removal allowance at different locations based on the actual three-dimensional model and the three-dimensional theoretical model;

[0047] Robotic grinding module: including a robot and a grinding disc designed at the end of the robot for performing grinding;

[0048] Robot grinding control module: used to build the contact force model between the grinding disc and the workpiece, and divide the robot grinding and polishing processing area according to the distribution of the removal margin and the size and shape of the high-speed rail head, and adopt the fuzzy variable resistance control method with self-disturbance rejection to achieve precise control of constant force grinding in local areas;

[0049] Robot processing parameter calculation module: used to establish a precise removal model for putty materials on the front of a high-speed railway, and to reversely obtain the robot grinding and polishing process parameters of the corresponding parts based on the removal allowance of the surface putty coating material on different parts.

[0050] As further preferred, the actual three-dimensional model reconstruction module includes:

[0051] Structured light projection system: used to obtain the structured light point cloud of the high-speed rail head and reconstruct the three-dimensional shape information of the head based on the structured light point cloud;

[0052] Photometric stereo vision camera system: used to obtain depth information and photometric information of the putty layer;

[0053] Synchronization system: used to synchronize the working time of the structured light projection system and the photometric stereo vision camera system;

[0054] Data processing module: used to fuse the 3D shape information and the depth information and photometric information of the putty layer, so as to reversely construct the actual 3D model of the high-speed rail head;

[0055] Preferably, the robot polishing module further includes horizontal guide rails and vertical guide rails disposed on both sides of the vehicle head for driving the robot to move.

[0056] In general, the above technical solution conceived by the present invention has the following technical advantages compared with the prior art:

[0057] 1. The present invention integrates structured light measurement and photometric stereo vision technology, and combines the improved K-means clustering algorithm to process three-dimensional morphological data, thus realizing accurate three-dimensional measurement and model reconstruction of putty coating on the surface of high-speed railway head. At the same time, by establishing an accurate contact force model and material removal theoretical model, adaptive grinding force control of different areas of high-speed railway head is realized, which improves grinding accuracy and surface quality, ensures the stability and consistency of the processing process, and thus significantly improves the overall efficiency and quality of high-speed railway head coating processing.

[0058] 2. The present invention designs a multifunctional sensor measurement system that integrates structured light measurement and photometric stereo vision to accurately measure the three-dimensional morphology of the putty coating on the surface of the high-speed rail head. The improved K-means clustering algorithm is used to denoise, streamline, fit, splice and smooth the data, and reversely reconstruct the actual three-dimensional model of the high-speed rail head.

[0059] 3. The method for constructing the contact force between the disc grinding head and the workpiece when grinding the putty on the front of a high-speed railway according to the present invention establishes a contact force model from the perspective of energy, which can solve the problem that a certain error exists between the contact force model modeled by geometric relationship and the actual situation due to the poor adhesion between the grinding disc and the surface, and obtains a more accurate contact force between the grinding disc and the workpiece, which is beneficial to ensure the stability and consistency during the grinding process, thereby improving the surface quality and precision of the workpiece after grinding; subsequently, the stress cloud map can be drawn to observe the utilization rate of the grinding disc for sandpaper, which can reflect the grinding efficiency to a certain extent.

[0060] 4. The high-speed railway head putty coating material removal depth model of the present invention can help predict and control the amount of material removal during the processing, thereby improving the processing accuracy and surface quality; it can identify and optimize the key parameters that affect the material removal efficiency, such as grinding and polishing force, feed speed and tool speed.

[0061] 5. The present invention divides the high-speed rail head processing area into regions and adopts different grinding forces according to the specific characteristics of each region, which helps to ensure that each region can achieve the ideal grinding accuracy, avoid over-grinding or under-grinding, ensure the flatness and smoothness of the head surface, and improve the overall grinding quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1It is a flow chart of a method for adaptive grinding and polishing of a surface coating of a high-speed railway head by a robot according to an embodiment of the present invention;

[0063] Figure 2 It is a schematic diagram of the overall layout structure of the high-speed railway head robot grinding involved in an embodiment of the present invention;

[0064] Figure 3 It is a flow chart of generating a three-dimensional model by robot scanning according to an embodiment of the present invention;

[0065] Figure 4 It is a schematic diagram of the contact force analysis between the robot end grinding head and the high-speed railway head component involved in the embodiment of the present invention;

[0066] Figure 5 is a flow chart of a method for establishing contact force between a disc grinding head and a workpiece according to an embodiment of the present invention;

[0067] Figure 6 It is a schematic diagram of the division of the grinding area of ​​the high-speed railway locomotive robot in an embodiment of the present invention. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0069] Example 1

[0070] like Figure 1 As shown, an embodiment of the present invention provides a method for adaptive grinding and polishing of a high-speed railway head surface coating by a robot, comprising the following steps:

[0071] Step 1: reconstruct the actual 3D model of the high-speed train head according to the 3D theoretical model of the high-speed train head and the 3D shape data obtained by measurement. Step 1 includes the following steps:

[0072] (11) Import the three-dimensional theoretical model of the high-speed rail head and build a robot measurement workstation for the high-speed rail head in the simulation software to generate the robot's scanning path;

[0073] (22) obtaining time-synchronized three-dimensional shape information of the high-speed train head and depth information and photometric information of the putty layer;

[0074] (23) The three-dimensional shape information and the depth information of the putty layer are fused with the photometric information. During the fusion process, the improved K-means clustering algorithm is used to establish the spatial topological relationship of the point cloud. Then, the noise points of each type of point cloud after clustering are identified and removed. The data is then denoised, streamlined, fitted, spliced, and smoothed to obtain a point cloud model of the putty layer on the surface of the high-speed rail head.

[0075] (24) Based on the point cloud model, gradient model, photometric information, and depth information of the putty layer on the surface of the high-speed train head, the actual three-dimensional model of the high-speed train head is reconstructed in reverse.

[0076] Step 2: Based on the actual three-dimensional model and the three-dimensional theoretical model, a topological map of the distribution of the surface putty coating material removal residue at different parts is obtained.

[0077] Step 2 includes the following steps:

[0078] Based on the Preston wear law, a theoretical model of material removal is established. A microelement of dx*dy is taken on the surface to be polished. In this theoretical model of material removal, the relative sliding velocity v a Approximately equal to the linear velocity v of the abrasive s (x,y), let D l is the removal depth per unit contact length between the polishing disc and the rubber surface along the polishing path, and can be obtained as:

[0079]

[0080] Among them, v f is the feed rate, β is the Preston wear coefficient;

[0081] Based on the constructed actual 3D model and 3D theoretical model, the distribution topology of the surface putty coating material removal residue in different parts is accurately obtained.

[0082] Step 3: Construct a contact force model between the grinding disc at the end of the robot and the workpiece at the front of the high-speed railway;

[0083] Step three, constructing a contact force model between the grinding head and the workpiece includes:

[0084] (31) Using the Galerkin method, assuming that the left and right boundaries of the circular thin plate are in the shape of a quadratic curve after contact deformation, the displacement of each point on the thin plate is calculated to satisfy:

[0085]

[0086] Among them, h corresponds to the thickness of the circular thin plate after deformation, and d corresponds to the thickness of the circular thin plate compressed in the z-axis direction;

[0087] (32) Assuming that all mechanical energy is converted into shear potential energy, the shear potential energy is derived from the shear strain:

[0088]

[0089] Among them, G is the shear modulus, and R corresponds to the radius of the bottom surface of the circular thin plate before deformation;

[0090] (33) According to the relationship between energy and force, the contact force on the grinding disc can be obtained:

[0091]

[0092] Among them, F N is the contact force on the grinding disc, and U is the total energy.

[0093] Step three also includes: analyzing the stress distribution of the grinding disc, specifically:

[0094] Using the infinitesimal method, take an infinitesimal element with a length of Δr for the portion where r=r0, and calculate the stress distribution at the bottom of the grinding disc according to the calculation method of the shear potential energy and energy in step (32) and the calculation method of the contact force on the grinding disc in step (33):

[0095]

[0096] Among them, σ(r,0) is the stress of one microelement at the bottom of the grinding disk.

[0097] Step three also includes:

[0098] The circular thin plate model includes the sponge pad and sandpaper on the grinding head. According to the deformation relationship between the sponge pad and the sandpaper, the comprehensive shear modulus of the grinding disc is derived as follows:

[0099]

[0100] Among them, E1 and E2 correspond to the elastic modulus of the sponge pad and sandpaper, d1 and d2 correspond to the original thickness of the sponge pad and sandpaper, and v1 and v2 correspond to the Poisson's ratio of the sponge pad and sandpaper.

[0101] Step 4: According to the distribution of the removed excess and the size and shape of the high-speed rail head, the robot grinding and polishing processing area is divided, and a fuzzy variable resistance control method with self-disturbance rejection is used to achieve precise control of constant force grinding. In this step, the fuzzy variable resistance control method with self-disturbance rejection includes a fuzzy variable impedance controller and an self-disturbance rejection grinding controller, wherein the inner loop uses a fuzzy variable impedance controller and the outer loop uses an self-disturbance rejection grinding controller, and the two constitute a closed-loop control circuit.

[0102] The fuzzy variable resistance control method with self-disturbance rejection to achieve precise control of constant force grinding includes:

[0103] (41) Given the desired position, the anti-disturbance grinding controller estimates the grinding model and robot disturbance state online and compensates the signal based on the current working position of the grinding tool fed back by the position sensor;

[0104] (42) The output value of the anti-disturbance grinding controller is used as the input of the impedance model. The relationship between the telescopic displacement of the end effector and the contact force is established through the model. The output force correction value compensates the force to make the force deviation close to zero, thereby preventing over-cutting caused by overshoot of the contact force.

[0105] (43) The impedance parameters of the impedance model are fuzzy controlled and combined with the force feedback information considering the environmental stiffness compensation to optimize the output.

[0106] Step five: Establish a precise removal model for putty materials on the front of a high-speed railway train, and reversely obtain the robot grinding and polishing process parameters of the corresponding parts based on the removal amount of surface putty coating materials in different parts.

[0107] Example 2

[0108] In this embodiment, a high-speed railway head surface coating robot adaptive grinding and polishing system is provided, comprising:

[0109] Actual 3D model reconstruction module: used to reconstruct the actual 3D model of the high-speed train head according to the 3D theoretical model of the high-speed train head and the 3D shape data obtained by measurement;

[0110] Material removal amount calculation module: used to obtain a distribution topological map of the surface putty coating material removal allowance at different locations based on the actual three-dimensional model and the three-dimensional theoretical model;

[0111] Robotic grinding module: including a robot and a grinding disc designed at the end of the robot for performing grinding;

[0112] Robot grinding control module: used to build the contact force model between the grinding disc and the workpiece, and divide the robot grinding and polishing processing area according to the distribution of the removal margin and the size and shape of the high-speed rail head, and adopt the fuzzy variable resistance control method with self-disturbance rejection to achieve precise control of constant force grinding in local areas;

[0113] Robot processing parameter calculation module: used to establish a precise removal model for putty materials on the front of a high-speed railway, and to reversely obtain the robot grinding and polishing process parameters of the corresponding parts based on the removal allowance of the surface putty coating material on different parts.

[0114] Among them, the actual three-dimensional model reconstruction module includes: a structured light projection system, which is used to obtain the structured light point cloud of the high-speed rail front and reconstruct the three-dimensional shape information of the front based on the structured light point cloud; a photometric stereo vision camera system, which is used to obtain the depth information and photometric information of the putty layer; a synchronization system, which is used to synchronize the working time of the structured light projection system and the photometric stereo vision camera system; a data processing module, which is used to fuse the three-dimensional shape information and the depth information and photometric information of the putty layer to inversely reconstruct the actual three-dimensional model of the high-speed rail front.

[0115] In addition, in this embodiment, the guide rail system in the robot grinding module is composed of parallel horizontal guide rails 7, 14, vertical guide rails 5, 12 and guide rail sealing tracks 6, 15. The horizontal guide rails 7, 14 provide a horizontal movement path for the robot, so that it can perform grinding operations at different positions; the vertical guide rails 5, 12 can move on the horizontal guide rails 7, 14 respectively, and are used to provide vertical movement guidance for the robot, further expanding the coverage of grinding; the guide rail sealing tracks 6, 15 are used to protect the guide rail system, preventing dust, iron filings and other impurities from entering the guide rails and affecting normal operation.

[0116] The robot part consists of robots 1 and 8, curved surface flexible grinding heads 3 and 10, and shape measurement devices 2 and 9. Robots 1 and 8 are high-performance industrial robots with multiple joint degrees of freedom, and are the core executors of grinding operations. With advanced control systems and flexible mechanical structures, they can accurately perform various grinding tasks in different working scenarios; curved surface flexible grinding heads 3 and 10 are installed at the ends of robots 1 and 8 respectively, and are used for high-speed rail head curved surface grinding. They have good flexibility and adaptability, can fit closely to the head surface, and achieve uniform and efficient grinding effects.

[0117] Example 3

[0118] In this embodiment, the self-adaptive grinding and polishing method of the high-speed railway head surface coating robot is as follows:

[0119] First, the 3D theoretical model of the high-speed train head is imported, a robot measurement workstation is built in the simulation software, and the robot scanning path is generated, so as to quickly and accurately scan the 3D morphology of the high-speed train head and obtain the 3D point cloud data of the actual component. Then, the scanned 3D surface data is denoised, streamlined, fitted, spliced ​​and smoothed by the improved K-means clustering algorithm, and the actual 3D model of the high-speed train head is reversely reconstructed.

[0120] Then, a theoretical model of material removal was established according to the Preston wear law. Based on the constructed three-dimensional model and theoretical model, the topological map of the excess distribution of putty coating materials on the surface of different parts was accurately obtained. After that, the contact force model between the disc grinding head and the workpiece when grinding the putty of the high-speed rail body was constructed, and the contact force and stress distribution were derived from the energy perspective, and comprehensive physical parameters such as the comprehensive shear modulus were calculated. The robot grinding and polishing processing area was accurately divided based on the removal excess distribution and the size and shape of the high-speed rail head. Different grinding forces were used in different areas, and a fuzzy variable impedance control method with self-anti-interference was proposed to achieve precise control of the grinding force. At the same time, a precise material removal model for the high-speed rail head was established, and the robot grinding and polishing process parameters were reversely obtained based on the material removal excess of different parts.

[0121] Finally, the automatic grinding program of the high-speed rail head component robot is executed. After grinding, the surface morphology is measured and the processing allowance is calculated again. If all areas meet the requirements, the grinding is completed and enters the next process; if not, the area continues to execute the grinding program until the material removal amount and surface consistency requirements are met.

[0122] Specifically, Figure 2 As shown, in the above method, the robot grinding module is introduced as follows: the guide rail system is composed of parallel horizontal guide rails 7, 14, vertical guide rails 5, 12 and guide rail sealing tracks 6, 15. The horizontal guide rails 7, 14 provide a horizontal movement path for the robot, so that it can perform grinding operations at different positions; the vertical guide rails 5, 12 can move on the horizontal guide rails 7, 14 respectively, and are used to provide vertical movement guidance for the robot, further expanding the coverage of grinding; the guide rail sealing tracks 6, 15 are used to protect the guide rail system to prevent dust, iron filings and other impurities from entering the guide rails and affecting normal operation.

[0123] The robot part consists of robots 1 and 8, curved surface flexible grinding heads 3 and 10, and shape measuring devices 2 and 9. Robots 1 and 8 are high-performance industrial robots with multiple joint degrees of freedom, and are the core executors of grinding operations. With advanced control systems and flexible mechanical structures, they can accurately perform various grinding tasks in different working scenarios; curved surface flexible grinding heads 3 and 10 are respectively installed at the ends of robots 1 and 8, and are used for high-speed rail head curved surface grinding. They have good flexibility and adaptability, can fit the head surface tightly, and achieve uniform and efficient grinding effects; shape measuring devices 2 and 9 are respectively installed on robots 1 and 8, and are used to measure the surface of the high-speed rail head in real time, obtain surface shape information, accurately detect the surface state before and after grinding, and provide feedback data for the system. The control system adjusts the robot's motion trajectory and grinding parameters based on these feedback information to ensure that the grinding quality meets the requirements.

[0124] The robot fixed bases 4 and 11 provide stable support for the robots 1 and 8, ensuring that they will not shake or move during the working process, thereby ensuring the accuracy and stability of grinding.

[0125] When the system starts working, the robot 1 moves to the corresponding position of the high-speed rail head 16 under the guidance of the horizontal guide rail 14 and the vertical guide rail 11. The curved surface flexible grinding head 3 is adjusted according to the shape of the workpiece and the grinding requirements, and is in close contact with the surface of the workpiece. During the grinding process, the morphology measurement device 2 continuously monitors the morphology changes of the workpiece and transmits the data to the control system. The control system adjusts the movement and grinding parameters of the robot 1 in real time according to the feedback information to ensure the consistency and high quality of the grinding effect. At the same time, the robot fixed base 4 ensures the stability and accuracy of the robot during work, making the entire grinding process efficient and reliable.

[0126] In order to construct the actual 3D model of the high-speed train head, it is first necessary to import the 3D theoretical model of the high-speed train head and build a robot measurement workstation of the high-speed train head in the simulation software to generate the robot's scanning path. Due to the weak surface texture of the putty coating, small light intensity difference, flat surface, and unclear putty defect characteristics, the sensing measurement system of the scanning robot should combine structured light measurement and photometric stereo vision technology. Specifically, the system should include: an integrated structured light projection device for obtaining 3D shape information; a photometric stereo vision camera system for obtaining depth information and photometric information of the putty layer; and a synchronization system to ensure that the structured light projection system and the photometric stereo vision camera system can work synchronously so as to obtain information at the same time.

[0127] After collecting the three-dimensional shape information, depth information and photometric information of the high-speed rail surface, these data will be transmitted to the multifunctional sensor fusion and data processing system. First, the data of the structured light measurement system and the photometric stereo vision system are fused to comprehensively consider the advantages of the two sensors; then the data obtained by the two sensors are aligned to ensure their consistency in space; secondly, the three-dimensional shape information reconstructed by structured light is fused with the depth information estimated by stereo vision to obtain more accurate and complete surface shape data, and feature extraction and analysis are performed on the fused data; after that, the system is calibrated and calibrated to ensure the accurate correspondence between the structured light projection and the stereo vision system and the correct alignment of the photometric information and the depth information.

[0128] In order to extract the ideal target point cloud from the scattered point cloud data containing noise, the spatial topological relationship of the point cloud is established by the improved K-means clustering algorithm, and then the noise points of each type of point cloud after clustering are identified and removed, and then the data is denoised, streamlined, fitted, spliced ​​and smoothed. Finally, the three-dimensional morphology of the putty coating on the surface of the high-speed railway can be accurately measured under low light intensity differences. The point cloud model, gradient model, photometric information and depth information of the putty layer on the surface of the high-speed railway head are obtained through the measuring device, and the actual three-dimensional model of the high-speed railway head is reversely reconstructed, such as Figure 3 shown.

[0129] Furthermore, in this embodiment, the material removal theoretical model is established according to the Preston wear law. Take a microelement of dx*dy on the polished surface. In the working condition of this embodiment, the feed speed is much smaller than the linear speed of the polishing disc, so the relative sliding speed v in this model is a Approximately equal to the linear velocity v of the abrasive s (x, y), combined with the stress distribution at the bottom of the grinding disc that has been derived in the previous part, let D l is the removal depth per unit contact length between the polishing disc and the rubber surface along the polishing path, and can be obtained as:

[0130]

[0131] Where dh is the material removal depth of the microelement, v f is the feed rate, and β is the Preston wear coefficient.

[0132] Based on the constructed three-dimensional model and theoretical model, the distribution topology of the surface putty coating material removal residue in different parts is accurately obtained.

[0133] After obtaining a relatively accurate material removal allowance model, a suitable grinding and polishing force model needs to be established. In the scenario of grinding the putty on the head of a train, the current grinding and polishing force model has two problems: first, the putty removal allowance on the surface of the high-speed train head is very small and unevenly distributed, and the grinding disc rotates at a high speed, which makes it difficult to remove the material accurately and unevenly, seriously affecting the surface processing consistency of the high-speed train head; second, the high-speed train head is a streamlined structure with large changes in curvature, such as Figure 4 As shown in the figure, the actual grinding point is inconsistent with the theoretical grinding point. The center line of the robot end at the actual processing position contact point and the workpiece normal form an angle α, and α>0, resulting in a certain error in the contact force model modeled by geometric relationships.

[0134] In addition, in this embodiment, a method for constructing the contact force between a disc grinding head and a workpiece when grinding putty on a high-speed railway head is provided. The method mainly includes two aspects: on the one hand, a normal contact force model between the grinding disc and the workpiece is established; on the other hand, comprehensive physical parameters of the grinding disc are calculated. Figure 5 shown.

[0135] Simulate the force scenario when the sponge pad and sandpaper on the grinding disc are in direct contact with the workpiece, i.e. the putty surface: During the process of grinding putty, the contact that occurs is the direct contact between the sponge pad and sandpaper on the grinding disc surface and the putty. Due to the large volume difference between the grinding disc and the workpiece, the sponge pad and sandpaper can be regarded as an elastic circular thin plate, and the material is regarded as incompressible; the deformation degree of the putty, i.e. the workpiece surface, relative to the sponge pad and sandpaper can be ignored, that is, the workpiece surface can be regarded as a rigid plane; the upper surface of the sponge pad is regarded as a fixed plane.

[0136] Using the Galerkin method, assuming that the left and right boundaries of the circular thin plate are in the shape of a quadratic curve after contact deformation, the displacement of each point on the thin plate can be calculated to satisfy:

[0137]

[0138] Among them, h corresponds to the thickness of the circular thin plate after deformation, and d corresponds to the thickness of the circular thin plate compressed in the z-axis direction.

[0139] Shearing is the most important mechanical behavior in the grinding process. Therefore, in this contact engineering, for circular thin plates, it can be assumed that all mechanical energy is converted into shear potential energy. The shear potential energy, i.e. the total energy, is derived from the shear strain:

[0140]

[0141] Among them, G is the shear modulus, and R corresponds to the bottom radius of the circular thin plate before deformation.

[0142] According to the relationship between energy and force, the contact force on the grinding disc can be obtained:

[0143]

[0144] In actual grinding, in order to ensure the grinding quality and the service life of the grinding disc, it is necessary to analyze the stress distribution of the grinding disc. Using the infinitesimal method, take a infinitesimal element with a length of Δr for the part where r=r0, and use the above energy and contact force calculation method to deduce the stress distribution at the bottom of the grinding disc:

[0145]

[0146] In the expressions of contact force and stress distribution derived above, the shear modulus is an important coefficient that is positively correlated with the force. At the same time, in this application scenario, the circular thin plate model is composed of the sponge pad and sandpaper on the grinding head, so it is necessary to make appropriate processing and calculations for various physical parameters including the shear modulus. This deformation process is elastic deformation. According to the deformation relationship between the sponge pad and the sandpaper, the comprehensive shear modulus of the grinding disc can be deduced as:

[0147]

[0148] Among them, E1 and E2 correspond to the elastic modulus of the sponge pad and sandpaper, d1 and d2 correspond to the original thickness of the sponge pad and sandpaper, and v1 and v2 correspond to the Poisson's ratio of the sponge pad and sandpaper.

[0149] After calculating the shear modulus based on the physical parameters of the material, it can be substituted into the contact force and stress distribution formula to obtain more accurate contact force and stress distribution results.

[0150] Furthermore, in this embodiment, in order to realize constant force tracking grinding, a fuzzy variable impedance control method with self-disturbance rejection is proposed. The inner loop uses a fuzzy variable impedance controller, and the outer loop uses a self-disturbance rejection grinding controller, which constitute a closed-loop control circuit. The specific process is as follows:

[0151] 1) Given the desired position, the anti-disturbance grinding controller estimates the grinding model and robot disturbance state online and compensates the signal based on the current working position of the grinding tool fed back by the position sensor.

[0152] 2) The output value of the anti-disturbance grinding controller is used as the input of the impedance model. The relationship between the telescopic displacement of the end effector and the contact force is established through the model. The output force correction amount compensates the force to make the force deviation close to zero, thereby preventing overcutting caused by overshoot of the contact force.

[0153] 3) The impedance parameters of the impedance model are fuzzy controlled and the output is optimized by combining the force feedback information with environmental stiffness compensation.

[0154] In the constructed self-disturbance rejection fuzzy variable impedance closed-loop control circuit, the outer-loop self-disturbance rejection grinding controller estimates and compensates the position error based on the expected position and error disturbance at the beginning and during grinding to solve the disturbance problem of the grinding control system; the inner-loop fuzzy variable impedance controller dynamically adjusts the force correction amount by changing parameters to adapt to the variable contact environment of grinding and improve the control effect.

[0155] According to the removal allowance distribution and the size and shape of the high-speed rail head, the robot grinding and polishing processing area and the robot grinding and polishing processing trajectory of the corresponding area can be accurately divided. Figure 6 By dividing the polished area of ​​the high-speed rail head into regions, defects and uneven parts on the surface of the high-speed rail head can be identified and processed more accurately. Setting different polishing forces for different areas helps to improve the polishing accuracy and efficiency and ensure the polishing quality.

[0156] More specifically, for regions 1 and 8, the radius of curvature is relatively large and smooth, and is less affected by external forces. The main purpose is to ensure the flatness and glossiness of the overall appearance, and the flatness after grinding is required to be high. Therefore, a smaller grinding force is set to avoid excessive grinding of the plane.

[0157] For areas 2 and 7, the curvature changes are complex and the surface curvature changes sharply. Set a medium to small grinding force to remove minor surface flaws and avoid deformation of the curved surface caused by excessive grinding. At the same time, improve the efficiency of grinding, reduce excessive removal of the putty layer, and retain the appropriate thickness for subsequent repair and painting work.

[0158] For areas 3 and 6, the radius of curvature is relatively large, but they are close to the ground and the putty layer is severely damaged. A larger grinding force is required to quickly remove the deeper scratches, dents and uneven parts to restore the good shape of the front end of the train. Therefore, a larger grinding force is set to more effectively remove the severely damaged putty layer and prepare for subsequent repair and painting work. At the same time, due to the importance of this area in the operation of high-speed rail, it is necessary to ensure that its surface quality meets high standards to reduce air resistance and improve operation safety.

[0159] For areas 4 and 5, the damage is relatively moderate, and it is necessary to remove the slight wear and unevenness of the surface while ensuring a natural transition with other areas. Therefore, setting a medium grinding force can achieve a certain grinding effect without destroying the coordination with other areas due to excessive grinding force.

[0160] Furthermore, this embodiment also establishes a precise removal model of putty material for high-speed rail head, and reversely obtains the robot grinding and polishing process parameters of these parts based on the material removal margin of different parts, such as robot feed rate; grinding and polishing disc speed, diameter and direction, etc.

[0161] Furthermore, this embodiment also needs to execute the high-speed rail head component robot automatic grinding program, and perform surface morphology measurement and processing allowance calculation again after grinding. If all areas meet the requirements, the grinding is completed and enters the next process; if not, the area continues to execute the grinding program until the material removal amount and surface consistency requirements are met.

[0162] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A robot adaptive grinding and polishing method for high-speed railway head surface coating, characterized in that: The following steps are involved: Step 1: reconstructing the actual 3D model of the high-speed train head according to the 3D theoretical model of the high-speed train head and the 3D shape data obtained by measurement; Step 2: Based on the actual three-dimensional model and the three-dimensional theoretical model, a distribution topological map of the surface putty coating material removal residue at different parts is obtained; Step 3: Construct a contact force model between the grinding disc at the end of the robot and the workpiece at the front of the high-speed railway; Step 4: According to the distribution of the removed excess and the size and shape of the high-speed rail head, the robot grinding and polishing processing area is divided, and a fuzzy variable resistance control method with self-disturbance rejection is used to achieve precise control of constant force grinding in local areas; Step five: Establish a precise removal model for putty materials on the front of a high-speed railway, obtain the nonlinear relationship between the material removal allowance and the robot grinding and polishing process parameters, and reversely obtain the robot grinding and polishing process parameters of the corresponding parts based on the surface putty coating material removal allowance of different parts.

2. The method for adaptive grinding and polishing of the surface coating of a high-speed railway head by a robot according to claim 1 is characterized in that: Step one includes the following steps: (11) Import the three-dimensional theoretical model of the high-speed rail head and build a robot measurement workstation for the high-speed rail head in the simulation software to generate the robot's scanning path; (22) obtaining time-synchronized three-dimensional shape information of the high-speed train head and depth information and photometric information of the putty layer; (23) The three-dimensional shape information and the depth information of the putty layer are fused with the photometric information. During the fusion process, the improved K-means clustering algorithm is used to establish the spatial topological relationship of the point cloud. Then, the noise points of each type of point cloud after clustering are identified and removed. The data is then denoised, streamlined, fitted, spliced, and smoothed to obtain a point cloud model of the putty layer on the surface of the high-speed rail head. (24) Based on the point cloud model, gradient model, photometric information, and depth information of the putty layer on the surface of the high-speed train head, the actual three-dimensional model of the high-speed train head is reconstructed in reverse.

3. The method for adaptive grinding and polishing of the surface coating of a high-speed railway head by a robot according to claim 1 is characterized in that: Step 2 includes the following steps: Based on the Preston wear law, a material removal theoretical model is constructed. An infinitesimal element of dx*dy is taken on the polished surface. In this material removal theoretical model, the relative sliding velocity v a Approximately equal to the linear velocity v of the abrasive s (x,y), let D l is the removal depth per unit contact length between the polishing disc and the rubber surface along the polishing path, and can be obtained as: Among them, v f is the feed rate, β is the Preston wear coefficient; Based on the constructed actual 3D model and 3D theoretical model, the surface putty coating material removal amount of different parts can be accurately obtained.

4. The method for adaptive grinding and polishing of a high-speed railway head surface coating by a robot according to claim 1, characterized in that: Step 3, the construction of the contact force model between the robot end disc grinding head and the high-speed rail head workpiece includes: (31) Using the Galerkin method, assuming that the left and right boundaries of the circular thin plate are in the shape of a quadratic curve after contact deformation, the displacement of each point on the thin plate is calculated to satisfy: Among them, h corresponds to the thickness of the circular thin plate after deformation, and d corresponds to the thickness of the circular thin plate compressed in the z-axis direction; (32) Assuming that all mechanical energy is converted into shear potential energy, the shear potential energy, i.e., the total energy, is derived from the shear strain: Among them, G is the shear modulus, and R corresponds to the radius of the bottom surface of the circular thin plate before deformation; (33) According to the relationship between energy and force, the contact force on the grinding disc can be obtained: Among them, F N is the contact force on the grinding disc, and U is the total energy.

5. The method for adaptive grinding and polishing of the surface coating of a high-speed railway head by a robot according to claim 4 is characterized in that: Step three also includes: analyzing the stress distribution of the grinding disc, specifically: Using the infinitesimal method, take an infinitesimal element with a length of Δr for the portion where r=r0, and calculate the stress distribution at the bottom of the grinding disc according to the calculation method of the shear potential energy and energy in step (32) and the calculation method of the contact force on the grinding disc in step (33): Among them, σ(r,0) is the stress of one microelement at the bottom of the grinding disk.

6. The method for adaptive grinding and polishing of the surface coating of a high-speed railway head by a robot according to claim 5 is characterized in that: Step three also includes: The circular thin plate model includes the sponge pad and sandpaper on the grinding head. According to the deformation relationship between the sponge pad and the sandpaper, the comprehensive shear modulus of the grinding disc is derived as follows: Among them, E1 and E2 correspond to the elastic modulus of the sponge pad and sandpaper, d1 and d2 correspond to the original thickness of the sponge pad and sandpaper, and v1 and v2 correspond to the Poisson's ratio of the sponge pad and sandpaper.

7. The method for adaptive grinding and polishing of the surface coating of a high-speed railway head by a robot according to claim 1, characterized in that: In step 4, the fuzzy variable resistance control method with auto-disturbance rejection includes a fuzzy variable impedance controller and an auto-disturbance rejection polishing controller, wherein the fuzzy variable impedance controller is used for the inner loop and the auto-disturbance rejection polishing controller is used for the outer loop, and the two constitute a closed-loop control circuit.

8. The method for adaptive grinding and polishing of the surface coating of a high-speed railway head by a robot according to claim 7 is characterized in that: The fuzzy variable resistance control method with self-disturbance rejection to achieve precise control of constant force grinding in a local area includes: (41) Given the desired position, the anti-disturbance grinding controller estimates the grinding model and robot disturbance state online and compensates the signal based on the current working position of the grinding tool fed back by the position sensor; (42) The output value of the anti-disturbance grinding controller is used as the input of the impedance model. The relationship between the telescopic displacement of the end effector and the contact force is established through the model. The output force correction value compensates the force to make the force deviation close to zero, thereby preventing over-cutting caused by overshoot of the contact force. (43) The impedance parameters of the impedance model are fuzzy controlled and combined with the force feedback information considering the environmental stiffness compensation to optimize the output.

9. A high-speed railway head surface coating robot adaptive grinding and polishing system, characterized in that: include: Actual 3D model reconstruction module: used to reconstruct the actual 3D model of the high-speed train head according to the 3D theoretical model of the high-speed train head and the 3D shape data obtained by measurement; Material removal amount calculation module: used to obtain a distribution topological map of the surface putty coating material removal allowance at different parts based on the actual three-dimensional model and the three-dimensional theoretical model; Robotic grinding module: including a robot and a grinding disc designed at the end of the robot for performing grinding; Robot grinding control module: used to build the contact force model between the grinding disc and the workpiece, and divide the robot grinding and polishing processing area according to the distribution of the removal margin and the size and shape of the high-speed rail head, and adopt the fuzzy variable resistance control method with self-disturbance rejection to achieve precise control of constant force grinding in local areas; Robot processing parameter calculation module: used to establish a precise removal model for putty materials on the front of a high-speed railway, and to reversely obtain the robot grinding and polishing process parameters of the corresponding parts based on the removal allowance of the surface putty coating material on different parts.

10. The high-speed railway head surface coating robot adaptive grinding and polishing system according to claim 9, characterized in that: The actual three-dimensional model reconstruction module includes: Structured light projection system: used to obtain the structured light point cloud of the high-speed rail head and reconstruct the three-dimensional shape information of the head based on the structured light point cloud; Photometric stereo vision camera system: used to obtain depth information and photometric information of the putty layer; Synchronization system: used to synchronize the working time of the structured light projection system and the photometric stereo vision camera system; Data processing module: used to fuse the 3D shape information and the depth information and photometric information of the putty layer, so as to reversely construct the actual 3D model of the high-speed rail head; Preferably, the robot polishing module further includes horizontal guide rails and vertical guide rails disposed on both sides of the vehicle head for driving the robot to move.

Images